freebsd-skq/usr.sbin/ypserv/yp_dblookup.c

736 lines
16 KiB
C
Raw Normal View History

/*-
* SPDX-License-Identifier: BSD-4-Clause
*
* Copyright (c) 1995
* Bill Paul <wpaul@ctr.columbia.edu>. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by Bill Paul.
* 4. Neither the name of the author nor the names of any co-contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY Bill Paul AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL Bill Paul OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <db.h>
#include <errno.h>
#include <fcntl.h>
#include <limits.h>
#include <paths.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <sys/stat.h>
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#include <sys/param.h>
#include <rpcsvc/yp.h>
#include "yp_extern.h"
int ypdb_debug = 0;
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
enum ypstat yp_errno = YP_TRUE;
#define PERM_SECURE (S_IRUSR|S_IWUSR)
HASHINFO openinfo = {
4096, /* bsize */
32, /* ffactor */
256, /* nelem */
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
2048 * 512, /* cachesize */
NULL, /* hash */
0, /* lorder */
};
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#ifdef DB_CACHE
#include <sys/queue.h>
#ifndef MAXDBS
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#define MAXDBS 20
#endif
static int numdbs = 0;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
struct dbent {
DB *dbp;
char *name;
char *key;
int size;
int flags;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
};
2000-12-29 20:33:28 +00:00
static TAILQ_HEAD(circlehead, circleq_entry) qhead;
struct circleq_entry {
struct dbent *dbptr;
2000-12-29 20:33:28 +00:00
TAILQ_ENTRY(circleq_entry) links;
};
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
/*
* Initialize the circular queue.
*/
void
yp_init_dbs(void)
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
{
2000-12-29 20:33:28 +00:00
TAILQ_INIT(&qhead);
return;
}
/*
* Dynamically allocate an entry for the circular queue.
* Return a NULL pointer on failure.
*/
static struct circleq_entry *
yp_malloc_qent(void)
{
register struct circleq_entry *q;
q = malloc(sizeof(struct circleq_entry));
if (q == NULL) {
yp_error("failed to malloc() circleq entry");
return(NULL);
}
bzero((char *)q, sizeof(struct circleq_entry));
q->dbptr = malloc(sizeof(struct dbent));
if (q->dbptr == NULL) {
yp_error("failed to malloc() circleq entry");
free(q);
return(NULL);
}
bzero((char *)q->dbptr, sizeof(struct dbent));
return(q);
}
/*
* Free a previously allocated circular queue
* entry.
*/
static void
yp_free_qent(struct circleq_entry *q)
{
/*
* First, close the database. In theory, this is also
* supposed to free the resources allocated by the DB
* package, including the memory pointed to by q->dbptr->key.
* This means we don't have to free q->dbptr->key here.
*/
if (q->dbptr->dbp) {
(void)(q->dbptr->dbp->close)(q->dbptr->dbp);
q->dbptr->dbp = NULL;
}
/*
* Then free the database name, which was strdup()'ed.
*/
free(q->dbptr->name);
/*
* Free the rest of the dbent struct.
*/
free(q->dbptr);
q->dbptr = NULL;
/*
* Free the circleq struct.
*/
free(q);
q = NULL;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
return;
}
/*
* Zorch a single entry in the dbent queue and release
* all its resources. (This always removes the last entry
* in the queue.)
*/
static void
yp_flush(void)
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
{
register struct circleq_entry *qptr;
2000-12-29 20:33:28 +00:00
qptr = TAILQ_LAST(&qhead, circlehead);
TAILQ_REMOVE(&qhead, qptr, links);
yp_free_qent(qptr);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
numdbs--;
return;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
/*
* Close all databases, erase all database names and empty the queue.
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
*/
void
yp_flush_all(void)
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
{
register struct circleq_entry *qptr;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
while (!TAILQ_EMPTY(&qhead)) {
2000-12-29 20:33:28 +00:00
qptr = TAILQ_FIRST(&qhead); /* save this */
TAILQ_REMOVE(&qhead, qptr, links);
yp_free_qent(qptr);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
numdbs = 0;
return;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
static char *inter_string = "YP_INTERDOMAIN";
static char *secure_string = "YP_SECURE";
static int inter_sz = sizeof("YP_INTERDOMAIN") - 1;
static int secure_sz = sizeof("YP_SECURE") - 1;
static int
yp_setflags(DB *dbp)
{
DBT key = { NULL, 0 }, data = { NULL, 0 };
int flags = 0;
key.data = inter_string;
key.size = inter_sz;
if (!(dbp->get)(dbp, &key, &data, 0))
flags |= YP_INTERDOMAIN;
key.data = secure_string;
key.size = secure_sz;
if (!(dbp->get)(dbp, &key, &data, 0))
flags |= YP_SECURE;
return(flags);
}
int
yp_testflag(char *map, char *domain, int flag)
{
char buf[MAXPATHLEN + 2];
register struct circleq_entry *qptr;
if (map == NULL || domain == NULL)
return(0);
strcpy(buf, domain);
strcat(buf, "/");
strcat(buf, map);
2000-12-29 20:33:28 +00:00
TAILQ_FOREACH(qptr, &qhead, links) {
if (!strcmp(qptr->dbptr->name, buf)) {
if (qptr->dbptr->flags & flag)
return(1);
else
return(0);
}
}
if (yp_open_db_cache(domain, map, NULL, 0) == NULL)
return(0);
2000-12-29 20:33:28 +00:00
if (TAILQ_FIRST(&qhead)->dbptr->flags & flag)
return(1);
return(0);
}
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
/*
* Add a DB handle and database name to the cache. We only maintain
* fixed number of entries in the cache, so if we're asked to store
* a new entry when all our slots are already filled, we have to kick
* out the entry in the last slot to make room.
*/
static int
yp_cache_db(DB *dbp, char *name, int size)
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
{
register struct circleq_entry *qptr;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
if (numdbs == MAXDBS) {
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
if (ypdb_debug)
yp_error("queue overflow -- releasing last slot");
yp_flush();
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
/*
* Allocate a new queue entry.
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
*/
if ((qptr = yp_malloc_qent()) == NULL) {
yp_error("failed to allocate a new cache entry");
return(1);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
qptr->dbptr->dbp = dbp;
qptr->dbptr->name = strdup(name);
qptr->dbptr->size = size;
qptr->dbptr->key = NULL;
qptr->dbptr->flags = yp_setflags(dbp);
2000-12-29 20:33:28 +00:00
TAILQ_INSERT_HEAD(&qhead, qptr, links);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
numdbs++;
return(0);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
/*
* Search the list for a database matching 'name.' If we find it,
* move it to the head of the list and return its DB handle. If
* not, just fail: yp_open_db_cache() will subsequently try to open
* the database itself and call yp_cache_db() to add it to the
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
* list.
*
* The search works like this:
*
* - The caller specifies the name of a database to locate. We try to
* find an entry in our queue with a matching name.
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
*
* - If the caller doesn't specify a key or size, we assume that the
* first entry that we encounter with a matching name is returned.
* This will result in matches regardless of the key/size values
* stored in the queue entry.
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
*
* - If the caller also specifies a key and length, we check to see
* if the key and length saved in the queue entry also matches.
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
* This lets us return a DB handle that's already positioned at the
* correct location within a database.
*
* - Once we have a match, it gets migrated to the top of the queue
* so that it will be easier to find if another request for
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
* the same database comes in later.
*/
static DB *
yp_find_db(const char *name, const char *key, int size)
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
{
register struct circleq_entry *qptr;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
2000-12-29 20:33:28 +00:00
TAILQ_FOREACH(qptr, &qhead, links) {
if (!strcmp(qptr->dbptr->name, name)) {
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
if (size) {
if (size != qptr->dbptr->size ||
strncmp(qptr->dbptr->key, key, size))
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
continue;
} else {
if (qptr->dbptr->size)
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
continue;
}
2000-12-29 20:33:28 +00:00
if (qptr != TAILQ_FIRST(&qhead)) {
TAILQ_REMOVE(&qhead, qptr, links);
TAILQ_INSERT_HEAD(&qhead, qptr, links);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
return(qptr->dbptr->dbp);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
}
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
return(NULL);
}
/*
* Open a DB database and cache the handle for later use. We first
* check the cache to see if the required database is already open.
* If so, we fetch the handle from the cache. If not, we try to open
* the database and save the handle in the cache for later use.
*/
DB *
yp_open_db_cache(const char *domain, const char *map, const char *key,
const int size)
{
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
DB *dbp = NULL;
char buf[MAXPATHLEN + 2];
/*
snprintf(buf, sizeof(buf), "%s/%s", domain, map);
*/
yp_errno = YP_TRUE;
strcpy(buf, domain);
strcat(buf, "/");
strcat(buf, map);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
if ((dbp = yp_find_db(buf, key, size)) != NULL) {
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
return(dbp);
} else {
if ((dbp = yp_open_db(domain, map)) != NULL) {
if (yp_cache_db(dbp, buf, size)) {
(void)(dbp->close)(dbp);
yp_errno = YP_YPERR;
return(NULL);
}
}
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
return (dbp);
}
#endif
/*
* Open a DB database.
*/
DB *
yp_open_db(const char *domain, const char *map)
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
{
DB *dbp = NULL;
char buf[MAXPATHLEN + 2];
yp_errno = YP_TRUE;
if (map[0] == '.' || strchr(map, '/')) {
yp_errno = YP_BADARGS;
return (NULL);
}
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#ifdef DB_CACHE
if (yp_validdomain(domain)) {
yp_errno = YP_NODOM;
return(NULL);
}
#endif
snprintf(buf, sizeof(buf), "%s/%s/%s", yp_dir, domain, map);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#ifdef DB_CACHE
again:
#endif
dbp = dbopen(buf, O_RDONLY, PERM_SECURE, DB_HASH, NULL);
if (dbp == NULL) {
switch (errno) {
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#ifdef DB_CACHE
case ENFILE:
/*
* We ran out of file descriptors. Nuke an
* open one and try again.
*/
yp_error("ran out of file descriptors");
yp_flush();
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
goto again;
break;
#endif
case ENOENT:
yp_errno = YP_NOMAP;
break;
case EFTYPE:
yp_errno = YP_BADDB;
break;
default:
yp_errno = YP_YPERR;
break;
}
}
return (dbp);
}
/*
* Database access routines.
*
* - yp_get_record(): retrieve an arbitrary key/data pair given one key
* to match against.
*
* - yp_first_record(): retrieve first key/data base in a database.
*
* - yp_next_record(): retrieve key/data pair that sequentially follows
* the supplied key value in the database.
*/
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
#ifdef DB_CACHE
int
yp_get_record(DB *dbp, const DBT *key, DBT *data, int allow)
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
#else
int
yp_get_record(const char *domain, const char *map,
const DBT *key, DBT *data, int allow)
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
#endif
{
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
#ifndef DB_CACHE
DB *dbp;
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
#endif
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
int rval = 0;
#ifndef DB_CACHE
static unsigned char buf[YPMAXRECORD];
#endif
if (ypdb_debug)
yp_error("looking up key [%.*s]",
(int)key->size, (char *)key->data);
/*
* Avoid passing back magic "YP_*" entries unless
* the caller specifically requested them by setting
* the 'allow' flag.
*/
if (!allow && !strncmp(key->data, "YP_", 3))
return(YP_NOKEY);
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
#ifndef DB_CACHE
if ((dbp = yp_open_db(domain, map)) == NULL) {
return(yp_errno);
}
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
#endif
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
if ((rval = (dbp->get)(dbp, key, data, 0)) != 0) {
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
TAILQ_FIRST(&qhead)->dbptr->size = 0;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#else
(void)(dbp->close)(dbp);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#endif
if (rval == 1)
return(YP_NOKEY);
else
return(YP_BADDB);
}
if (ypdb_debug)
yp_error("result of lookup: key: [%.*s] data: [%.*s]",
(int)key->size, (char *)key->data,
(int)data->size, (char *)data->data);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
if (TAILQ_FIRST(&qhead)->dbptr->size) {
TAILQ_FIRST(&qhead)->dbptr->key = "";
TAILQ_FIRST(&qhead)->dbptr->size = 0;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
#else
bcopy(data->data, &buf, data->size);
data->data = &buf;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
(void)(dbp->close)(dbp);
#endif
return(YP_TRUE);
}
int
yp_first_record(const DB *dbp, DBT *key, DBT *data, int allow)
{
int rval;
#ifndef DB_CACHE
static unsigned char buf[YPMAXRECORD];
#endif
if (ypdb_debug)
yp_error("retrieving first key in map");
if ((rval = (dbp->seq)(dbp,key,data,R_FIRST)) != 0) {
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
TAILQ_FIRST(&qhead)->dbptr->size = 0;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#endif
if (rval == 1)
return(YP_NOKEY);
else
return(YP_BADDB);
}
/* Avoid passing back magic "YP_*" records. */
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
while (!strncmp(key->data, "YP_", 3) && !allow) {
if ((rval = (dbp->seq)(dbp,key,data,R_NEXT)) != 0) {
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
TAILQ_FIRST(&qhead)->dbptr->size = 0;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#endif
if (rval == 1)
return(YP_NOKEY);
else
return(YP_BADDB);
}
}
if (ypdb_debug)
yp_error("result of lookup: key: [%.*s] data: [%.*s]",
(int)key->size, (char *)key->data,
(int)data->size, (char *)data->data);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
if (TAILQ_FIRST(&qhead)->dbptr->size) {
TAILQ_FIRST(&qhead)->dbptr->key = key->data;
TAILQ_FIRST(&qhead)->dbptr->size = key->size;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
#else
bcopy(data->data, &buf, data->size);
data->data = &buf;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#endif
return(YP_TRUE);
}
int
yp_next_record(const DB *dbp, DBT *key, DBT *data, int all, int allow)
{
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
static DBT lkey = { NULL, 0 };
static DBT ldata = { NULL, 0 };
int rval;
#ifndef DB_CACHE
static unsigned char keybuf[YPMAXRECORD];
static unsigned char datbuf[YPMAXRECORD];
#endif
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
if (key == NULL || !key->size || key->data == NULL) {
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
rval = yp_first_record(dbp,key,data,allow);
if (rval == YP_NOKEY)
return(YP_NOMORE);
else {
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
TAILQ_FIRST(&qhead)->dbptr->key = key->data;
TAILQ_FIRST(&qhead)->dbptr->size = key->size;
#endif
return(rval);
}
}
if (ypdb_debug)
yp_error("retrieving next key, previous was: [%.*s]",
(int)key->size, (char *)key->data);
if (!all) {
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
if (TAILQ_FIRST(&qhead)->dbptr->key == NULL) {
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#endif
(dbp->seq)(dbp,&lkey,&ldata,R_FIRST);
while (key->size != lkey.size ||
strncmp(key->data, lkey.data,
(int)key->size))
if ((dbp->seq)(dbp,&lkey,&ldata,R_NEXT)) {
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
TAILQ_FIRST(&qhead)->dbptr->size = 0;
#endif
return(YP_NOKEY);
}
#ifdef DB_CACHE
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
#endif
}
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
if ((dbp->seq)(dbp,key,data,R_NEXT)) {
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
TAILQ_FIRST(&qhead)->dbptr->size = 0;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#endif
return(YP_NOMORE);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
/* Avoid passing back magic "YP_*" records. */
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
while (!strncmp(key->data, "YP_", 3) && !allow)
if ((dbp->seq)(dbp,key,data,R_NEXT)) {
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
TAILQ_FIRST(&qhead)->dbptr->size = 0;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#endif
return(YP_NOMORE);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
if (ypdb_debug)
yp_error("result of lookup: key: [%.*s] data: [%.*s]",
(int)key->size, (char *)key->data,
(int)data->size, (char *)data->data);
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#ifdef DB_CACHE
2000-12-29 20:33:28 +00:00
if (TAILQ_FIRST(&qhead)->dbptr->size) {
TAILQ_FIRST(&qhead)->dbptr->key = key->data;
TAILQ_FIRST(&qhead)->dbptr->size = key->size;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
}
#else
bcopy(key->data, &keybuf, key->size);
lkey.data = &keybuf;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
lkey.size = key->size;
bcopy(data->data, &datbuf, data->size);
data->data = &datbuf;
Performance enhancements (I hope) and new stuff: yp_dblookup.c: - Implement database handle caching. What this means is that instead of opening and closing map databases for each request, we open a database and save the handle (and, if requested, the key index) in an array. This saves a bit of overhead on things like repeated YPPROC_NEXT calls, such as you'd get from getpwent(). Normally, each YPPROC_NEXT would require open()ing the database, seeking to the location supplied by the caller (which is time consuming with hash databases as the R_CURSOR flag doesn't work), reading the data, close()ing the database and then shipping the data off to the caller. The system call overhead is prohibitive, especially with very large maps. By caching the handle to an open database, we elimitate at least the open()/close() system calls, as well as the associated DB setup and tear-down operations, for a large percentage of the time. This improves performance substantially at the cost of consuming a little more memory than before. Note that all the caching support is surrounded by #ifdef DB_CACHE so that this same source module can still be used by other programs that don't need it. - Make yp_open_db() call yp_validdomain(). Doing it here saves cycles when caching is enabled since a hit on the map cache list by definition means that the domain being referenced is valid. - Also make yp_open_db() check for exhaustion of file descriptors, just in case. yp_server.c: - Reorganize things a little to take advantage of the database handle caching. Add a call to yp_flush_all() in ypproc_clear_2_svc(). - Remove calls to yp_validdomain() from some of the service procedures. yp_validdomain() is called inside yp_open_db() now, so procedures that call into the database package don't need to use yp_validdomain() themselves. - Fix a bogosity in ypproc_maplist_2_svc(): don't summarily initiallize the result.maps pointer to NULL. This causes yp_maplist_free() to fail and leaks memory. - Make ypproc_master_2_svc() copy the string it gets from the database package into a private static buffer before trying to NUL terminate it. This is necessary with the DB handle caching: stuffing a NUL into the data returned by DB package will goof it up internally. yp_main.c: - Stuff for DB handle caching: call yp_init_dbs() to clear the handle array and add call to yp_flush_all() to the SIGHUP signal handler. Makefile.yp: - Reorganize to deal with database caching. yp_mkdb(8) can now be used to send a YPPROC_CLEAR signal to ypserv(8). Call it after each map is created to refresh ypserv's cache. - Add support for mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). - Make default location for the netgroups source file be /var/yp/netgroup instead of /etc/netgroup. mkaliases: - New file: script to generate mail.alias map. Contributed by Mike Murphy (mrm@sceard.com). Makefile: - Install Makefile.yp as /var/yp/Makefile.dist and link it to /var/yp/Makefile only if /var/yp/Makefile doesn't already exist. Suggested by Peter Wemm. - Install new mkaliases script in /usr/libexec along with mknetid. - Use somewhat saner approach to generating rpcgen-dependent files as suggested by Garrett Wollman.
1996-04-28 04:38:52 +00:00
#endif
return(YP_TRUE);
}
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
#ifdef DB_CACHE
/*
* Database glue functions.
*/
static DB *yp_currmap_db = NULL;
static int yp_allow_db = 0;
ypstat
yp_select_map(char *map, char *domain, keydat *key, int allow)
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
{
if (key == NULL)
yp_currmap_db = yp_open_db_cache(domain, map, NULL, 0);
else
yp_currmap_db = yp_open_db_cache(domain, map,
key->keydat_val,
key->keydat_len);
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
yp_allow_db = allow;
return(yp_errno);
}
ypstat
yp_getbykey(keydat *key, valdat *val)
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
{
DBT db_key = { NULL, 0 }, db_val = { NULL, 0 };
ypstat rval;
db_key.data = key->keydat_val;
db_key.size = key->keydat_len;
rval = yp_get_record(yp_currmap_db,
&db_key, &db_val, yp_allow_db);
if (rval == YP_TRUE) {
val->valdat_val = db_val.data;
val->valdat_len = db_val.size;
}
return(rval);
}
ypstat
yp_firstbykey(keydat *key, valdat *val)
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
{
DBT db_key = { NULL, 0 }, db_val = { NULL, 0 };
ypstat rval;
rval = yp_first_record(yp_currmap_db, &db_key, &db_val, yp_allow_db);
if (rval == YP_TRUE) {
key->keydat_val = db_key.data;
key->keydat_len = db_key.size;
val->valdat_val = db_val.data;
val->valdat_len = db_val.size;
}
return(rval);
}
ypstat
yp_nextbykey(keydat *key, valdat *val)
Big round o changes: - yp_dblookup.c: Create non-DB specific database access functions. Using these allows access to the underlying database functions without needing explicit knowledge of Berkeley DB. (These are used only when DB_CACHE is #defined. Other programs that use the non-caching functions (yp_mkdb, ypxfr, yppush, rpc.yppasswdd) shouldn't notice the difference.) - yp_dnslookup: Implement async DNS lookups. We send our own DNS requests using UDP and put the request in a queue. When the response arrives, we use the ID in the header to find the corresponsing queue entry and then send the response to the client. We can go about our business and handle other YP requests in the meantime. This way, we can deal with time consuming DNS requests without blocking and without forking. - yp_server.c: Convert to using new non-DB-specific database access functions. This simplifies the code a bit and removes the need for this module to know anything about Berkeley DB. Also convert the ypproc_match_2_svc() function to use the async DNS lookup routines. - yp_main.c: tweak yp_svc_run() to add the resolver socket to the set of descriptors monitored in the select() loop. Also add a timeout to select(); we may get stale DNS requests stuck in the queue which we want to invalidate after a while. If the timeout hits, we decrement the ttl on all pending DNS requests and nuke those requests that aren't handled before ttl hits zero. - yp_extern.h: Add prototypes for new stuff. - yp_svc_udp.c (new file): The async resolver code needs to be able to rummage around inside the RPC UDP transport handle in order to work correcty. There's basically one transport handle, and each time a request comes in, the transaction ID in the handle is changed. This means that if we queue a DNS request, then we handle some other unrelated requests, we will be unable to send the DNS response because the transaction ID and remote address of the client that made the DNS request will have been lost. What we need to do is save the client address and transaction ID in the queue entry for the DNS request, then put the transaction ID and address back in the transport handle when we're ready to reply. (And then we have to undo the change so as not to confuse any other part of the server.) The trouble is that the transaction ID is hidden in an opaque part of the transport handle, and only the code in the svc_udp module in the RPC library knows how to handle it. This file contains a couple of functions that let us read and set the transaction ID in spite of this. This is really a dirty trick and I should be taken out and shot for even thinking about it, but there's no other way to get this stuff to work. - Makefile: add yp_svc_udp.c to SRCS.
1996-12-22 22:30:58 +00:00
{
DBT db_key = { NULL, 0 }, db_val = { NULL, 0 };
ypstat rval;
db_key.data = key->keydat_val;
db_key.size = key->keydat_len;
rval = yp_next_record(yp_currmap_db, &db_key, &db_val, 0, yp_allow_db);
if (rval == YP_TRUE) {
key->keydat_val = db_key.data;
key->keydat_len = db_key.size;
val->valdat_val = db_val.data;
val->valdat_len = db_val.size;
}
return(rval);
}
#endif