JAIL(8) BSD System Manager's Manual JAIL(8)NAMEjail — create or modify a system jailSYNOPSISjail [-dhi] [-J jid_file] [-l -u username | -U username] [-c | -m]
[parameter=value ...]
jail [-hi] [-n jailname] [-J jid_file] [-s securelevel]
[-l -u username | -U username] [path hostname [ip[,..]] command ...]
jail [-r jail]
DESCRIPTION
The jail utility creates a new jail or modifies an existing jail, option‐
ally imprisoning the current process (and future descendants) inside it.
The options are as follows:
-d Allow making changes to a dying jail.
-h Resolve the host.hostname parameter (or hostname) and add all IP
addresses returned by the resolver to the list of ip addresses
for this prison. This may affect default address selection for
outgoing IPv4 connections of prisons. The address first returned
by the resolver for each address family will be used as primary
address. See the ip4.addr and ip6.addr parameters further down
for details.
-i Output the jail identifier of the newly created jail.
-n jailname
Set the jail's name. This is deprecated and is equivalent to
setting the name parameter.
-J jid_file
Write a jid_file file, containing jail identifier, path, host‐
name, IP and command used to start the jail.
-l Run program in the clean environment. The environment is dis‐
carded except for HOME, SHELL, TERM and USER. HOME and SHELL are
set to the target login's default values. USER is set to the
target login. TERM is imported from the current environment.
The environment variables from the login class capability data‐
base for the target login are also set.
-s securelevel
Set the kern.securelevel MIB entry to the specified value inside
the newly created jail. This is deprecated and is equivalent to
setting the securelevel parameter.
-u username
The user name from host environment as whom the command should
run.
-U username
The user name from jailed environment as whom the command should
run.
-c Create a new jail. The jid and name parameters (if specified)
must not refer to an existing jail.
-m Modify an existing jail. One of the jid or name parameters must
exist and refer to an existing jail.
-cm Create a jail if it does not exist, or modify a jail if it does
exist.
-r Remove the jail specified by jid or name. All jailed processes
are killed, and all children of this jail are also removed.
At least one of the -c, -m or -r options must be specified.
Parameters are listed in “name=value” form, following the options. Some
parameters are boolean, and do not have a value but are set by the name
alone with or without a “no” prefix, e.g. persist or nopersist. Any
parameters not set will be given default values, often based on the cur‐
rent environment.
The pseudo-parameter command specifies that the current process should
enter the new (or modified) jail, and run the specified command. It must
be the last parameter specified, because it includes not only the value
following the ‘=’ sign, but also passes the rest of the arguments to the
command.
Instead of supplying named parameters, four fixed parameters may be sup‐
plied in order on the command line: path, hostname, ip, and command. As
the jid and name parameters aren't in this list, this mode will always
create a new jail, and the -c and -m options don't apply (and must not
exist).
Jails have a set a core parameters, and modules can add their own jail
parameters. The current set of available parameters can be retrieved via
“sysctl -d security.jail.param”. The core parameters are:
jid The jail identifier. This will be assigned automatically to a
new jail (or can be explicitly set), and can be used to identify
the jail for later modification, or for such commands as jls(8)
or jexec(8).
name The jail name. This is an arbitrary string that identifies a
jail (except it may not contain a ‘.’). Like the jid, it can be
passed to later jail commands, or to jls(8) or jexec(8). If no
name is supplied, a default is assumed that is the same as the
jid.
path Directory which is to be the root of the prison. The command (if
any) is run from this directory, as are commands from jexec(8).
ip4.addr
A comma-separated list of IPv4 addresses assigned to the prison.
If this is set, the jail is restricted to using only these
address. Any attempts to use other addresses fail, and attempts
to use wildcard addresses silently use the jailed address
instead. For IPv4 the first address given will be kept used as
the source address in case source address selection on unbound
sockets cannot find a better match. It is only possible to start
multiple jails with the same IP address, if none of the jails has
more than this single overlapping IP address assigned to itself.
ip4.saddrsel
A boolean option to change the formerly mentioned behaviour and
disable IPv4 source address selection for the prison in favour of
the primary IPv4 address of the jail. Source address selection
is enabled by default for all jails and a ip4.nosaddrsel setting
of a parent jail is not inherited for any child jails.
ip4 Control the availablity of IPv4 addresses. Possible values are
“inherit” to allow unrestricted access to all system addresses,
“new” to restrict addresses via ip4.addr above, and “disable” to
stop the jail from using IPv4 entirely. Setting the ip4.addr
parameter implies a value of “new”.
ip6.addr, ip6.saddrsel, ip6
A set of IPv6 options for the prison, the counterparts to
ip4.addr, ip4.saddrsel and ip4 above.
host.hostname
Hostname of the prison. Other similar parameters are
host.domainname, host.hostuuid and host.hostid.
host Set the origin of hostname and related information. Possible
values are “inherit” to use the system information and “new” for
the jail to use the information from the above fields. Setting
any of the above fields implies a value of “new”.
securelevel
The value of the jail's kern.securelevel sysctl. A jail never
has a lower securelevel than the default system, but by setting
this parameter it may have a higher one. If the system
securelevel is changed, any jail securelevels will be at least as
secure.
children.max
The number of child jails allowed to be created by this jail (or
by other jails under this jail). This limit is zero by default,
indicating the jail is not allowed to create child jails. See
the Hierarchical Jails section for more information.
children.cur
The number of descendents of this jail, including its own child
jails and any jails created under them.
enforce_statfs
This determines which information processes in a jail are able to
get about mount points. It affects the behaviour of the follow‐
ing syscalls: statfs(2), fstatfs(2), getfsstat(2) and fhstatfs(2)
(as well as similar compatibility syscalls). When set to 0, all
mount points are available without any restrictions. When set to
1, only mount points below the jail's chroot directory are visi‐
ble. In addition to that, the path to the jail's chroot direc‐
tory is removed from the front of their pathnames. When set to 2
(default), above syscalls can operate only on a mount-point where
the jail's chroot directory is located.
persist
Setting this boolean parameter allows a jail to exist without any
processes. Normally, a jail is destroyed as its last process
exits. A new jail must have either the persist parameter or
command pseudo-parameter set.
cpuset.id
The ID of the cpuset associated with this jail (read-only).
dying This is true if the jail is in the process of shutting down
(read-only).
parent The jid of the parent of this jail, or zero if this is a top-
level jail (read-only).
allow.*
Some restrictions of the jail environment may be set on a per-
jail basis. With the exception of allow.set_hostname, these
boolean parameters are off by default.
allow.set_hostname
The jail's hostname may be changed via hostname(1) or
sethostname(3).
allow.sysvipc
A process within the jail has access to System V IPC
primitives. In the current jail implementation, System V
primitives share a single namespace across the host and
jail environments, meaning that processes within a jail
would be able to communicate with (and potentially inter‐
fere with) processes outside of the jail, and in other
jails.
allow.raw_sockets
The prison root is allowed to create raw sockets. Set‐
ting this parameter allows utilities like ping(8) and
traceroute(8) to operate inside the prison. If this is
set, the source IP addresses are enforced to comply with
the IP address bound to the jail, regardless of whether
or not the IP_HDRINCL flag has been set on the socket.
Since raw sockets can be used to configure and interact
with various network subsystems, extra caution should be
used where privileged access to jails is given out to
untrusted parties.
allow.chflags
Normally, privileged users inside a jail are treated as
unprivileged by chflags(2). When this parameter is set,
such users are treated as privileged, and may manipulate
system file flags subject to the usual constraints on
kern.securelevel.
allow.mount
privileged users inside the jail will be able to mount
and unmount file system types marked as jail-friendly.
The lsvfs(1) command can be used to find file system
types available for mount from within a jail.
allow.quotas
The prison root may administer quotas on the jail's
filesystem(s). This includes filesystems that the jail
may share with other jails or with non-jailed parts of
the system.
allow.socket_af
Sockets within a jail are normally restricted to IPv4,
IPv6, local (UNIX), and route. This allows access to
other protocol stacks that have not had jail functional‐
ity added to them.
Jails are typically set up using one of two philosophies: either to con‐
strain a specific application (possibly running with privilege), or to
create a “virtual system image” running a variety of daemons and ser‐
vices. In both cases, a fairly complete file system install of FreeBSD
is required, so as to provide the necessary command line tools, daemons,
libraries, application configuration files, etc. However, for a virtual
server configuration, a fair amount of additional work is required so as
to configure the “boot” process. This manual page documents the configu‐
ration steps necessary to support either of these steps, although the
configuration steps may be refined based on local requirements.
EXAMPLES
Setting up a Jail Directory Tree
To set up a jail directory tree containing an entire FreeBSD distribu‐
tion, the following sh(1) command script can be used:
D=/here/is/the/jail
cd /usr/src
mkdir -p $D
make world DESTDIR=$D
make distribution DESTDIR=$D
mount -t devfs devfs $D/dev
NOTE: It is important that only appropriate device nodes in devfs be
exposed to a jail; access to disk devices in the jail may permit pro‐
cesses in the jail to bypass the jail sandboxing by modifying files out‐
side of the jail. See devfs(8) for information on how to use devfs rules
to limit access to entries in the per-jail devfs. A simple devfs ruleset
for jails is available as ruleset #4 in /etc/defaults/devfs.rules.
In many cases this example would put far more in the jail than needed.
In the other extreme case a jail might contain only one file: the exe‐
cutable to be run in the jail.
We recommend experimentation and caution that it is a lot easier to start
with a “fat” jail and remove things until it stops working, than it is to
start with a “thin” jail and add things until it works.
Setting Up a Jail
Do what was described in Setting Up a Jail Directory Tree to build the
jail directory tree. For the sake of this example, we will assume you
built it in /data/jail/192.0.2.100, named for the jailed IP address.
Substitute below as needed with your own directory, IP address, and host‐
name.
Setting up the Host Environment
First, you will want to set up your real system's environment to be
“jail-friendly”. For consistency, we will refer to the parent box as the
“host environment”, and to the jailed virtual machine as the “jail
environment”. Since jail is implemented using IP aliases, one of the
first things to do is to disable IP services on the host system that lis‐
ten on all local IP addresses for a service. If a network service is
present in the host environment that binds all available IP addresses
rather than specific IP addresses, it may service requests sent to jail
IP addresses if the jail did not bind the port. This means changing
inetd(8) to only listen on the appropriate IP address, and so forth. Add
the following to /etc/rc.conf in the host environment:
sendmail_enable="NO"
inetd_flags="-wW -a 192.0.2.23"
rpcbind_enable="NO"
192.0.2.23 is the native IP address for the host system, in this example.
Daemons that run out of inetd(8) can be easily set to use only the speci‐
fied host IP address. Other daemons will need to be manually configured—
for some this is possible through the rc.conf(5) flags entries; for oth‐
ers it is necessary to modify per-application configuration files, or to
recompile the applications. The following frequently deployed services
must have their individual configuration files modified to limit the
application to listening to a specific IP address:
To configure sshd(8), it is necessary to modify /etc/ssh/sshd_config.
To configure sendmail(8), it is necessary to modify
/etc/mail/sendmail.cf.
For named(8), it is necessary to modify /etc/namedb/named.conf.
In addition, a number of services must be recompiled in order to run them
in the host environment. This includes most applications providing ser‐
vices using rpc(3), such as rpcbind(8), nfsd(8), and mountd(8). In gen‐
eral, applications for which it is not possible to specify which IP
address to bind should not be run in the host environment unless they
should also service requests sent to jail IP addresses. Attempting to
serve NFS from the host environment may also cause confusion, and cannot
be easily reconfigured to use only specific IPs, as some NFS services are
hosted directly from the kernel. Any third-party network software run‐
ning in the host environment should also be checked and configured so
that it does not bind all IP addresses, which would result in those ser‐
vices' also appearing to be offered by the jail environments.
Once these daemons have been disabled or fixed in the host environment,
it is best to reboot so that all daemons are in a known state, to reduce
the potential for confusion later (such as finding that when you send
mail to a jail, and its sendmail is down, the mail is delivered to the
host, etc.).
Configuring the Jail
Start any jail for the first time without configuring the network inter‐
face so that you can clean it up a little and set up accounts. As with
any machine (virtual or not) you will need to set a root password, time
zone, etc. Some of these steps apply only if you intend to run a full
virtual server inside the jail; others apply both for constraining a par‐
ticular application or for running a virtual server.
Start a shell in the jail:
jail-c path=/data/jail/192.0.2.100 host.hostname=testhostname \
ip4.addr=192.0.2.100 command=/bin/sh
Assuming no errors, you will end up with a shell prompt within the jail.
You can now run /usr/sbin/sysinstall and do the post-install configura‐
tion to set various configuration options, or perform these actions manu‐
ally by editing /etc/rc.conf, etc.
· Create an empty /etc/fstab to quell startup warnings about
missing fstab (virtual server only)
· Disable the port mapper (/etc/rc.conf: rpcbind_enable="NO")
(virtual server only)
· Configure /etc/resolv.conf so that name resolution within the
jail will work correctly
· Run newaliases(1) to quell sendmail(8) warnings.
· Disable interface configuration to quell startup warnings about
ifconfig(8) (network_interfaces="") (virtual server only)
· Set a root password, probably different from the real host sys‐
tem
· Set the timezone
· Add accounts for users in the jail environment
· Install any packages the environment requires
You may also want to perform any package-specific configuration (web
servers, SSH servers, etc), patch up /etc/syslog.conf so it logs as you
would like, etc. If you are not using a virtual server, you may wish to
modify syslogd(8) in the host environment to listen on the syslog socket
in the jail environment; in this example, the syslog socket would be
stored in /data/jail/192.0.2.100/var/run/log.
Exit from the shell, and the jail will be shut down.
Starting the Jail
You are now ready to restart the jail and bring up the environment with
all of its daemons and other programs. If you are running a single
application in the jail, substitute the command used to start the appli‐
cation for /etc/rc in the examples below. To start a virtual server
environment, /etc/rc is run to launch various daemons and services. To
do this, first bring up the virtual host interface, and then start the
jail's /etc/rc script from within the jail.
ifconfig ed0 inet alias 192.0.2.100/32
mount -t procfs proc /data/jail/192.0.2.100/proc
jail-c path=/data/jail/192.0.2.100 host.hostname=testhostname \
ip4.addr=192.0.2.100 command=/bin/sh /etc/rc
A few warnings will be produced, because most sysctl(8) configuration
variables cannot be set from within the jail, as they are global across
all jails and the host environment. However, it should all work prop‐
erly. You should be able to see inetd(8), syslogd(8), and other pro‐
cesses running within the jail using ps(1), with the ‘J’ flag appearing
beside jailed processes. To see an active list of jails, use the jls(8)
utility. You should also be able to telnet(1) to the hostname or IP
address of the jailed environment, and log in using the accounts you cre‐
ated previously.
It is possible to have jails started at boot time. Please refer to the
“jail_*” variables in rc.conf(5) for more information. The rc(8)jail
script provides a flexible system to start/stop jails:
/etc/rc.d/jail start
/etc/rc.d/jail stop
/etc/rc.d/jail start myjail
/etc/rc.d/jail stop myjail
Managing the Jail
Normal machine shutdown commands, such as halt(8), reboot(8), and
shutdown(8), cannot be used successfully within the jail. To kill all
processes in a jail, you may log into the jail and, as root, use one of
the following commands, depending on what you want to accomplish:
kill -TERM -1
kill -KILL -1
This will send the SIGTERM or SIGKILL signals to all processes in the
jail from within the jail. Depending on the intended use of the jail,
you may also want to run /etc/rc.shutdown from within the jail. To kill
processes from outside the jail, use the jexec(8) utility in conjunction
with the one of the kill(1) commands above. You may also remove the jail
with jail-r, which will killall the jail's processes with SIGKILL.
The /proc/pid/status file contains, as its last field, the name of the
jail in which the process runs, or “-” to indicate that the process is
not running within a jail. The ps(1) command also shows a ‘J’ flag for
processes in a jail.
You can also list/kill processes based on their jail ID. To show pro‐
cesses and their jail ID, use the following command:
ps ax -o pid,jid,args
To show and then kill processes in jail number 3 use the following com‐
mands:
pgrep -lfj 3
pkill -j 3
or:
killall -j 3
Jails and File Systems
It is not possible to mount(8) or umount(8) any file system inside a jail
unless the file system is marked jail-friendly and the jail's allow.mount
parameter is set.
Multiple jails sharing the same file system can influence each other.
For example a user in one jail can fill the file system also leaving no
space for processes in the other jail. Trying to use quota(1) to prevent
this will not work either as the file system quotas are not aware of
jails but only look at the user and group IDs. This means the same user
ID in two jails share the same file system quota. One would need to use
one file system per jail to make this work.
Sysctl MIB Entries
The read-only entry security.jail.jailed can be used to determine if a
process is running inside a jail (value is one) or not (value is zero).
The variable security.jail.max_af_ips determines how may address per
address family a prison may have. The default is 255.
Some MIB variables have per-jail settings. Changes to these variables by
a jailed process do not effect the host environment, only the jail envi‐
ronment. These variables are kern.securelevel, kern.hostname,
kern.domainname, kern.hostid, and kern.hostuuid.
Hierarchical Jails
By setting a jail's children.max parameter, processes within a jail may
be able to create jails of their own. These child jails are kept in a
hierarchy, with jails only able to see and/or modify the jails they cre‐
ated (or those jails' children). Each jail has a read-only parent param‐
eter, containing the jid of the jail that created it; a jid of 0 indi‐
cates the jail is a child of the current jail (or is a top-level jail if
the current process isn't jailed).
Jailed processes are not allowed to confer greater permissions than they
themselves are given, e.g. if a jail is created with allow.nomount, it is
not able to create a jail with allow.mount set. Similarly, such restric‐
tions as ip4.addr and securelevel may not be bypassed in child jails.
A child jail may in turn create its own child jails if its own
children.max parameter is set (remember it is zero by default). These
jails are visible to and can be modified by their parent and all ances‐
tors.
Jail names reflect this hierarchy, with a full name being an MIB-type
string separated by dots. For example, if a base system process creates
a jail “foo”, and a process under that jail creates another jail “bar”,
then the second jail will be seen as “foo.bar” in the base system (though
it is only seen as “bar” to any processes inside jail “foo”). Jids on
the other hand exist in a single space, and each jail must have a unique
jid.
Like the names, a child jail's path is relative to its creator's own
path. This is by virtue of the child jail being created in the chrooted
environment of the first jail.
SEE ALSOkillall(1), lsvfs(1), newaliases(1), pgrep(1), pkill(1), ps(1), quota(1),
chroot(2), jail_set(2), jail_attach(2), procfs(5), rc.conf(5),
sysctl.conf(5), devfs(8), halt(8), inetd(8), jexec(8), jls(8), mount(8),
named(8), reboot(8), rpcbind(8), sendmail(8), shutdown(8), sysctl(8),
syslogd(8), umount(8)HISTORY
The jail utility appeared in FreeBSD 4.0. Hierarchical/extensible jails
were introduced in FreeBSD 8.0.
AUTHORS
The jail feature was written by Poul-Henning Kamp for R&D Associates
http://www.rndassociates.com/ who contributed it to FreeBSD.
Robert Watson wrote the extended documentation, found a few bugs, added a
few new features, and cleaned up the userland jail environment.
Bjoern A. Zeeb added multi-IP jail support for IPv4 and IPv6 based on a
patch originally done by Pawel Jakub Dawidek for IPv4.
James Gritton added the extensible jail parameters and hierchical jails.
BUGS
Jail currently lacks the ability to allow access to specific jail infor‐
mation via ps(1) as opposed to procfs(5). Similarly, it might be a good
idea to add an address alias flag such that daemons listening on all IPs
(INADDR_ANY) will not bind on that address, which would facilitate build‐
ing a safe host environment such that host daemons do not impose on ser‐
vices offered from within jails. Currently, the simplest answer is to
minimize services offered on the host, possibly limiting it to services
offered from inetd(8) which is easily configurable.
BSD January 17, 2010 BSD