freebsd-dev/sys/kern/kern_proc.c
Poul-Henning Kamp 75c1354190 This Implements the mumbled about "Jail" feature.
This is a seriously beefed up chroot kind of thing.  The process
is jailed along the same lines as a chroot does it, but with
additional tough restrictions imposed on what the superuser can do.

For all I know, it is safe to hand over the root bit inside a
prison to the customer living in that prison, this is what
it was developed for in fact:  "real virtual servers".

Each prison has an ip number associated with it, which all IP
communications will be coerced to use and each prison has its own
hostname.

Needless to say, you need more RAM this way, but the advantage is
that each customer can run their own particular version of apache
and not stomp on the toes of their neighbors.

It generally does what one would expect, but setting up a jail
still takes a little knowledge.

A few notes:

   I have no scripts for setting up a jail, don't ask me for them.

   The IP number should be an alias on one of the interfaces.

   mount a /proc in each jail, it will make ps more useable.

   /proc/<pid>/status tells the hostname of the prison for
   jailed processes.

   Quotas are only sensible if you have a mountpoint per prison.

   There are no privisions for stopping resource-hogging.

   Some "#ifdef INET" and similar may be missing (send patches!)

If somebody wants to take it from here and develop it into
more of a "virtual machine" they should be most welcome!

Tools, comments, patches & documentation most welcome.

Have fun...

Sponsored by:   http://www.rndassociates.com/
Run for almost a year by:       http://www.servetheweb.com/
1999-04-28 11:38:52 +00:00

597 lines
15 KiB
C

/*
* Copyright (c) 1982, 1986, 1989, 1991, 1993
* The Regents of the University of California. 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 the University of
* California, Berkeley and its contributors.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS 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 THE REGENTS 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.
*
* @(#)kern_proc.c 8.7 (Berkeley) 2/14/95
* $Id: kern_proc.c,v 1.46 1999/02/19 14:25:34 luoqi Exp $
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/sysctl.h>
#include <sys/proc.h>
#include <sys/malloc.h>
#include <sys/filedesc.h>
#include <sys/tty.h>
#include <sys/signalvar.h>
#include <vm/vm.h>
#include <sys/lock.h>
#include <vm/pmap.h>
#include <vm/vm_map.h>
#include <sys/user.h>
#include <vm/vm_zone.h>
static MALLOC_DEFINE(M_PGRP, "pgrp", "process group header");
MALLOC_DEFINE(M_SESSION, "session", "session header");
static MALLOC_DEFINE(M_PROC, "proc", "Proc structures");
MALLOC_DEFINE(M_SUBPROC, "subproc", "Proc sub-structures");
struct prochd qs[NQS]; /* as good a place as any... */
struct prochd rtqs[NQS]; /* Space for REALTIME queues too */
struct prochd idqs[NQS]; /* Space for IDLE queues too */
static void pgdelete __P((struct pgrp *));
/*
* Structure associated with user cacheing.
*/
struct uidinfo {
LIST_ENTRY(uidinfo) ui_hash;
uid_t ui_uid;
long ui_proccnt;
};
#define UIHASH(uid) (&uihashtbl[(uid) & uihash])
static LIST_HEAD(uihashhead, uidinfo) *uihashtbl;
static u_long uihash; /* size of hash table - 1 */
static void orphanpg __P((struct pgrp *pg));
/*
* Other process lists
*/
struct pidhashhead *pidhashtbl;
u_long pidhash;
struct pgrphashhead *pgrphashtbl;
u_long pgrphash;
struct proclist allproc;
struct proclist zombproc;
vm_zone_t proc_zone;
/*
* Initialize global process hashing structures.
*/
void
procinit()
{
LIST_INIT(&allproc);
LIST_INIT(&zombproc);
pidhashtbl = hashinit(maxproc / 4, M_PROC, &pidhash);
pgrphashtbl = hashinit(maxproc / 4, M_PROC, &pgrphash);
uihashtbl = hashinit(maxproc / 16, M_PROC, &uihash);
proc_zone = zinit("PROC", sizeof (struct proc), 0, 0, 5);
}
/*
* Change the count associated with number of processes
* a given user is using.
*/
int
chgproccnt(uid, diff)
uid_t uid;
int diff;
{
register struct uidinfo *uip;
register struct uihashhead *uipp;
uipp = UIHASH(uid);
for (uip = uipp->lh_first; uip != 0; uip = uip->ui_hash.le_next)
if (uip->ui_uid == uid)
break;
if (uip) {
uip->ui_proccnt += diff;
if (uip->ui_proccnt > 0)
return (uip->ui_proccnt);
if (uip->ui_proccnt < 0)
panic("chgproccnt: procs < 0");
LIST_REMOVE(uip, ui_hash);
FREE(uip, M_PROC);
return (0);
}
if (diff <= 0) {
if (diff == 0)
return(0);
panic("chgproccnt: lost user");
}
MALLOC(uip, struct uidinfo *, sizeof(*uip), M_PROC, M_WAITOK);
LIST_INSERT_HEAD(uipp, uip, ui_hash);
uip->ui_uid = uid;
uip->ui_proccnt = diff;
return (diff);
}
/*
* Is p an inferior of the current process?
*/
int
inferior(p)
register struct proc *p;
{
for (; p != curproc; p = p->p_pptr)
if (p->p_pid == 0)
return (0);
return (1);
}
/*
* Locate a process by number
*/
struct proc *
pfind(pid)
register pid_t pid;
{
register struct proc *p;
for (p = PIDHASH(pid)->lh_first; p != 0; p = p->p_hash.le_next)
if (p->p_pid == pid)
return (p);
return (NULL);
}
/*
* Locate a process group by number
*/
struct pgrp *
pgfind(pgid)
register pid_t pgid;
{
register struct pgrp *pgrp;
for (pgrp = PGRPHASH(pgid)->lh_first; pgrp != 0;
pgrp = pgrp->pg_hash.le_next)
if (pgrp->pg_id == pgid)
return (pgrp);
return (NULL);
}
/*
* Move p to a new or existing process group (and session)
*/
int
enterpgrp(p, pgid, mksess)
register struct proc *p;
pid_t pgid;
int mksess;
{
register struct pgrp *pgrp = pgfind(pgid);
KASSERT(pgrp == NULL || !mksess,
("enterpgrp: setsid into non-empty pgrp"));
KASSERT(!SESS_LEADER(p),
("enterpgrp: session leader attempted setpgrp"));
if (pgrp == NULL) {
pid_t savepid = p->p_pid;
struct proc *np;
/*
* new process group
*/
KASSERT(p->p_pid == pgid,
("enterpgrp: new pgrp and pid != pgid"));
MALLOC(pgrp, struct pgrp *, sizeof(struct pgrp), M_PGRP,
M_WAITOK);
if ((np = pfind(savepid)) == NULL || np != p)
return (ESRCH);
if (mksess) {
register struct session *sess;
/*
* new session
*/
MALLOC(sess, struct session *, sizeof(struct session),
M_SESSION, M_WAITOK);
sess->s_leader = p;
sess->s_sid = p->p_pid;
sess->s_count = 1;
sess->s_ttyvp = NULL;
sess->s_ttyp = NULL;
bcopy(p->p_session->s_login, sess->s_login,
sizeof(sess->s_login));
p->p_flag &= ~P_CONTROLT;
pgrp->pg_session = sess;
KASSERT(p == curproc,
("enterpgrp: mksession and p != curproc"));
} else {
pgrp->pg_session = p->p_session;
pgrp->pg_session->s_count++;
}
pgrp->pg_id = pgid;
LIST_INIT(&pgrp->pg_members);
LIST_INSERT_HEAD(PGRPHASH(pgid), pgrp, pg_hash);
pgrp->pg_jobc = 0;
SLIST_INIT(&pgrp->pg_sigiolst);
} else if (pgrp == p->p_pgrp)
return (0);
/*
* Adjust eligibility of affected pgrps to participate in job control.
* Increment eligibility counts before decrementing, otherwise we
* could reach 0 spuriously during the first call.
*/
fixjobc(p, pgrp, 1);
fixjobc(p, p->p_pgrp, 0);
LIST_REMOVE(p, p_pglist);
if (p->p_pgrp->pg_members.lh_first == 0)
pgdelete(p->p_pgrp);
p->p_pgrp = pgrp;
LIST_INSERT_HEAD(&pgrp->pg_members, p, p_pglist);
return (0);
}
/*
* remove process from process group
*/
int
leavepgrp(p)
register struct proc *p;
{
LIST_REMOVE(p, p_pglist);
if (p->p_pgrp->pg_members.lh_first == 0)
pgdelete(p->p_pgrp);
p->p_pgrp = 0;
return (0);
}
/*
* delete a process group
*/
static void
pgdelete(pgrp)
register struct pgrp *pgrp;
{
/*
* Reset any sigio structures pointing to us as a result of
* F_SETOWN with our pgid.
*/
funsetownlst(&pgrp->pg_sigiolst);
if (pgrp->pg_session->s_ttyp != NULL &&
pgrp->pg_session->s_ttyp->t_pgrp == pgrp)
pgrp->pg_session->s_ttyp->t_pgrp = NULL;
LIST_REMOVE(pgrp, pg_hash);
if (--pgrp->pg_session->s_count == 0)
FREE(pgrp->pg_session, M_SESSION);
FREE(pgrp, M_PGRP);
}
/*
* Adjust pgrp jobc counters when specified process changes process group.
* We count the number of processes in each process group that "qualify"
* the group for terminal job control (those with a parent in a different
* process group of the same session). If that count reaches zero, the
* process group becomes orphaned. Check both the specified process'
* process group and that of its children.
* entering == 0 => p is leaving specified group.
* entering == 1 => p is entering specified group.
*/
void
fixjobc(p, pgrp, entering)
register struct proc *p;
register struct pgrp *pgrp;
int entering;
{
register struct pgrp *hispgrp;
register struct session *mysession = pgrp->pg_session;
/*
* Check p's parent to see whether p qualifies its own process
* group; if so, adjust count for p's process group.
*/
if ((hispgrp = p->p_pptr->p_pgrp) != pgrp &&
hispgrp->pg_session == mysession)
if (entering)
pgrp->pg_jobc++;
else if (--pgrp->pg_jobc == 0)
orphanpg(pgrp);
/*
* Check this process' children to see whether they qualify
* their process groups; if so, adjust counts for children's
* process groups.
*/
for (p = p->p_children.lh_first; p != 0; p = p->p_sibling.le_next)
if ((hispgrp = p->p_pgrp) != pgrp &&
hispgrp->pg_session == mysession &&
p->p_stat != SZOMB)
if (entering)
hispgrp->pg_jobc++;
else if (--hispgrp->pg_jobc == 0)
orphanpg(hispgrp);
}
/*
* A process group has become orphaned;
* if there are any stopped processes in the group,
* hang-up all process in that group.
*/
static void
orphanpg(pg)
struct pgrp *pg;
{
register struct proc *p;
for (p = pg->pg_members.lh_first; p != 0; p = p->p_pglist.le_next) {
if (p->p_stat == SSTOP) {
for (p = pg->pg_members.lh_first; p != 0;
p = p->p_pglist.le_next) {
psignal(p, SIGHUP);
psignal(p, SIGCONT);
}
return;
}
}
}
#include "opt_ddb.h"
#ifdef DDB
#include <ddb/ddb.h>
DB_SHOW_COMMAND(pgrpdump, pgrpdump)
{
register struct pgrp *pgrp;
register struct proc *p;
register int i;
for (i = 0; i <= pgrphash; i++) {
if ((pgrp = pgrphashtbl[i].lh_first) != NULL) {
printf("\tindx %d\n", i);
for (; pgrp != 0; pgrp = pgrp->pg_hash.le_next) {
printf(
"\tpgrp %p, pgid %ld, sess %p, sesscnt %d, mem %p\n",
(void *)pgrp, (long)pgrp->pg_id,
(void *)pgrp->pg_session,
pgrp->pg_session->s_count,
(void *)pgrp->pg_members.lh_first);
for (p = pgrp->pg_members.lh_first; p != 0;
p = p->p_pglist.le_next) {
printf("\t\tpid %ld addr %p pgrp %p\n",
(long)p->p_pid, (void *)p,
(void *)p->p_pgrp);
}
}
}
}
}
#endif /* DDB */
/*
* Fill in an eproc structure for the specified process.
*/
void
fill_eproc(p, ep)
register struct proc *p;
register struct eproc *ep;
{
register struct tty *tp;
bzero(ep, sizeof(*ep));
ep->e_paddr = p;
if (p->p_cred) {
ep->e_pcred = *p->p_cred;
if (p->p_ucred)
ep->e_ucred = *p->p_ucred;
}
if (p->p_procsig){
ep->e_procsig = *p->p_procsig;
}
if (p->p_stat != SIDL && p->p_stat != SZOMB && p->p_vmspace != NULL) {
register struct vmspace *vm = p->p_vmspace;
ep->e_vm = *vm;
ep->e_vm.vm_rssize = vmspace_resident_count(vm); /*XXX*/
}
if (p->p_pptr)
ep->e_ppid = p->p_pptr->p_pid;
if (p->p_pgrp) {
ep->e_pgid = p->p_pgrp->pg_id;
ep->e_jobc = p->p_pgrp->pg_jobc;
ep->e_sess = p->p_pgrp->pg_session;
if (ep->e_sess) {
bcopy(ep->e_sess->s_login, ep->e_login, sizeof(ep->e_login));
if (ep->e_sess->s_ttyvp)
ep->e_flag = EPROC_CTTY;
if (p->p_session && SESS_LEADER(p))
ep->e_flag |= EPROC_SLEADER;
}
}
if ((p->p_flag & P_CONTROLT) &&
(ep->e_sess != NULL) &&
((tp = ep->e_sess->s_ttyp) != NULL)) {
ep->e_tdev = tp->t_dev;
ep->e_tpgid = tp->t_pgrp ? tp->t_pgrp->pg_id : NO_PID;
ep->e_tsess = tp->t_session;
} else
ep->e_tdev = NODEV;
if (p->p_wmesg) {
strncpy(ep->e_wmesg, p->p_wmesg, WMESGLEN);
ep->e_wmesg[WMESGLEN] = 0;
}
}
static struct proc *
zpfind(pid_t pid)
{
struct proc *p;
for (p = zombproc.lh_first; p != 0; p = p->p_list.le_next)
if (p->p_pid == pid)
return (p);
return (NULL);
}
static int
sysctl_out_proc(struct proc *p, struct sysctl_req *req, int doingzomb)
{
struct eproc eproc;
int error;
pid_t pid = p->p_pid;
fill_eproc(p, &eproc);
error = SYSCTL_OUT(req,(caddr_t)p, sizeof(struct proc));
if (error)
return (error);
error = SYSCTL_OUT(req,(caddr_t)&eproc, sizeof(eproc));
if (error)
return (error);
if (!doingzomb && pid && (pfind(pid) != p))
return EAGAIN;
if (doingzomb && zpfind(pid) != p)
return EAGAIN;
return (0);
}
static int
sysctl_kern_proc SYSCTL_HANDLER_ARGS
{
int *name = (int*) arg1;
u_int namelen = arg2;
struct proc *p;
int doingzomb;
int error = 0;
if (oidp->oid_number == KERN_PROC_PID) {
if (namelen != 1)
return (EINVAL);
p = pfind((pid_t)name[0]);
if (!p)
return (0);
if (!PRISON_CHECK(curproc, p))
return (0);
error = sysctl_out_proc(p, req, 0);
return (error);
}
if (oidp->oid_number == KERN_PROC_ALL && !namelen)
;
else if (oidp->oid_number != KERN_PROC_ALL && namelen == 1)
;
else
return (EINVAL);
if (!req->oldptr) {
/* overestimate by 5 procs */
error = SYSCTL_OUT(req, 0, sizeof (struct kinfo_proc) * 5);
if (error)
return (error);
}
for (doingzomb=0 ; doingzomb < 2 ; doingzomb++) {
if (!doingzomb)
p = allproc.lh_first;
else
p = zombproc.lh_first;
for (; p != 0; p = p->p_list.le_next) {
/*
* Skip embryonic processes.
*/
if (p->p_stat == SIDL)
continue;
/*
* TODO - make more efficient (see notes below).
* do by session.
*/
switch (oidp->oid_number) {
case KERN_PROC_PGRP:
/* could do this by traversing pgrp */
if (p->p_pgrp == NULL ||
p->p_pgrp->pg_id != (pid_t)name[0])
continue;
break;
case KERN_PROC_TTY:
if ((p->p_flag & P_CONTROLT) == 0 ||
p->p_session == NULL ||
p->p_session->s_ttyp == NULL ||
p->p_session->s_ttyp->t_dev != (dev_t)name[0])
continue;
break;
case KERN_PROC_UID:
if (p->p_ucred == NULL ||
p->p_ucred->cr_uid != (uid_t)name[0])
continue;
break;
case KERN_PROC_RUID:
if (p->p_ucred == NULL ||
p->p_cred->p_ruid != (uid_t)name[0])
continue;
break;
}
if (!PRISON_CHECK(curproc, p))
continue;
error = sysctl_out_proc(p, req, doingzomb);
if (error)
return (error);
}
}
return (0);
}
SYSCTL_NODE(_kern, KERN_PROC, proc, CTLFLAG_RD, 0, "Process table");
SYSCTL_PROC(_kern_proc, KERN_PROC_ALL, all, CTLFLAG_RD|CTLTYPE_STRUCT,
0, 0, sysctl_kern_proc, "S,proc", "");
SYSCTL_NODE(_kern_proc, KERN_PROC_PGRP, pgrp, CTLFLAG_RD,
sysctl_kern_proc, "Process table");
SYSCTL_NODE(_kern_proc, KERN_PROC_TTY, tty, CTLFLAG_RD,
sysctl_kern_proc, "Process table");
SYSCTL_NODE(_kern_proc, KERN_PROC_UID, uid, CTLFLAG_RD,
sysctl_kern_proc, "Process table");
SYSCTL_NODE(_kern_proc, KERN_PROC_RUID, ruid, CTLFLAG_RD,
sysctl_kern_proc, "Process table");
SYSCTL_NODE(_kern_proc, KERN_PROC_PID, pid, CTLFLAG_RD,
sysctl_kern_proc, "Process table");