|
Unix VPS
Support
FAQ
Network
Miscellaneous
|
| iocell(8) | FreeBSD System Manager's Manual | iocell(8) |
NAME
iocell - jail manager amalgamating ZFS, VNET and resource limits
SYNOPSIS
iocell [-v] activate ZPOOL iocell [-v] cap UUID|TAG iocell [-v] clean [-f] [-a|-r|-j|-t] iocell [-v] clone UUID|TAG [UUID|TAG@snapshot] [property=value] iocell [-v] console [-f] UUID|TAG iocell [-v] create [-e] [release=[RELEASE|NAME]] [pkglist=file] [property=value] [count=value] iocell [-v] deactivate ZPOOL iocell [-v] defaults iocell [-v] destroy [-f] UUID|TAG iocell [-v] df iocell [-v] exec [-u username | -U username] UUID|TAG|ALL command [arg ...] iocell [-v] export UUID|TAG iocell [-v] fetch [-p|-P] [release=RELEASE | ftphost=ftp.hostname.org |
ftpdir=/dir/ | ftpfiles="base.txz lib32.txz src.txz"]
[ ftplocaldir=/dir/ ] [ compression=ALGO ] iocell [-v] get [-r] property|all UUID|TAG iocell [-v] help iocell [-v] import UUID [property=value] iocell [-v] init-host IP ZPOOL iocell [-v] inuse UUID|TAG iocell [-v] pkg TAG pkg_cmd iocell [-v] limits [UUID|TAG] iocell [-v] list [-t|-r|-s jid|uuid|boot|state|tag|type|ip4] iocell [-v] promote UUID|TAG iocell [-v] rcboot iocell [-v] reboot|restart [-s] UUID|TAG iocell [-v] rcshutdown iocell [-v] reset UUID|TAG|ALL iocell [-v] rollback UUID|TAG@snapshotname iocell [-v] runtime UUID|TAG iocell [-v] send [-c|-i|-I|-h|-u|-m] POOL iocell [-v] set property=value [property=value] UUID|TAG iocell [-v] snaplist UUID|TAG iocell [-v] snapmount UUID|TAG@snapshotname DESTINATION iocell [-v] snapremove UUID|TAG@snapshotname|ALL iocell [-v] snapshot|snap [-f|-r] UUID|TAG [UUID|TAG@snapshotname] iocell [-v] snapumount UUID|TAG@snapshotname iocell [-v] start [-f] UUID|TAG iocell [-v] stop UUID|TAG|ALL iocell [-v] uncap UUID|TAG iocell [-v] update [-p|-P] UUID|TAG|RELEASE iocell [-v] upgrade UUID|TAG [release=RELEASE] iocell [-v] version | --version
DESCRIPTION
iocell is a system administration tool for jails designed to simplify jail management tasks. It abstracts away the management of ZFS backed jails running VNET or shared IP networking with optional support for resource limits.
Both, shared IP based jails and VNET enabled jails are supported.
Each jail has a unique ID (UUID) automatically generated at creation time. Using the UUID as a jail identifier means that a jail can be replicated in a distributed environment with greater flexibility. This also eliminates potential naming clashes on large scale deployments and helps reduce operator error.
You can use either the full UUID or a jail's TAG with every operation.
To ease jail identification a TAG field is included in list mode which can be set to any string (hostname, label, note, etc.). By default if unset the TAG field contains the creation date and time stamp.
Properties are stored inside ZFS custom fields. This eliminates the need for any configuration files and jails can be easily moved with ZFS send and receive preserving all of their properties automatically.
iocell relies on ZFS and at least one ZFS pool must be present on the host system. To enable all the features iocell supports, consider the following optional kernel options and system requirements:
Calling iocell with the -v switch before a command will allow you to see the how the shell is running each of the commands. It's very verbose.SUBCOMMANDS
activate ZPOOL
cap UUID|TAG
Intended for automation tools. The pool can be activated for iocell jails
without requiring user input.
clean [-f] [-a|-r|-j|-t]
Reapply resource limits on jail while it is running.
clone UUID|TAG [UUID|TAG@snapshot] [property=value]
This removes the ZFS datasets associated with the flag called. The -a flag
will destroy everything associated with iocell. The -r flag will destroy
all RELEASEs that have been fetched. Basejails rely on these to exist,
so be sure you don't use this flag if you have any. The -j flag will
destroy the jails dataset. The -t flag will destroy the templates dataset.
Same warning applies here as with cleaning the RELEASE dataset.
console [-f] UUID|TAG
Clone jail identified by UUID (ZFS clone). All properties will be reset on
the clone, defaults can be overridden by specifying properties on the fly.
Custom point-in-time snapshots can be used as a source for cloning in the
form of UUID@snapshot or TAG@snapshot.
Examples:
Clone the current state of the jail:
iocell clone UUID tag=www02
Clone a jail from a custom snapshot (snapshot taken previously):
iocell clone UUID@snapshotname tag=www02
create [-e] [base=[RELEASE|NAME]] [property=value] [property=value] [count=value]
Execute login to have a shell inside the jail. If the -f flag is specified,
and the jail is not running it will be started for the console command.
deactivate ZPOOL
By default create will deploy a new jail based on the host operating
system's release. This can be changed by specifying the base option.
By default a basejail is created that has a common shared base.
The -e switch will create an empty jail which can be used for unsupported
or custom jails.
Example: iocell create tag=www01 pkglist=$HOME/my-pkgs.txt
iocell create tag=mybasejail
iocell create tag=mybasejail base=10.2-RELEASE
iocell create tag=mycustombasejail base=nginx_postgres
By default only one jail will be created, but if you specify count=NUMBER,
then that many jails will be created.
Example: iocell create tag="mybasejail" host_hostname="mybasejail" count=10
This will create 10 jails with the tag mybasejail_{1-10}. The tag and
hostname will be incremented up with the number of jails you wish to create.
defaults
Remove the comment that iocell uses to locate a usable pool.
destroy [-f] UUID|TAG
Display all defaults set in the default dataset.
df
Destroy given jail. This is irreversible, use with caution. If the jail is
running destroy action will fail.
With -f the jail will be destroyed without confirmation.
exec [-u username | -U username] UUID|TAG|ALL command [arg ...]
List disk space related information. Available fields:
CRT - compression ratio
RES - reserved space
QTA - disk quota
USE - used space
AVA - available space
TAG - jail name
export UUID|TAG
Execute command inside the jail. This is simply an iocell UUID/tag wrapper
for jexec(8).
fetch [-p|-P] [release=RELEASE | ftphost=ftp.hostname.org | ftpdir=/dir/ | ftpfiles="base.txz lib32.txz src.txz"] [ ftplocaldir=/dir/ ] [ compression=ALGO ]
Export a complete jail. An archive file is created in /iocell/images with
SHA256 checksum. Jail must be in stopped state before exporting.
Used for downloading and updating/patching releases.
fetch must be executed as the first command on a pristine system. By default fetch will download the host node's RELEASE for deployment. If other releases are required, this can be changed by supplying the required release property or just selecting the appropriate RELEASE from the menu list.
Example: iocell fetch release=13.2-RELEASE
fetch can also used to update already downloaded releases. To update a local release already present in iocell (iocell list -r) run:
This example will apply latest patches to 13.2-RELEASE base. Newly created jails or basejails will automatically have the latest updates applied.
iocell fetch release=13.2-RELEASE
Specifying -p or -P to fetch will have it also fetch a ports tree that all basejails will share with that base.
compression: Compression is lz4 by default; see zfs(8) for the list of compression algorithms available.
ftplocaldir: Specifying ftplocaldir=/dev/ will fetch release tar files from a local system directory. This is useful when a user wants to use a custom build release or use iocell offline.
Example: fetch release=13.2-RELEASE ftplocaldir=/fakeftp/release/13.2-RELEASE
get [-r] property|all UUID|TAG
help
Get named property or if "all" keyword is specified dump all properties
known to iocell.
To display whether resource limits are enforced for a jail:
iocell get rlimits UUID|TAG
iocell get -r rlimits
iocell get all UUID|TAG
import UUID [property=value]
This help file you are reading.
init-host IP ZPOOL
Import full jail images or differential packages. Images need to be
present in /iocell/images along with checksum files. You can use short
UUIDs - do not specify the full filename only the UUID. These jails
become what we call "thickjails" and are not supported for all operations.
inuse [UUID|TAG]
Initialize a remote host for iocell. iocell will send all required ZFS
filesystems over the network to the host. This requires SSH working keys
installed for root.
pkg TAG pkg_cmd
Display consumed resources for a jail. Without UUID or TAG dump all
resources for all running jails in a comma delimited form.
limits [UUID|TAG]
A convinience wrapper for pkg to manage your jails. Converts the given tag
to the JID of a running jail and executes `pkg -j JID pkg_cmd`.
list [-t|-r]
Display active resource limits for a jail or all jails. With no UUID
supplied display all limits active for all jails.
promote UUID|TAG
List all jails, if -t is specified list only templates,
with -r list downloaded releases.
Non iocell jails will be listed only if the jail is running.
rcboot
Promote a cloned jail to a fully independent copy. For more details please
read zfs(8).
reboot UUID|TAG
Start all jails with "boot" property set to "on". Intended for boot time
execution. Jails will be started in an ordered fashion based on their
"priority" property.
rcshutdown
Fully stops the jail with `iocell stop` and, upon success, restarts it with
`iocell start`. If `-s` is specified it will instead do a soft restart
on the jail. The soft method will restart the jail without destroying
the jail's networking or the jail process itself. All processes are
gracefully restarted inside the jail. This is useful for quick and graceful
restarts.
reset UUID|TAG|ALL
Stop all jails with "boot" property set to "on". Intended for full host
shutdown.
Jails will be stopped in an ordered fashion based on their "priority"
property.
restart UUID|TAG
This will reset a jail's properties back to the defaults.
It reads from the properties set on the "default" dataset. TAG, UUID and
generated vnet mac addresses are carried forward.
Those will retain their values, even if you reset the jail.
You can also reset every jail to the default properties by using the
keyword "ALL".
rollback UUID|TAG@snapshotname
Soft restart jail. Soft method will restart the jail without destroying
the jail's networking and the jail process itself. All processes are
gracefully restarted inside the jail. Useful for quick and graceful
restarts.
runtime UUID|TAG
Rollback to an existing snapshot. Any intermediate snapshots will be
destroyed. For more information on this functionality please read zfs(8).
send [-c|-i|-I|-h|-u|-m] POOL
Show runtime configuration of a jail. Useful for debugging.
set property=value [property=value] UUID|TAG
Will take a recursive snapshot of the POOL/iocell dataset and send it to
the specified pool. Supports remote, incremental and movement operations.
Moving (-m) is only available for local operations. Remote (-r) requires a
user (-u) and a host (-h) to be specified. It also requires publickey
authentication. If using incremental (-i) or Incremental (-I) be sure to
supply the flag for each snapshot.
If clean (-c) is provided, it will remove all snapshots from both the
source and destination pools. This cannot be used in combination with
incremental (-i) or Incremental (-i).
Local Example: iocell send newpool
Incremental Example: iocell send -i oldpool/iocell@snap1 -i oldpool/iocell@snap2 newpool
Remote Example: iocell send -r -u root -h some.host newpool
The received pool is not mounted if received locally, but is mounted if
received remotely.
snaplist UUID|TAG
Set one or more properties to the supplied value.
snapmount UUID|TAG@snapshotname DESTINATION
List all snapshots belonging to jail.
NAME - snapshot name
CREATED - creation time
RSIZE - referenced size
USED - used space
snapremove UUID|TAG@snapshotname|ALL
This will mount the specified jail's snapshot on the location given.
snapshot|snap [-f|-r] UUID|TAG [UUID|TAG@snapshotname]
Destroy specified jail snapshot. If the keyword ALL is specified all
snapshots will be destroyed for the jail.
snapumount UUID|TAG@snapshotname
Create a ZFS snapshot for jail. If the -r flag is specified it will by
recursively snapshot POOL/iocell/jails/UUID. Otherwise it will only snapshot
POOL/iocell/jails/UUID/root. If no snapshot name is specified defaults
to auto snapshot name based on current date and time.
The -f flag is to bypass the interactive question if you have a running
database in the jail. Make sure to pass -f on it's own and first before any
other flag like -r.
Example: iocell snapshot -f -r JAIL
start [-f] UUID|TAG
Will unmount the specified jail's snapshot.
stop UUID|TAG|ALL
Start jail identified by UUID or TAG. If the -f switch is supplied, it will
force start a template for editing. This flag does not apply to normal
jails.
uncap UUID|TAG
Stop jail identified by UUID or TAG. If ALL is specified, it will stop all
jails.
update [-p|-P] UUID|TAG|RELEASE
Release all resource limits, disable limits on the fly.
upgrade UUID|TAG [release=RELEASE]
Will reclone the jail if it is a basejail, otherwise it updates the jail
to the latest patch level. A back-out snapshot is created to allow
safe update/rollback.
If -p or -P along with a RELEASE are specified instead of a jail, it will
update that RELEASEs shared ports tree.
version | --version
By default this will upgrade jail RELEASE to match the host's RELEASE
unless another RELEASE is specified with the "release" property.
Please note: The upgrade process for basejails is done by setting the jail's
"release" property to the required new RELEASE and then merging /etc.
If you wish to do this yourself without having /etc merged do:
iocell set release=10.1-RELEASE UUID|TAG
For this the RELEASE must be locally available. The basejail
will re-clone its filesystems from the new release at next jail start.
WARNING: Cloned jails (deprecated) [-c] are not supported for upgrades.
Please use thick (deprecated) [-t] or our standard basejails for
this feature.
List version number.
PROPERTIES
For more information on properties please check the relevant man page which is noted under each property in the form of "Source: manpage". Source "local" marks iocell specific properties.
pkglist=none | path-to-file
vnet=on | off
A text file containing one package per line. These will be auto installed
when a jail is created. Works only in combination with the create
subcommand.
Default: none
Source: local
bpf=on | off
This controls whether to start the jail with VNET or a shared IP
configuration. Default is to auto-guess from a sysctl. If you don't
need a fully virtualized per jail network stack set it to off.
Default: auto-guess
Source: local
dhcp=on | off
This controls whether to start the jail with BPF devices enabled.
Default is off.
Default: off
Source: local
ip4_addr="interface|ip-address/netmask"
This controls whether to start the jail with DHCP.
Default is off, but if you want it on, vnet and bpf are required.
Default: off
Source: local
ip4_saddrsel=1 | 0
The IPv4 address for VNET and shared IP jails.
Form is: interface|ip-address/netmask
Multiple interfaces:
"interface|ip-address/netmask,interface|ip-address/netmask"
For shared IP jails if an interface is given before
the IP address, an alias for the address will be added to that
interface, as it is with the interface parameter. If the DEFAULT
tag is used instead of an interface, the interface will be
automatically assigned based upon the system's default interface.
If a netmask in either dotted-quad or CIDR form is given after IP
address, it will be used when adding the IP alias.
If the AUTOIP4 tag is used in place of an ip-address, the IP address
will be automatically assigned at first start of the jail. This
requires that the ip4_autostart and ip4_autoend variables are set on
the 'default' property source. If used, the IP4 address will be set
to the first available based upon the given range and already created
jails.
Example:
# iocell set ip4_autostart="100" default
# iocell set ip4_autoend="150" default
# iocell set ip4_autosubnet="24" default
This will result in the automatic IPv4 address being assigned in the
base range of the default network interface. I.E. if the local default
NIC is set to 192.168.0.XXX, then the new address will be
192.168.0.[100-150]/24
For VNET jails the interface will be configured with the IP addresses
listed.
Example: "vnet0|192.168.0.10/24,vnet1|10.1.1.10/24"
This would configure interfaces vnet0 and vnet1 in a VNET jail. In this
case no network configuration is necessary in the jail's rc.conf file.
Default: none
Source: jail(8)
ip4=new | disable | inherit
Only takes effect when vnet=off.
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 the ip4_nosaddrsel
settting of a parent jail is not inherited for any child jails.
Default: 1
Source: jail(8)
createbridge=on | off
Only takes effect when vnet=off.
Control the availability 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".
Default: new
Source: jail(8)
keepbridge=on | off
Controls whether default bridge0 and bridge1 will be created
on jails's start.
Default: on
defaultrouter=none | ipaddress
Controls whether bridges bridge0 and bridge1 will be destroyed
on jail's shutdown.
Default: off
defaultrouter6=none | ip6address
Setting this property to anything other than none will try to configure a
default route inside a VNET jail.
resolver=none | nameserver IP;nameserver IP;search domain.local
Setting this property to anything other than none will try to configure a
default IPv6 route inside a VNET jail.
ip6.addr, ip6.saddrsel, ip6
Sets the jail's resolver (resolv.conf). Fields must be ; delimited
which are translated to newlines in resolv.conf.
If the resolver is set to none (default) the jail will inherit
the hosts resolv.conf file.
interfaces=vnet0:bridge0,vnet1:bridge1 | vnet0:bridge0
A set of IPv6 options for the prison, the counterparts to
ip4.addr, ip4.saddrsel and ip4 above.
host_domainname="NIS domainname"
By default there are two interfaces specified with their bridge
association. Up to ten interfaces are supported. Interface configurations
are separated by commas. Format is interface:bridge, where left value is
the virtual VNET interface name, right value is the bridge name where the
virtual interface should be attached.
Default: vnet0:bridge0,vnet1:bridge1
Source: local
host_hostname=UUID
The NIS domainname of the jail.
Default: none
Source: jail(8)
exec_fib=0 | 1 ..
The hostname of the jail.
Default: UUID
Source: jail(8)
devfs_ruleset=4 | 0 ..
The FIB (routing table) to set when running commands inside the jail.
Default: 0
Source: jail(8)
mount_devfs=1 | 0
The number of the devfs ruleset that is enforced for mounting
devfs in this jail. A value of zero (default) means no ruleset
is enforced. Descendant jails inherit the parent jail's devfs
ruleset enforcement. Mounting devfs inside a jail is possible
only if the allow_mount and allow_mount_devfs permissions are
effective and enforce_statfs is set to a value lower than 2.
Devfs rules and rulesets cannot be viewed or modified from inside
a jail.
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
processes in the jail to bypass the jail sandboxing by modifying
files outside 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
Default: 4
Source: jail(8)
exec_start="/bin/sh /etc/rc"
Mount a devfs(5) filesystem on the chrooted /dev directory, and
apply the ruleset in the devfs_ruleset parameter (or a default of
ruleset 4: devfsrules_jail) to restrict the devices visible
inside the jail.
Default: 1
Source: jail(8)
exec_stop="/bin/sh /etc/rc.shutdown"
Command(s) to run in the prison environment when a jail is created.
A typical command to run is "sh /etc/rc".
Default: /bin/sh /etc/rc
Source: jail(8)
exec_prestart="/usr/bin/true"
Command(s) to run in the prison environment before a jail is
removed, and after any exec_prestop commands have completed.
A typical command to run is "sh /etc/rc.shutdown".
Default: /bin/sh /etc/rc.shutdown
Source: jail(8)
exec_prestop="/usr/bin/true"
Command(s) to run in the system environment before a jail is started.
Default: /usr/bin/true
Source: jail(8)
exec_poststop="/usr/bin/true"
Command(s) to run in the system environment before a jail is stopped.
Default: /usr/bin/true
Source: jail(8)
exec_poststart="/usr/bin/true"
Command(s) to run in the system environment after a jail is stopped.
Default: /usr/bin/true
Source: jail(8)
exec_clean=1 | 0
Command(s) to run in the system environment after a jail is started,
and after any exec_start commands have completed.
Default: /usr/bin/true
Source: jail(8)
exec_timeout=60 | 30 ..
Run commands in a clean environment. The environment is discarded
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 database
for the target login are also set.
Default: 1
Source: jail(8)
stop_timeout=30 | 60 ..
The maximum amount of time to wait for a command to complete. If
a command is still running after this many seconds have passed,
the jail will be terminated.
Default: 60
Source: jail(8)
exec_jail_user=root
The maximum amount of time to wait for a jail's processes to
exit after sending them a SIGTERM signal (which happens after the
exec_stop commands have completed). After this many seconds have
passed, the jail will be removed, which will kill any remaining
processes. If this is set to zero, no SIGTERM is sent and the
prison is immediately removed.
Default: 30
Source: jail(8)
exec_system_jail_user=0 | 1
The user to run commands as, when running in the jail environment.
Default: root
Source: jail(8)
exec_system_user=root
This boolean option looks for the exec_jail_user in the system
passwd(5) file, instead of in the jail's file.
Default: 0
Source: jail(8)
mount_fdescfs=1 | 0
The user to run commands as, when running in the system environment.
The default is to run the commands as the current user.
Default: root
Source: jail(8)
mount_procfs=0 | 1
Mount a fdescfs(5) filesystem in the jail's /dev/fd directory.
Note: This is not supported on FreeBSD 9.3.
Default: 1
Source: jail(8)
mount_linprocfs=0 | 1
Mount a procfs(5) filesystem in the jail's /dev/proc directory.
Default: 0
Source: local
enforce_statfs=2 | 1 | 0
Mount a linprocfs(5) filesystem in the jail's /compat/linux/proc directory.
Default: 0
Source: local
children_max=0 | ..
This determines which information processes in a jail are able to
get about mount points. It affects the behaviour of the following
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 visible
In addition to that, the path to the jail's chroot directory
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.
Default: 2. jail(8)
login_flags="-f root"
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 in jail(8).
Default: 0
Source: jail(8)
jail_zfs=on | off
Supply these flags to login when logging in to jails with the console
function.
Default: -f root
Source: login(1)
jail_zfs_dataset="zfs_filesystem ..."
Enables automatic ZFS jailing inside the jail. Assigned ZFS dataset will
be fully controlled by the jail.
NOTE: Setting this to "on" automatically enables allow_mount=1
enforce_statfs=1 and allow_mount_zfs=1! These are dependent options
required for ZFS management inside a jail.
Default: off
Source: local
jail_zfs_mountpoint=/path/in/jail
This is a space-separated list of datasets to be jailed and fully handed
over to a jail.
Takes a ZFS filesystem name; if it begins with "iocell/jails", it will be
created on the ZFS pool that iocell uses.
NOTE: only valid if jail_zfs=on. By default the mountpoint is set to none,
to mount this dataset set its mountpoint inside the jail i.e. "zfs set
mountpoint=/data full-dataset-name" and issue "mount -a".
Default: iocell/jails/UUID/root/data
Source: local
securelevel=3 | 2 | 1 | 0 | -1
If only a single ZFS filesystem is assigned to a jail via the
jail_zfs_dataset setting, this property may be used to manipulate its
mountpoint within the jail.
The default value of "none" allows the mountpoint to be managed
from within the jail by setting the ZFS "mountpoint" property on the
dataset directly.
When multiple ZFS filesystesm are assigned to a jail, this property
must be "none", and the filesystem mountpoints are manipulated directly
using the ZFS "mountpoint" properties for the assigned datasets.
Default: none
Source: local
allow_set_hostname=1 | 0
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.
Default: 2
Source: jail(8)
allow_sysvipc=0 | 1
The jail's hostname may be changed via hostname(1) or sethostname(3).
Default: 1
Source: jail(8)
sysvmsg=disable | inherit | new
A process within the jail has access to System V IPC
primitives. Prior to FreeBSD 11.0, 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 interfere with)
processes outside of the jail, and in other jails. In
FreeBSD 11.0 and later, this setting is deprecated in favor
of sysvmsg, sysvsem, and sysvshm.
Default: 0
Source: jail(8)
sysvsem=disable | inherit | new
Allow access to SYSV IPC message primitives. If set to
inherit, all IPC objects on the system are visible to this
jail, whether they were created by the jail itself, the base
system, or other jails. If set to new, the jail will have
its own key namespace, and can only see the objects that it
has created; the system (or parent jail) has access to the
jail's objects, but not to its keys. If set to disable, the
jail cannot perform any sysvmsg-related system calls.
Ignored in FreeBSD 10.3 and earlier.
Default: new
Source: jail(8)
sysvshm=disable | inherit | new
Allow access to SYSV IPC semaphore primitives in the same
manner as sysvmsg. Ignored in FreeBSD 10.3 and earlier.
Default: new
Source: jail(8)
allow_raw_sockets=0 | 1
Allow access to SYSV IPC shared memory primitives in the
same manner as sysvmsg. Ignored in FreeBSD 10.3 and
earlier.
Default: new
Source: jail(8)
allow_chflags=0 | 1
The prison root is allowed to create raw sockets. Setting
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.
Default: 0
Source: jail(8)
allow_mount=0 | 1
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.
Default: 0
Source: jail(8)
allow_mount_devfs=0 | 1
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. This permission
is effective only if enforce_statfs is set to a
value lower than 2.
Default: 0
Source: jail(8)
allow_mount_nullfs=0 | 1
privileged users inside the jail will be able to mount
and unmount the devfs file system. This permission is
effective only together with allow.mount and if
enforce_statfs is set to a value lower than 2. Please
consider restricting the devfs ruleset with the
devfs_ruleset option.
Default: 0
Source: jail(8)
allow_mount_procfs=0 | 1
privileged users inside the jail will be able to mount
and unmount the nullfs file system. This permission is
effective only together with allow_mount and if
enforce_statfs is set to a value lower than 2.
Default: 0
Source: jail(8)
allow_mount_tmpfs=0 | 1
privileged users inside the jail will be able to mount
and unmount the procfs file system. This permission is
effective only together with allow.mount and if
enforce_statfs is set to a value lower than 2.
Default: 0
Source: jail(8)
allow_mount_zfs=0 | 1
privileged users inside the jail will be able to mount
and unmount the tmpfs file system. This permission is
effective only together with allow.mount and if
enforce_statfs is set to a value lower than 2.
Note: This is not supported on FreeBSD 9.3.
Default: 0
Source: jail(8)
allow_quotas=0 | 1
privileged users inside the jail will be able to mount
and unmount the ZFS file system. This permission is
effective only together with allow.mount and if
enforce_statfs is set to a value lower than 2. See
zfs(8) for information on how to configure the ZFS
filesystem to operate from within a jail.
Default: 0
Source: jail(8)
allow_socket_af=0 | 1
The jail 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.
Default: 0
Source: jail(8)
host_hostuuid=UUID
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 functionality
added to them.
Default: 0
Source: jail(8)
tag="any string"
Default: UUID
Source: jail(8)
istemplate=yes | no
Custom string for aliasing jails. Must be unique.
Default: date@time
Source: local
boot=on | off
This property controls whether the jail is a template. Templates are not
started by iocell. Set to yes if you wish to convert the jail into
a template.
(See EXAMPLES section)
Default: no
Source: local
notes="any string"
If set to "on" jail will be auto-started at boot time (rcboot subcommand)
and stopped at shutdown time (rcshutdown subcommand). Jails will be started
and stopped based on their priority value.
Default: off
Source: local
owner=root
Custom notes for miscellaneous tagging.
Default: none
Source: local
priority=99:0 Start priority at boot time, smaller value means higher priority. Also, for shutdown the order will be reversed.
The owner of the jail, can be any string.
Default: root
Source: local
Default: 99. Max: 99. Lowest: 0.
Anything set to 99 will automatically start in the background parallel with other jails set to 99. Anything less than 99 will start sequentially.
last_started
type=basejail
Last successful start time. Auto set every time jail starts.
Default: timestamp
Source: local
hostid=UUID
Currently only basejails, gitjails and empty jails are supported.
Default: basejail
Source: local
release=14.0-RELEASE | 13.2-RELEASE
The UUID of the host node. Jails won't start if this property differs from
the actual UUID of the host node. This is to safeguard jails from being
started on different nodes in case they are periodically replicated across.
Default: UUID of the host (taken from /etc/hostid)
Source: local
compression=on | off | lzjb | gzip | gzip-N | zle | lz4
The RELEASE used at creation time. Can be set to any string if needed.
Default: the host's RELEASE
Source: local
origin
Controls the compression algorithm used for this dataset. The lzjb
compression algorithm is optimized for performance while providing
decent data compression. Setting compression to on uses the lzjb
compression algorithm. The gzip compression algorithm uses the same
compression as the gzip(1) command. You can specify the gzip level by using
the value gzip-N where N is an integer from 1 (fastest) to 9 (best
compression ratio). Currently, gzip is equivalent to gzip-6 (which is also
the default for gzip(1)). The zle compression algorithm compresses runs of
zeros.
The lz4 compression algorithm is a high-performance replacement for
the lzjb algorithm. It features significantly faster compression and
decompression, as well as a moderately higher compression ratio than
lzjb, but can only be used on pools with the lz4_compress feature set
to enabled. See zpool-features(7) for details on ZFS feature flags
and the lz4_compress feature.
This property can also be referred to by its shortened column name
compress. Changing this property affects only newly-written data.
Default: lz4
Source: zfs(8)
quota=15G | 50G | ..
This is only set for clones. Read-only.
For cloned file systems or volumes, the snapshot from which the clone
was created. See also the clones property.
Default: -
Source: zfs(8)
mountpoint
Quota for jail.
Limits the amount of space a dataset and its descendents can consume.
This property enforces a hard limit on the amount of space used. This
includes all space consumed by descendents, including file systems
and snapshots. Setting a quota on a descendent of a dataset that
already has a quota does not override the ancestor's quota, but
rather imposes an additional limit.
Default: none
Source: zfs(8)
compressratio
Path for the jail's root filesystem. Don't tweak this or jail won't start!
Default: set to jail's root
Source: zfs(8)
available
Compression ratio. Read-only.
For non-snapshots, the compression ratio achieved for the used space
of this dataset, expressed as a multiplier. The used property
includes descendant datasets, and, for clones, does not include the
space shared with the origin snapshot.
Source: zfs(8)
used
Available space in jail's dataset.
The amount of space available to the dataset and all its children,
assuming that there is no other activity in the pool. Because space
is shared within a pool, availability can be limited by any number of
factors, including physical pool size, quotas, reservations, or other
datasets within the pool.
Source: zfs(8)
dedup=on | off | verify | sha256[,verify]
Used space by jail. Read-only.
The amount of space consumed by this dataset and all its descendents.
This is the value that is checked against this dataset's quota and
reservation. The space used does not include this dataset's reservation,
but does take into account the reservations of any descendent
datasets. The amount of space that a dataset consumes from its parent,
as well as the amount of space that are freed if this dataset is
recursively destroyed, is the greater of its space used and its
reservation.
When snapshots (see the "Snapshots" section) are created, their space
is initially shared between the snapshot and the file system, and
possibly with previous snapshots. As the file system changes, space
that was previously shared becomes unique to the snapshot, and
counted in the snapshot's space used. Additionally, deleting snapshots
can increase the amount of space unique to (and used by) other
snapshots.
The amount of space used, available, or referenced does not take into
account pending changes. Pending changes are generally accounted for
within a few seconds. Committing a change to a disk using fsync(2) or
O_SYNC does not necessarily guarantee that the space usage information
is updated immediately.
Source: zfs(8)
reservation=size | none
Deduplication for jail.
Configures deduplication for a dataset. The default value is off.
The default deduplication checksum is sha256 (this may change in the
future). When dedup is enabled, the checksum defined here overrides
the checksum property. Setting the value to verify has the same
effect as the setting sha256,verify.
If set to verify, ZFS will do a byte-to-byte comparsion in case of
two blocks having the same signature to make sure the block contents
are identical.
Default: off.
Source: zfs(8)
sync_target
Reserved space for jail.
The minimum amount of space guaranteed to a dataset and its descendents.
When the amount of space used is below this value, the dataset
is treated as if it were taking up the amount of space specified by
its reservation. Reservations are accounted for in the parent
datasets' space used, and count against the parent datasets' quotas
and reservations.
Default: none
Source: zfs(8)
sync_tgt_zpool
This is for future use, currently not supported.
rlimits=on | off
This is for future use, currently not supported.
cpuset=1 | 1,2,3,4 | 1-2 | off
If set to "on" resource limits will be enforced.
Default: off
Source: local
Controls the jail's CPU affinity. For more details please refer to
cpuset(1).
Default: off
Source: cpuset(1)
RESOURCE LIMITS
Resource limits (except cpuset and rlimits) use the following value field formatting in the property: limit:action.
Limit defines how much of the resource a process can use before the defined action triggers.
Action defines what will happen when a process exceeds the allowed amount.
Valid actions are:
To better understand what this means please read rctl(8) before enabling any limits.
The following resource limits are supported:
memoryuse=limit:action | off
memorylocked=limit:action | off
Limits the resident set size (DRAM).
Default: 8G:log
Source: rctl(8)
vmemoryuse=limit:action | off
Limits locked memory.
Default: off
Source: rctl(8)
maxproc=limit:action | off
Virtual memory limit (swap + DRAM combined)
Default: off
Source: rctl(8)
cputime=limit:action | off
Limit maximum number of processes.
Default: off
Source: rctl(8)
pcpu=limit:action | off
Limit CPU time, in seconds.
Default: off
Source: rctl(8)
datasize=limit:action | off
Limit %CPU, in percents of a single CPU core or hardware thread.
Default: off
Source: rctl(8)
stacksize=limit:action | off
Limit data size.
Default: off
Source: rctl(8)
coredumpsize=limit:action | off
Limit stack size.
Default: off
Source: rctl(8)
openfiles=limit:action | off
Limit core dump size.
Default: off
Source: rctl(8)
pseudoterminals=limit:action | off
Limit file descriptor table size (number of open files).
Default: off
Source: rctl(8)
swapuse=limit:action | off
Limit number of PTYs.
Default: off
Source: rctl(8)
nthr=limit:action | off
Limit swap usage.
Default: off
Source: rctl(8)
msgqqueued=limit:action | off
Limit number of threads.
Default: off
Source: rctl(8)
msgqsize=limit:action | off
Limit number of queued SysV messages.
Default: off
Source: rctl(8)
nmsgq=limit:action | off
Limit SysV message queue size.
Default: off
Source: rctl(8)
nsemop=limit:action | off
Limit number of SysV message queues.
Default: off
Source: rctl(8)
nshm=limit:action | off
Limit number of SysV semaphores modified in a single semop(2) call.
Default: off
Source: rctl(8)
shmsize=limit:action | off
Limit number of SysV shared memory segments.
Default: off
Source: rctl(8)
wallclock=limit:action | off
Limit SysV shared memory size.
Default: off
Source: rctl(8)
Limit wallclock time.
Default: off
Source: rctl(8)
EXAMPLES
Set up iocell from scratch:
Create first jail:
iocell fetch
List jails:
iocell create tag=myjail
Start jail:
iocell list
Turn on resource limits and apply them:
iocell start UUID
Display resource usage:
iocell set rlimits=on UUID
iocell cap UUID
Convert jail into template:
iocell inuse UUID
List templates:
iocell set istemplate=yes UUID
Clone jail from template:
iocell list -t
Import package on another host
iocell clone UUID-of-template tag=myjail
Get the last successful start time for all jails
iocell import UUID
iocell get -r last_started
HINTS
iocell marks a ZFS pool in the pool's comment field and identifies the active pool for use based on this string.
If using VNET don't forget to add the node's physical NIC into one of the bridges if you need an outside connection. Also read bridge(4) to see how traffic is handled if you are not familiar with this concept (in a nutshell: bridge behaves like a network switch).
PF firewall is not supported inside VNET jails as of July 2014. PF can be enabled for the host however. IPFW is fully supported inside a VNET jail.
Property validation is not handled by iocell (to keep it simple) so please make sure your property values are supported before configuring any properties.
The actual jail name in the jls(8) output is set to ioc-UUID. This is a required workaround as jails will refuse to start with jail(8) when name starts with a "0".
To prevent dmesg leak inside jails apply the following sysctl:
If using VNET consider applying these sysctl's as well:
security.bsd.unprivileged_read_msgbuf=0
For more information please visit:
net.inet.ip.forwarding=1
net.link.bridge.pfil_onlyip=0
net.link.bridge.pfil_bridge=0
net.link.bridge.pfil_member=0
https://github.com/bartekrutkowski/iocell
SEE ALSO
jail(8), ifconfig(8), epair(4), bridge(4), jexec(8), zfs(8), zpool(8), rctl(8), cpuset(1), freebsd-update(8), sysctl(8)
BUGS
In case of bugs/issues/feature requests, please open an issue at https://github.com/bartekrutkowski/iocell/issues
AUTHORS
Peter Toth <peter.toth198@gmail.com> Brandon Schneider <brandonschneider89@gmail.com> Bartek Rutkowski <contracting+github@pixeware.com>
SPECIAL THANKS
Sichendra Bista - for his ever willing attitude and ideas.
| 16 February 2024 |
Visit the GSP FreeBSD Man Page Interface.
Output converted with ManDoc.