1999-05-04 18:20:53 +00:00
|
|
|
/*
|
|
|
|
* ----------------------------------------------------------------------------
|
|
|
|
* "THE BEER-WARE LICENSE" (Revision 42):
|
|
|
|
* <phk@FreeBSD.ORG> wrote this file. As long as you retain this notice you
|
|
|
|
* can do whatever you want with this stuff. If we meet some day, and you think
|
|
|
|
* this stuff is worth it, you can buy me a beer in return. Poul-Henning Kamp
|
|
|
|
* ----------------------------------------------------------------------------
|
|
|
|
*/
|
|
|
|
|
2003-07-06 12:44:11 +00:00
|
|
|
#include <sys/cdefs.h>
|
|
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
|
2003-03-27 12:16:58 +00:00
|
|
|
#include <sys/param.h>
|
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
|
|
|
#include <sys/jail.h>
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
#include <sys/queue.h>
|
|
|
|
#include <sys/socket.h>
|
2006-05-11 13:04:23 +00:00
|
|
|
#include <sys/sysctl.h>
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
#include <sys/types.h>
|
2001-06-24 20:28:19 +00:00
|
|
|
|
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
|
|
|
#include <netinet/in.h>
|
2001-06-24 20:28:19 +00:00
|
|
|
#include <arpa/inet.h>
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
#include <netdb.h>
|
2001-06-24 20:28:19 +00:00
|
|
|
|
|
|
|
#include <err.h>
|
2004-05-29 18:39:27 +00:00
|
|
|
#include <errno.h>
|
2003-03-27 12:16:58 +00:00
|
|
|
#include <grp.h>
|
|
|
|
#include <login_cap.h>
|
2004-08-15 08:21:50 +00:00
|
|
|
#include <paths.h>
|
2003-03-27 12:16:58 +00:00
|
|
|
#include <pwd.h>
|
2001-06-24 20:28:19 +00:00
|
|
|
#include <stdio.h>
|
|
|
|
#include <stdlib.h>
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
#include <strings.h>
|
2001-06-24 20:28:19 +00:00
|
|
|
#include <string.h>
|
|
|
|
#include <unistd.h>
|
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
|
|
|
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
static void usage(void);
|
|
|
|
static int add_addresses(struct addrinfo *);
|
|
|
|
static struct in_addr *copy_addr4(void);
|
|
|
|
#ifdef INET6
|
|
|
|
static struct in6_addr *copy_addr6(void);
|
|
|
|
#endif
|
|
|
|
|
2004-08-15 08:21:50 +00:00
|
|
|
extern char **environ;
|
2003-03-27 12:16:58 +00:00
|
|
|
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
struct addr4entry {
|
|
|
|
STAILQ_ENTRY(addr4entry) addr4entries;
|
|
|
|
struct in_addr ip4;
|
|
|
|
int count;
|
|
|
|
};
|
|
|
|
struct addr6entry {
|
|
|
|
STAILQ_ENTRY(addr6entry) addr6entries;
|
|
|
|
#ifdef INET6
|
|
|
|
struct in6_addr ip6;
|
|
|
|
#endif
|
|
|
|
int count;
|
|
|
|
};
|
|
|
|
STAILQ_HEAD(addr4head, addr4entry) addr4 = STAILQ_HEAD_INITIALIZER(addr4);
|
|
|
|
STAILQ_HEAD(addr6head, addr6entry) addr6 = STAILQ_HEAD_INITIALIZER(addr6);
|
|
|
|
|
2004-05-29 18:39:27 +00:00
|
|
|
#define GET_USER_INFO do { \
|
|
|
|
pwd = getpwnam(username); \
|
|
|
|
if (pwd == NULL) { \
|
|
|
|
if (errno) \
|
|
|
|
err(1, "getpwnam: %s", username); \
|
|
|
|
else \
|
|
|
|
errx(1, "%s: no such user", username); \
|
|
|
|
} \
|
|
|
|
lcap = login_getpwclass(pwd); \
|
|
|
|
if (lcap == NULL) \
|
|
|
|
err(1, "getpwclass: %s", username); \
|
|
|
|
ngroups = NGROUPS; \
|
|
|
|
if (getgrouplist(username, pwd->pw_gid, groups, &ngroups) != 0) \
|
|
|
|
err(1, "getgrouplist: %s", username); \
|
|
|
|
} while (0)
|
|
|
|
|
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
|
|
|
int
|
|
|
|
main(int argc, char **argv)
|
|
|
|
{
|
2004-11-17 10:01:48 +00:00
|
|
|
login_cap_t *lcap = NULL;
|
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
|
|
|
struct jail j;
|
2004-11-17 10:01:48 +00:00
|
|
|
struct passwd *pwd = NULL;
|
2004-10-02 11:40:48 +00:00
|
|
|
gid_t groups[NGROUPS];
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
int ch, error, i, ngroups, securelevel;
|
|
|
|
int hflag, iflag, Jflag, lflag, uflag, Uflag;
|
|
|
|
char path[PATH_MAX], *jailname, *ep, *username, *JidFile, *ip;
|
2004-08-15 08:21:50 +00:00
|
|
|
static char *cleanenv;
|
2004-11-17 09:52:10 +00:00
|
|
|
const char *shell, *p = NULL;
|
2006-05-12 15:14:43 +00:00
|
|
|
long ltmp;
|
2005-12-03 17:32:39 +00:00
|
|
|
FILE *fp;
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
struct addrinfo hints, *res0;
|
2003-03-27 12:16:58 +00:00
|
|
|
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
hflag = iflag = Jflag = lflag = uflag = Uflag = 0;
|
2006-05-12 15:14:43 +00:00
|
|
|
securelevel = -1;
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
jailname = username = JidFile = cleanenv = NULL;
|
2005-12-03 17:32:39 +00:00
|
|
|
fp = NULL;
|
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
|
|
|
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
while ((ch = getopt(argc, argv, "hiln:s:u:U:J:")) != -1) {
|
2003-03-27 12:16:58 +00:00
|
|
|
switch (ch) {
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
case 'h':
|
|
|
|
hflag = 1;
|
|
|
|
break;
|
2003-04-09 03:04:12 +00:00
|
|
|
case 'i':
|
|
|
|
iflag = 1;
|
|
|
|
break;
|
2005-12-03 17:32:39 +00:00
|
|
|
case 'J':
|
|
|
|
JidFile = optarg;
|
|
|
|
Jflag = 1;
|
|
|
|
break;
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
case 'n':
|
|
|
|
jailname = optarg;
|
|
|
|
break;
|
2006-05-11 13:04:23 +00:00
|
|
|
case 's':
|
2006-05-12 15:14:43 +00:00
|
|
|
ltmp = strtol(optarg, &ep, 0);
|
|
|
|
if (*ep || ep == optarg || ltmp > INT_MAX || !ltmp)
|
|
|
|
errx(1, "invalid securelevel: `%s'", optarg);
|
|
|
|
securelevel = ltmp;
|
2006-05-11 13:04:23 +00:00
|
|
|
break;
|
2003-03-27 12:16:58 +00:00
|
|
|
case 'u':
|
|
|
|
username = optarg;
|
2004-05-29 18:39:27 +00:00
|
|
|
uflag = 1;
|
|
|
|
break;
|
|
|
|
case 'U':
|
|
|
|
username = optarg;
|
|
|
|
Uflag = 1;
|
2003-03-27 12:16:58 +00:00
|
|
|
break;
|
2004-08-15 08:21:50 +00:00
|
|
|
case 'l':
|
|
|
|
lflag = 1;
|
|
|
|
break;
|
2003-03-27 12:16:58 +00:00
|
|
|
default:
|
|
|
|
usage();
|
|
|
|
}
|
2003-04-09 03:04:12 +00:00
|
|
|
}
|
2003-03-27 12:16:58 +00:00
|
|
|
argc -= optind;
|
|
|
|
argv += optind;
|
|
|
|
if (argc < 4)
|
|
|
|
usage();
|
2004-05-29 18:39:27 +00:00
|
|
|
if (uflag && Uflag)
|
|
|
|
usage();
|
2004-08-15 08:21:50 +00:00
|
|
|
if (lflag && username == NULL)
|
|
|
|
usage();
|
2004-05-29 18:39:27 +00:00
|
|
|
if (uflag)
|
|
|
|
GET_USER_INFO;
|
2004-06-27 10:10:16 +00:00
|
|
|
if (realpath(argv[0], path) == NULL)
|
|
|
|
err(1, "realpath: %s", argv[0]);
|
|
|
|
if (chdir(path) != 0)
|
|
|
|
err(1, "chdir: %s", path);
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
/* Initialize struct jail. */
|
1999-09-19 08:36:37 +00:00
|
|
|
memset(&j, 0, sizeof(j));
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
j.version = JAIL_API_VERSION;
|
2004-06-27 10:10:16 +00:00
|
|
|
j.path = path;
|
2003-03-27 12:16:58 +00:00
|
|
|
j.hostname = argv[1];
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
if (jailname != NULL)
|
|
|
|
j.jailname = jailname;
|
|
|
|
|
|
|
|
/* Handle IP addresses. If requested resolve hostname too. */
|
|
|
|
bzero(&hints, sizeof(struct addrinfo));
|
|
|
|
hints.ai_protocol = IPPROTO_TCP;
|
|
|
|
hints.ai_socktype = SOCK_STREAM;
|
|
|
|
if (JAIL_API_VERSION < 2)
|
|
|
|
hints.ai_family = PF_INET;
|
|
|
|
else
|
|
|
|
hints.ai_family = PF_UNSPEC;
|
|
|
|
/* Handle hostname. */
|
|
|
|
if (hflag != 0) {
|
|
|
|
error = getaddrinfo(j.hostname, NULL, &hints, &res0);
|
|
|
|
if (error != 0)
|
|
|
|
errx(1, "failed to handle hostname: %s",
|
|
|
|
gai_strerror(error));
|
|
|
|
error = add_addresses(res0);
|
|
|
|
freeaddrinfo(res0);
|
|
|
|
if (error != 0)
|
|
|
|
errx(1, "failed to add addresses.");
|
|
|
|
}
|
|
|
|
/* Handle IP addresses. */
|
|
|
|
hints.ai_flags = AI_NUMERICHOST;
|
|
|
|
ip = strtok(argv[2], ",");
|
|
|
|
while (ip != NULL) {
|
|
|
|
error = getaddrinfo(ip, NULL, &hints, &res0);
|
|
|
|
if (error != 0)
|
|
|
|
errx(1, "failed to handle ip: %s", gai_strerror(error));
|
|
|
|
error = add_addresses(res0);
|
|
|
|
freeaddrinfo(res0);
|
|
|
|
if (error != 0)
|
|
|
|
errx(1, "failed to add addresses.");
|
|
|
|
ip = strtok(NULL, ",");
|
|
|
|
}
|
|
|
|
/* Count IP addresses and add them to struct jail. */
|
|
|
|
if (!STAILQ_EMPTY(&addr4)) {
|
|
|
|
j.ip4s = STAILQ_FIRST(&addr4)->count;
|
|
|
|
j.ip4 = copy_addr4();
|
|
|
|
if (j.ip4s > 0 && j.ip4 == NULL)
|
|
|
|
errx(1, "copy_addr4()");
|
|
|
|
}
|
|
|
|
#ifdef INET6
|
|
|
|
if (!STAILQ_EMPTY(&addr6)) {
|
|
|
|
j.ip6s = STAILQ_FIRST(&addr6)->count;
|
|
|
|
j.ip6 = copy_addr6();
|
|
|
|
if (j.ip6s > 0 && j.ip6 == NULL)
|
|
|
|
errx(1, "copy_addr6()");
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2005-12-03 17:32:39 +00:00
|
|
|
if (Jflag) {
|
|
|
|
fp = fopen(JidFile, "w");
|
|
|
|
if (fp == NULL)
|
|
|
|
errx(1, "Could not create JidFile: %s", JidFile);
|
|
|
|
}
|
2003-04-09 03:04:12 +00:00
|
|
|
i = jail(&j);
|
|
|
|
if (i == -1)
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
err(1, "syscall failed with");
|
2003-04-21 17:20:48 +00:00
|
|
|
if (iflag) {
|
2003-04-09 03:04:12 +00:00
|
|
|
printf("%d\n", i);
|
2003-04-21 17:20:48 +00:00
|
|
|
fflush(stdout);
|
|
|
|
}
|
2005-12-03 17:32:39 +00:00
|
|
|
if (Jflag) {
|
|
|
|
if (fp != NULL) {
|
|
|
|
fprintf(fp, "%d\t%s\t%s\t%s\t%s\n",
|
|
|
|
i, j.path, j.hostname, argv[2], argv[3]);
|
|
|
|
(void)fclose(fp);
|
|
|
|
} else {
|
|
|
|
errx(1, "Could not write JidFile: %s", JidFile);
|
|
|
|
}
|
|
|
|
}
|
2006-05-11 19:06:33 +00:00
|
|
|
if (securelevel > 0) {
|
|
|
|
if (sysctlbyname("kern.securelevel", NULL, 0, &securelevel,
|
|
|
|
sizeof(securelevel)))
|
|
|
|
err(1, "Can not set securelevel to %d", securelevel);
|
|
|
|
}
|
2003-03-27 12:16:58 +00:00
|
|
|
if (username != NULL) {
|
2004-05-29 18:39:27 +00:00
|
|
|
if (Uflag)
|
|
|
|
GET_USER_INFO;
|
2004-08-15 08:21:50 +00:00
|
|
|
if (lflag) {
|
|
|
|
p = getenv("TERM");
|
|
|
|
environ = &cleanenv;
|
|
|
|
}
|
2003-04-02 09:20:08 +00:00
|
|
|
if (setgroups(ngroups, groups) != 0)
|
|
|
|
err(1, "setgroups");
|
|
|
|
if (setgid(pwd->pw_gid) != 0)
|
|
|
|
err(1, "setgid");
|
|
|
|
if (setusercontext(lcap, pwd, pwd->pw_uid,
|
2006-04-16 12:32:04 +00:00
|
|
|
LOGIN_SETALL & ~LOGIN_SETGROUP & ~LOGIN_SETLOGIN) != 0)
|
2003-04-02 09:20:08 +00:00
|
|
|
err(1, "setusercontext");
|
2003-04-07 10:16:37 +00:00
|
|
|
login_close(lcap);
|
2003-03-27 12:16:58 +00:00
|
|
|
}
|
2004-08-15 08:21:50 +00:00
|
|
|
if (lflag) {
|
|
|
|
if (*pwd->pw_shell)
|
|
|
|
shell = pwd->pw_shell;
|
|
|
|
else
|
|
|
|
shell = _PATH_BSHELL;
|
|
|
|
if (chdir(pwd->pw_dir) < 0)
|
|
|
|
errx(1, "no home directory");
|
|
|
|
setenv("HOME", pwd->pw_dir, 1);
|
|
|
|
setenv("SHELL", shell, 1);
|
|
|
|
setenv("USER", pwd->pw_name, 1);
|
|
|
|
if (p)
|
|
|
|
setenv("TERM", p, 1);
|
|
|
|
}
|
2003-04-02 09:20:08 +00:00
|
|
|
if (execv(argv[3], argv + 3) != 0)
|
|
|
|
err(1, "execv: %s", argv[3]);
|
2003-04-09 03:04:12 +00:00
|
|
|
exit(0);
|
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
|
|
|
}
|
2003-03-27 12:16:58 +00:00
|
|
|
|
|
|
|
static void
|
|
|
|
usage(void)
|
|
|
|
{
|
|
|
|
|
2006-05-11 13:04:23 +00:00
|
|
|
(void)fprintf(stderr, "%s%s%s\n",
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
"usage: jail [-hi] [-n jailname] [-J jid_file] ",
|
|
|
|
"[-s securelevel] [-l -u username | -U username] ",
|
|
|
|
"path hostname [ip[,..]] command ...");
|
2003-04-02 09:20:08 +00:00
|
|
|
exit(1);
|
2003-03-27 12:16:58 +00:00
|
|
|
}
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
|
|
|
|
static int
|
|
|
|
add_addresses(struct addrinfo *res0)
|
|
|
|
{
|
|
|
|
int error;
|
|
|
|
struct addrinfo *res;
|
|
|
|
struct addr4entry *a4p;
|
|
|
|
struct sockaddr_in *sai;
|
|
|
|
#ifdef INET6
|
|
|
|
struct addr6entry *a6p;
|
|
|
|
struct sockaddr_in6 *sai6;
|
|
|
|
#endif
|
|
|
|
int count;
|
|
|
|
|
|
|
|
error = 0;
|
|
|
|
for (res = res0; res && error == 0; res = res->ai_next) {
|
|
|
|
switch (res->ai_family) {
|
|
|
|
case AF_INET:
|
|
|
|
sai = (struct sockaddr_in *)(void *)res->ai_addr;
|
|
|
|
STAILQ_FOREACH(a4p, &addr4, addr4entries) {
|
|
|
|
if (bcmp(&sai->sin_addr, &a4p->ip4,
|
|
|
|
sizeof(struct in_addr)) == 0) {
|
|
|
|
err(1, "Ignoring duplicate IPv4 address.");
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
a4p = (struct addr4entry *) malloc(
|
|
|
|
sizeof(struct addr4entry));
|
|
|
|
if (a4p == NULL) {
|
|
|
|
error = 1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
bzero(a4p, sizeof(struct addr4entry));
|
|
|
|
bcopy(&sai->sin_addr, &a4p->ip4,
|
|
|
|
sizeof(struct in_addr));
|
|
|
|
if (!STAILQ_EMPTY(&addr4))
|
|
|
|
count = STAILQ_FIRST(&addr4)->count;
|
|
|
|
else
|
|
|
|
count = 0;
|
|
|
|
STAILQ_INSERT_TAIL(&addr4, a4p, addr4entries);
|
|
|
|
STAILQ_FIRST(&addr4)->count = count + 1;
|
|
|
|
break;
|
|
|
|
#ifdef INET6
|
|
|
|
case AF_INET6:
|
|
|
|
sai6 = (struct sockaddr_in6 *)(void *)res->ai_addr;
|
|
|
|
STAILQ_FOREACH(a6p, &addr6, addr6entries) {
|
|
|
|
if (bcmp(&sai6->sin6_addr, &a6p->ip6,
|
|
|
|
sizeof(struct in6_addr)) == 0) {
|
|
|
|
err(1, "Ignoring duplicate IPv6 address.");
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
a6p = (struct addr6entry *) malloc(
|
|
|
|
sizeof(struct addr6entry));
|
|
|
|
if (a6p == NULL) {
|
|
|
|
error = 1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
bzero(a6p, sizeof(struct addr6entry));
|
|
|
|
bcopy(&sai6->sin6_addr, &a6p->ip6,
|
|
|
|
sizeof(struct in6_addr));
|
|
|
|
if (!STAILQ_EMPTY(&addr6))
|
|
|
|
count = STAILQ_FIRST(&addr6)->count;
|
|
|
|
else
|
|
|
|
count = 0;
|
|
|
|
STAILQ_INSERT_TAIL(&addr6, a6p, addr6entries);
|
|
|
|
STAILQ_FIRST(&addr6)->count = count + 1;
|
|
|
|
break;
|
|
|
|
#endif
|
|
|
|
default:
|
|
|
|
err(1, "Address family %d not supported. Ignoring.\n",
|
|
|
|
res->ai_family);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct in_addr *
|
|
|
|
copy_addr4(void)
|
|
|
|
{
|
|
|
|
size_t len;
|
|
|
|
struct in_addr *ip4s, *p, ia;
|
|
|
|
struct addr4entry *a4p;
|
|
|
|
|
|
|
|
if (STAILQ_EMPTY(&addr4))
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
len = STAILQ_FIRST(&addr4)->count * sizeof(struct in_addr);
|
|
|
|
|
|
|
|
ip4s = p = (struct in_addr *)malloc(len);
|
|
|
|
if (ip4s == NULL)
|
|
|
|
return (NULL);
|
|
|
|
|
|
|
|
bzero(p, len);
|
|
|
|
|
|
|
|
while (!STAILQ_EMPTY(&addr4)) {
|
|
|
|
a4p = STAILQ_FIRST(&addr4);
|
|
|
|
STAILQ_REMOVE_HEAD(&addr4, addr4entries);
|
|
|
|
ia.s_addr = a4p->ip4.s_addr;
|
|
|
|
bcopy(&ia, p, sizeof(struct in_addr));
|
|
|
|
p++;
|
|
|
|
free(a4p);
|
|
|
|
}
|
|
|
|
|
|
|
|
return (ip4s);
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef INET6
|
|
|
|
static struct in6_addr *
|
|
|
|
copy_addr6(void)
|
|
|
|
{
|
|
|
|
size_t len;
|
|
|
|
struct in6_addr *ip6s, *p;
|
|
|
|
struct addr6entry *a6p;
|
|
|
|
|
|
|
|
if (STAILQ_EMPTY(&addr6))
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
len = STAILQ_FIRST(&addr6)->count * sizeof(struct in6_addr);
|
|
|
|
|
|
|
|
ip6s = p = (struct in6_addr *)malloc(len);
|
|
|
|
if (ip6s == NULL)
|
|
|
|
return (NULL);
|
|
|
|
|
|
|
|
bzero(p, len);
|
|
|
|
|
|
|
|
while (!STAILQ_EMPTY(&addr6)) {
|
|
|
|
a6p = STAILQ_FIRST(&addr6);
|
|
|
|
STAILQ_REMOVE_HEAD(&addr6, addr6entries);
|
|
|
|
bcopy(&a6p->ip6, p, sizeof(struct in6_addr));
|
|
|
|
p++;
|
|
|
|
free(a6p);
|
|
|
|
}
|
|
|
|
|
|
|
|
return (ip6s);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|