freebsd-nq/sys/kern/kern_ktrace.c

532 lines
12 KiB
C
Raw Normal View History

1994-05-24 10:09:53 +00:00
/*
* Copyright (c) 1989, 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_ktrace.c 8.2 (Berkeley) 9/23/93
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
* $Id: kern_ktrace.c,v 1.25 1998/12/10 01:47:41 rvb Exp $
1994-05-24 10:09:53 +00:00
*/
#include "opt_ktrace.h"
1994-05-24 10:09:53 +00:00
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/sysproto.h>
#include <sys/kernel.h>
1994-05-24 10:09:53 +00:00
#include <sys/proc.h>
#include <sys/fcntl.h>
#include <sys/lock.h>
1994-05-24 10:09:53 +00:00
#include <sys/namei.h>
#include <sys/vnode.h>
#include <sys/ktrace.h>
#include <sys/malloc.h>
#include <sys/syslog.h>
static MALLOC_DEFINE(M_KTRACE, "KTRACE", "KTRACE");
#ifdef KTRACE
static struct ktr_header *ktrgetheader __P((int type));
static void ktrwrite __P((struct vnode *, struct ktr_header *));
static int ktrcanset __P((struct proc *,struct proc *));
static int ktrsetchildren __P((struct proc *,struct proc *,int,int,struct vnode *));
static int ktrops __P((struct proc *,struct proc *,int,int,struct vnode *));
static struct ktr_header *
1994-05-24 10:09:53 +00:00
ktrgetheader(type)
int type;
{
register struct ktr_header *kth;
struct proc *p = curproc; /* XXX */
1995-05-30 08:16:23 +00:00
MALLOC(kth, struct ktr_header *, sizeof (struct ktr_header),
M_KTRACE, M_WAITOK);
1994-05-24 10:09:53 +00:00
kth->ktr_type = type;
microtime(&kth->ktr_time);
kth->ktr_pid = p->p_pid;
bcopy(p->p_comm, kth->ktr_comm, MAXCOMLEN);
return (kth);
}
void
1994-05-24 10:09:53 +00:00
ktrsyscall(vp, code, narg, args)
struct vnode *vp;
int code, narg, args[];
{
struct ktr_header *kth;
struct ktr_syscall *ktp;
1998-02-20 13:37:40 +00:00
register int len = sizeof(struct ktr_syscall) + (narg * sizeof(int));
1994-05-24 10:09:53 +00:00
struct proc *p = curproc; /* XXX */
int *argp, i;
p->p_traceflag |= KTRFAC_ACTIVE;
kth = ktrgetheader(KTR_SYSCALL);
MALLOC(ktp, struct ktr_syscall *, len, M_KTRACE, M_WAITOK);
1994-05-24 10:09:53 +00:00
ktp->ktr_code = code;
ktp->ktr_narg = narg;
argp = (int *)((char *)ktp + sizeof(struct ktr_syscall));
for (i = 0; i < narg; i++)
*argp++ = args[i];
kth->ktr_buf = (caddr_t)ktp;
kth->ktr_len = len;
ktrwrite(vp, kth);
FREE(ktp, M_KTRACE);
FREE(kth, M_KTRACE);
1994-05-24 10:09:53 +00:00
p->p_traceflag &= ~KTRFAC_ACTIVE;
}
void
1994-05-24 10:09:53 +00:00
ktrsysret(vp, code, error, retval)
struct vnode *vp;
int code, error, retval;
{
struct ktr_header *kth;
struct ktr_sysret ktp;
struct proc *p = curproc; /* XXX */
p->p_traceflag |= KTRFAC_ACTIVE;
kth = ktrgetheader(KTR_SYSRET);
ktp.ktr_code = code;
ktp.ktr_error = error;
ktp.ktr_retval = retval; /* what about val2 ? */
kth->ktr_buf = (caddr_t)&ktp;
kth->ktr_len = sizeof(struct ktr_sysret);
ktrwrite(vp, kth);
FREE(kth, M_KTRACE);
1994-05-24 10:09:53 +00:00
p->p_traceflag &= ~KTRFAC_ACTIVE;
}
void
1994-05-24 10:09:53 +00:00
ktrnamei(vp, path)
struct vnode *vp;
char *path;
{
struct ktr_header *kth;
struct proc *p = curproc; /* XXX */
p->p_traceflag |= KTRFAC_ACTIVE;
kth = ktrgetheader(KTR_NAMEI);
kth->ktr_len = strlen(path);
kth->ktr_buf = path;
ktrwrite(vp, kth);
FREE(kth, M_KTRACE);
1994-05-24 10:09:53 +00:00
p->p_traceflag &= ~KTRFAC_ACTIVE;
}
void
1994-05-24 10:09:53 +00:00
ktrgenio(vp, fd, rw, iov, len, error)
struct vnode *vp;
int fd;
enum uio_rw rw;
register struct iovec *iov;
int len, error;
{
struct ktr_header *kth;
register struct ktr_genio *ktp;
register caddr_t cp;
register int resid = len, cnt;
struct proc *p = curproc; /* XXX */
1995-05-30 08:16:23 +00:00
1994-05-24 10:09:53 +00:00
if (error)
return;
p->p_traceflag |= KTRFAC_ACTIVE;
kth = ktrgetheader(KTR_GENIO);
MALLOC(ktp, struct ktr_genio *, sizeof(struct ktr_genio) + len,
M_KTRACE, M_WAITOK);
1994-05-24 10:09:53 +00:00
ktp->ktr_fd = fd;
ktp->ktr_rw = rw;
cp = (caddr_t)((char *)ktp + sizeof (struct ktr_genio));
while (resid > 0) {
if ((cnt = iov->iov_len) > resid)
cnt = resid;
if (copyin(iov->iov_base, cp, (unsigned)cnt))
goto done;
cp += cnt;
resid -= cnt;
iov++;
}
kth->ktr_buf = (caddr_t)ktp;
kth->ktr_len = sizeof (struct ktr_genio) + len;
ktrwrite(vp, kth);
done:
FREE(kth, M_KTRACE);
FREE(ktp, M_KTRACE);
1994-05-24 10:09:53 +00:00
p->p_traceflag &= ~KTRFAC_ACTIVE;
}
void
1994-05-24 10:09:53 +00:00
ktrpsig(vp, sig, action, mask, code)
struct vnode *vp;
int sig;
sig_t action;
int mask, code;
{
struct ktr_header *kth;
struct ktr_psig kp;
struct proc *p = curproc; /* XXX */
p->p_traceflag |= KTRFAC_ACTIVE;
kth = ktrgetheader(KTR_PSIG);
kp.signo = (char)sig;
kp.action = action;
kp.mask = mask;
kp.code = code;
kth->ktr_buf = (caddr_t)&kp;
kth->ktr_len = sizeof (struct ktr_psig);
ktrwrite(vp, kth);
FREE(kth, M_KTRACE);
1994-05-24 10:09:53 +00:00
p->p_traceflag &= ~KTRFAC_ACTIVE;
}
void
1994-05-24 10:09:53 +00:00
ktrcsw(vp, out, user)
struct vnode *vp;
int out, user;
{
struct ktr_header *kth;
struct ktr_csw kc;
struct proc *p = curproc; /* XXX */
p->p_traceflag |= KTRFAC_ACTIVE;
kth = ktrgetheader(KTR_CSW);
kc.out = out;
kc.user = user;
kth->ktr_buf = (caddr_t)&kc;
kth->ktr_len = sizeof (struct ktr_csw);
ktrwrite(vp, kth);
FREE(kth, M_KTRACE);
1994-05-24 10:09:53 +00:00
p->p_traceflag &= ~KTRFAC_ACTIVE;
}
#endif
1994-05-24 10:09:53 +00:00
/* Interface and common routines */
/*
* ktrace system call
*/
#ifndef _SYS_SYSPROTO_H_
1994-05-24 10:09:53 +00:00
struct ktrace_args {
char *fname;
int ops;
int facs;
int pid;
};
#endif
1994-05-24 10:09:53 +00:00
/* ARGSUSED */
int
ktrace(curp, uap)
1994-05-24 10:09:53 +00:00
struct proc *curp;
register struct ktrace_args *uap;
{
#ifdef KTRACE
1994-05-24 10:09:53 +00:00
register struct vnode *vp = NULL;
register struct proc *p;
struct pgrp *pg;
int facs = uap->facs & ~KTRFAC_ROOT;
int ops = KTROP(uap->ops);
int descend = uap->ops & KTRFLAG_DESCEND;
int ret = 0;
int error = 0;
struct nameidata nd;
curp->p_traceflag |= KTRFAC_ACTIVE;
if (ops != KTROP_CLEAR) {
/*
* an operation which requires a file argument.
*/
NDINIT(&nd, LOOKUP, FOLLOW, UIO_USERSPACE, uap->fname, curp);
error = vn_open(&nd, FREAD|FWRITE, 0);
if (error) {
1994-05-24 10:09:53 +00:00
curp->p_traceflag &= ~KTRFAC_ACTIVE;
return (error);
}
vp = nd.ni_vp;
VOP_UNLOCK(vp, 0, curp);
1994-05-24 10:09:53 +00:00
if (vp->v_type != VREG) {
(void) vn_close(vp, FREAD|FWRITE, curp->p_ucred, curp);
curp->p_traceflag &= ~KTRFAC_ACTIVE;
return (EACCES);
}
}
/*
* Clear all uses of the tracefile
*/
if (ops == KTROP_CLEARFILE) {
for (p = allproc.lh_first; p != 0; p = p->p_list.le_next) {
1994-05-24 10:09:53 +00:00
if (p->p_tracep == vp) {
if (ktrcanset(curp, p)) {
p->p_tracep = NULL;
p->p_traceflag = 0;
(void) vn_close(vp, FREAD|FWRITE,
p->p_ucred, p);
} else
error = EPERM;
}
}
goto done;
}
/*
* need something to (un)trace (XXX - why is this here?)
*/
if (!facs) {
error = EINVAL;
goto done;
}
1995-05-30 08:16:23 +00:00
/*
1994-05-24 10:09:53 +00:00
* do it
*/
if (uap->pid < 0) {
/*
* by process group
*/
pg = pgfind(-uap->pid);
if (pg == NULL) {
error = ESRCH;
goto done;
}
for (p = pg->pg_members.lh_first; p != 0; p = p->p_pglist.le_next)
1994-05-24 10:09:53 +00:00
if (descend)
ret |= ktrsetchildren(curp, p, ops, facs, vp);
1995-05-30 08:16:23 +00:00
else
1994-05-24 10:09:53 +00:00
ret |= ktrops(curp, p, ops, facs, vp);
1995-05-30 08:16:23 +00:00
1994-05-24 10:09:53 +00:00
} else {
/*
* by pid
*/
p = pfind(uap->pid);
if (p == NULL) {
error = ESRCH;
goto done;
}
if (descend)
ret |= ktrsetchildren(curp, p, ops, facs, vp);
else
ret |= ktrops(curp, p, ops, facs, vp);
}
if (!ret)
error = EPERM;
done:
if (vp != NULL)
(void) vn_close(vp, FWRITE, curp->p_ucred, curp);
curp->p_traceflag &= ~KTRFAC_ACTIVE;
return (error);
#else
return ENOSYS;
#endif
1994-05-24 10:09:53 +00:00
}
/*
* utrace system call
*/
/* ARGSUSED */
int
utrace(curp, uap)
struct proc *curp;
register struct utrace_args *uap;
{
#ifdef KTRACE
struct ktr_header *kth;
struct proc *p = curproc; /* XXX */
register caddr_t cp;
if (!KTRPOINT(p, KTR_USER))
return (0);
p->p_traceflag |= KTRFAC_ACTIVE;
kth = ktrgetheader(KTR_USER);
MALLOC(cp, caddr_t, uap->len, M_KTRACE, M_WAITOK);
if (!copyin(uap->addr, cp, uap->len)) {
kth->ktr_buf = cp;
kth->ktr_len = uap->len;
ktrwrite(p->p_tracep, kth);
}
FREE(kth, M_KTRACE);
FREE(cp, M_KTRACE);
p->p_traceflag &= ~KTRFAC_ACTIVE;
return (0);
#else
return (ENOSYS);
#endif
}
#ifdef KTRACE
static int
1994-05-24 10:09:53 +00:00
ktrops(curp, p, ops, facs, vp)
struct proc *p, *curp;
int ops, facs;
struct vnode *vp;
{
if (!ktrcanset(curp, p))
return (0);
if (ops == KTROP_SET) {
1995-05-30 08:16:23 +00:00
if (p->p_tracep != vp) {
1994-05-24 10:09:53 +00:00
/*
* if trace file already in use, relinquish
*/
if (p->p_tracep != NULL)
vrele(p->p_tracep);
VREF(vp);
p->p_tracep = vp;
}
p->p_traceflag |= facs;
if (curp->p_ucred->cr_uid == 0)
p->p_traceflag |= KTRFAC_ROOT;
1995-05-30 08:16:23 +00:00
} else {
1994-05-24 10:09:53 +00:00
/* KTROP_CLEAR */
if (((p->p_traceflag &= ~facs) & KTRFAC_MASK) == 0) {
/* no more tracing */
p->p_traceflag = 0;
if (p->p_tracep != NULL) {
vrele(p->p_tracep);
p->p_tracep = NULL;
}
}
}
return (1);
}
static int
1994-05-24 10:09:53 +00:00
ktrsetchildren(curp, top, ops, facs, vp)
struct proc *curp, *top;
int ops, facs;
struct vnode *vp;
{
register struct proc *p;
register int ret = 0;
p = top;
for (;;) {
ret |= ktrops(curp, p, ops, facs, vp);
/*
* If this process has children, descend to them next,
* otherwise do any siblings, and if done with this level,
* follow back up the tree (but not past top).
*/
if (p->p_children.lh_first)
p = p->p_children.lh_first;
1994-05-24 10:09:53 +00:00
else for (;;) {
if (p == top)
return (ret);
if (p->p_sibling.le_next) {
p = p->p_sibling.le_next;
1994-05-24 10:09:53 +00:00
break;
}
p = p->p_pptr;
1994-05-24 10:09:53 +00:00
}
}
/*NOTREACHED*/
}
static void
1994-05-24 10:09:53 +00:00
ktrwrite(vp, kth)
struct vnode *vp;
register struct ktr_header *kth;
{
struct uio auio;
struct iovec aiov[2];
register struct proc *p = curproc; /* XXX */
int error;
if (vp == NULL)
return;
auio.uio_iov = &aiov[0];
auio.uio_offset = 0;
auio.uio_segflg = UIO_SYSSPACE;
auio.uio_rw = UIO_WRITE;
aiov[0].iov_base = (caddr_t)kth;
aiov[0].iov_len = sizeof(struct ktr_header);
auio.uio_resid = sizeof(struct ktr_header);
auio.uio_iovcnt = 1;
auio.uio_procp = curproc;
1994-05-24 10:09:53 +00:00
if (kth->ktr_len > 0) {
auio.uio_iovcnt++;
aiov[1].iov_base = kth->ktr_buf;
aiov[1].iov_len = kth->ktr_len;
auio.uio_resid += kth->ktr_len;
}
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, p);
1994-05-24 10:09:53 +00:00
error = VOP_WRITE(vp, &auio, IO_UNIT|IO_APPEND, p->p_ucred);
VOP_UNLOCK(vp, 0, p);
1994-05-24 10:09:53 +00:00
if (!error)
return;
/*
* If error encountered, give up tracing on this vnode.
*/
log(LOG_NOTICE, "ktrace write failed, errno %d, tracing stopped\n",
error);
for (p = allproc.lh_first; p != 0; p = p->p_list.le_next) {
1994-05-24 10:09:53 +00:00
if (p->p_tracep == vp) {
p->p_tracep = NULL;
p->p_traceflag = 0;
vrele(vp);
}
}
}
/*
* Return true if caller has permission to set the ktracing state
* of target. Essentially, the target can't possess any
* more permissions than the caller. KTRFAC_ROOT signifies that
1995-05-30 08:16:23 +00:00
* root previously set the tracing status on the target process, and
1994-05-24 10:09:53 +00:00
* so, only root may further change it.
*
* TODO: check groups. use caller effective gid.
*/
static int
1994-05-24 10:09:53 +00:00
ktrcanset(callp, targetp)
struct proc *callp, *targetp;
{
register struct pcred *caller = callp->p_cred;
register struct pcred *target = targetp->p_cred;
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
if (!PRISON_CHECK(callp, targetp))
return (0);
1994-05-24 10:09:53 +00:00
if ((caller->pc_ucred->cr_uid == target->p_ruid &&
target->p_ruid == target->p_svuid &&
caller->p_rgid == target->p_rgid && /* XXX */
target->p_rgid == target->p_svgid &&
(targetp->p_traceflag & KTRFAC_ROOT) == 0) ||
caller->pc_ucred->cr_uid == 0)
return (1);
return (0);
}
#endif /* KTRACE */