X-Git-Url: http://git.linux-vserver.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=doc%2Fconfiguration.xml;h=fc7416fadd4a11f1745b6b9ddbaac62c4ee40c69;hb=710b574e268ecfa9705ac7c73332362da9563fbd;hp=1e06b84d49361056c7dd836164c7f6e4a2a3f5ab;hpb=a2bf8e7cc409ebe34d384fae7a13f4e294352e68;p=util-vserver.git
diff --git a/doc/configuration.xml b/doc/configuration.xml
index 1e06b84..fc7416f 100644
--- a/doc/configuration.xml
+++ b/doc/configuration.xml
@@ -37,14 +37,20 @@ the 'barrier' attribute. Else, common chroot(2) exploits are possible.
-
+
-Enable namespace cleanup globally. It can be overridden for a single vserver
-by setting the nonamespace-cleanup flag
+Disable namespace cleanup globally. It can be overridden for a single vserver
+by setting the namespace-cleanup flag
there.
+
+
+List of paths to skip during namespace cleanup.
+
+
+
Path of the vserver run reverse directory. This directory contains
@@ -56,12 +62,40 @@ NOTE: this link exists in 0.30.202+ only; in previous versions it was
a vserver specific setting.
+
+
+
+Contains the pathname of the shell which will be used by the "vserver
+... enter" command. Can be overridden by the
+per-guest shell.
+
+
+
+
+
+The default fstab file to put in newly built guests.
+
+
+
+
+
+If this file exists, kernel-side dynamic contexts will be used by the "vserver
+... build" command. Otherwise a context will be generated, based on the
+contents of context.next.
+
+
+
+
+
+The context id to use for the next guest created by "vserver ... build".
+
+ /vservers/.pkg
-
+
The default apt.conf which is going to be used. It is overridden by
distribution specific configuration file.
@@ -83,6 +117,33 @@ from this uri and install it at a temporary place.
+
+
+ 'x' is an arbitrary name, replace it with e.g. device names
+
+ When this file exists, the device can be opened
+
+
+ When this file exists, the device can be created (if the guest has CAP_MKNOD)
+
+
+
+When this file exists, target will
+have to exist as well and opening the device will in fact open the target device
+
+
+
+ This file will let you specify unimplemented flags manually
+
+
+ Contains the name of a device node
+
+
+ Contains the device node of the target node to open instead of the device when remap is set
+
+
+
+
@@ -95,7 +156,7 @@ is invoked from the kernel. This should point somewhere e.g. into
-See vshelper/action.
+See vshelper/action.
@@ -141,7 +202,7 @@ at startup via initscript.
-
+
The environment to set when starting guests. Contains one VAR=VAL
pair per line.
@@ -190,16 +251,105 @@ operation will fail. Most tools assume 'on' as the default value.
+
+
+
+
+When this file exists, vlogin isn't used on vserver <guest> enter.
+
+
+
+
+
+
+
+A list of default options to supply to vserver ... build, one option per line.
+
+
+
-
+
When this file exists, the steps which setup and destroy a VLAN
interface will be executed.
+
+
+ This directory contains some files which will be copied to the guests during build.
+
+ The default /etc/hosts file.
+
+
+ The default /etc/nsswitch.conf file.
+
+
+ The default /etc/krb5.conf file.
+
+
+ The default /etc/krb.conf file.
+
+
+ The default /etc/krb.realms file.
+
+
+ The default /etc/ldap.conf file.
+
+
+ The default /etc/localtime file.
+
+
+ The default /etc/resolv.conf file.
+
+
+
+
+
+This directory contains cgroup settings which should be applied to all guests.
+See your kernel documentation for what settings are valid with your
+configuration.
+
+
+
+The directory to mount the cgroup hierarchy at. The default is /dev/cgroup.
+
+
+
+
+List of subsystems to enable on the cgroup mount point. The default is "all".
+
+
+
+
+Some subsystems start out with clean slates, making it impossible to use the
+cgroup before certain things have been set. This is true for e.g. the cpuset
+subsystem. This file contains a list of filenames which should be explicitly
+inherited from the parent (root) cgroup, if not overridden elsewhere.
+The default is cpuset.cpus and cpuset.mems.
+
+
+
+
+If this file exists, all guests will be put in one cgroup named after the
+contents of this file. The default is to put each guest in a cgroup named the
+same thing as the guest.
+
+
+
+
+If this file exists, all cgroups will be created underneath this directory
+in the cgroup mount point.
+
+
+
+
+If this file exists, each subsystem will have its own mount point.
+
+
+
@@ -307,6 +457,11 @@ this distribution.
+
+
+Symlink to the default template for this distribution.
+
+
@@ -338,19 +493,32 @@ the vserver is stopped, this can be a dangling symlink.
-[experimental; name is subject of possible change] Contains the system capabilities. See
+Contains the system capabilities. See
lib/bcaps-v13.c
for possible values.
-[experimental; name is subject of possible change] Contains the
-context capabilities. See lib/ccaps-v13.c
for possible values.
+
+
+Contains the network capabilities. See lib/ncaps-net.c
+for possible values.
+
+
+
+
+Contains the namespaces allowed to be unshared. See lib/umask-v23.c
+for possible values.
+
+
@@ -368,22 +536,24 @@ the 'barrier' attribute. Else, common chroot(2) exploits are possible.
-
+
-Overrides the global namespace-cleanup flag and disables
+Overrides the global nonamespace-cleanup flag and enables
namespace cleanup for the current vserver.
-
+
-Enable namespace cleanup for the current vserver.
+List of paths to skip during namespace cleanup. This overrides the global
+namespace-cleanup-skip
+file.
-
+
-[experimental; name is subject of possible change] Contains the
+[deprecated; use sched instead] Contains the
scheduler parameters, one per line.
The Hard CPU limit uses a mechanism called a Token Bucket. the
@@ -449,7 +619,7 @@ Contains per line a capability. This file is used for the 2.4 kernel
only; for 2.6 use bcapabilities.
-
+
Contains the pathname of the shell which will be used by the "vserver
... enter" command.
@@ -463,57 +633,40 @@ is the personality-type followed by flags (one item per line). See
values.
-
+
Contains per line a flag. See lib/cflags-v13.c
+url="http://svn.linux-vserver.org/svn/util-vserver/trunk/lib/cflags-v13.c">lib/cflags-v13.c
+for possible values.
+
+
+
+
+Old name for cflags. Deprecated.
+
+
+
+
+Contains a network flag per line. See lib/nflags-net.c
for possible values.
-
-Contains the context which shall be used for the vserver.
+Contains the context id which shall be used for the vserver. Valid range is 2 through 49151.
+
+
+Contains the network context id which shall be used for the vserver. Valid range is 2 through 49151.
+
+
+
+
+Contains the filesystem tag which shall be used for the vserver. Valid range is 2 through 49151.
+
+
The fstab file for the vserver. Entries in this file will be mounted
@@ -526,11 +679,16 @@ the 'fstab' file should be used.
The fstab file for the vserver. Entries in this file will be mounted
-within the network context of the host; this means that mount will be
+within the network context of the guest; this means that mount will be
called as chbind <options> mount .... See
fstab also.
+
+
+Contains the bias to be applied to processes in this guest when the OOM-killer strikes.
+
+
@@ -638,14 +796,41 @@ vserver ids (one name per line).
-
+
The environment to set when starting the guest. Contains one VAR=VAL
pair per line.
-
+
+
+
+ 'x' is an arbitrary name, replace it with e.g. device names
+
+ When this file exists, the device can be opened
+
+
+ When this file exists, the device can be created (if the guest has CAP_MKNOD)
+
+
+
+When this file exists, target will
+have to exist as well and opening the device will in fact open the target device
+
+
+
+ This file will let you specify unimplemented flags manually
+
+
+ Contains the name of a device node
+
+
+ Contains the device node of the target node to open instead of the device when remap is set
+
+
+
+
30
@@ -656,7 +841,7 @@ assumed.
-
+ restart
The action which is going to be executed when a vshelper event
@@ -713,7 +898,7 @@ skipped.
-See vshelper/action.
+See vshelper/action.
@@ -792,7 +977,7 @@ environment variable must be set by one of the in-shellcontext scripts
(the non-executable ones).
-
+
The scriptlet which will be executed before the root filesystem is mounted and
the configuration has been loaded. Before executing the script, the
@@ -811,11 +996,11 @@ The name of the current vserver.
-
+
Repository of initialize like scripts. Before executing the script,
the configuration directory will be made the working directory.
-
+
See initialize.
@@ -856,7 +1041,7 @@ The name of the current vserver.
Repository of prepre-start like scripts. Before executing the script,
the configuration directory will be made the working directory.
-
+
See prepre-start.
@@ -898,7 +1083,7 @@ The name of the current vserver.
Repository of pre-start like scripts. Before executing these scripts,
the vserver root directory will be made the working directory.
-
+
See pre-start.
@@ -982,7 +1167,7 @@ The name of the current vserver.
Repository of pre-stop like scripts. Before executing the script, the
vserver root directory will be made the working directory.
-
+
See pre-stop.
@@ -1086,10 +1271,13 @@ The name of the current vserver.
- The default broadcast address.
+ The guest's broadcast address.
+
+
+ The guest's loopback address.
- The default network device.
+ The default network device. See iface/dev for more information.The default network prefix-length.
@@ -1100,7 +1288,7 @@ The name of the current vserver.
The default scope of the network interfaces.
-
+
When this file exists, the steps which setup and destroy a VLAN
interface will be executed for all interfaces of this vserver.
@@ -1128,13 +1316,10 @@ numbers like '00' are good names for these directories.
- The ip which will be assigned to this interface.
+ The IP address which will be assigned to this interface.
-
- The broadcast address.
-
-
- The network device.
+
+ The network device. When this is specified, the IP address will be assigned and removed when starting and stopping the guest.The network prefix-length.
@@ -1168,36 +1353,81 @@ interface will be skipped. This will override the global
vlandev.
-
+
When this file exists, the steps which setup and destroy a VLAN
interface will be executed.
+
+
+When this file exists, a tun interface will be created when the guest
+starts, and removed when the guest stops.
+
+
+
+
+When this file exists, a tap interface will be created when the guest
+starts, and removed when the guest stops.
+
+
+
+
+If a tun or tap interface is created, the presence of this file will
+disable checksumming on it.
+
+
+
+
+When this file exists, the tun/tap interface created will not be
+owned by this particular guest.
+
+
+
+
+Sets the ownership of tun/tap interfaces. Requires a user id, not a
+username.
+
+
+
+
+Sets the group ownership of tun/tap interfaces. Requires a group id,
+not a group name.
+
+
+
+
+Sets the link type of tun/tap interfaces.
+
+
+
+
+Sets the peer for a point-to-point link, such as a tun interface.
+
+
A directory with ulimits. Possible resources are cpu, data, fsize,
-locks, memlock, nofile, nproc, rss and/or stack. This configuration
-will be honored for kernel 2.4 only.
-
+locks, memlock, nofile, nproc, rss and/or stack.
+
A file which contains the hard- and soft-limit of the given resource
in the first line. The special keyword 'inf' is recognized.
-
+
-A file which contains the hard- of the given resource in the first
+A file which contains the hard-limit of the given resource in the first
line. The special keyword 'inf' is recognized.
-
+
-A file which contains the soft- of the given resource in the first
+A file which contains the soft-limit of the given resource in the first
line. The special keyword 'inf' is recognized.
@@ -1206,30 +1436,31 @@ line. The special keyword 'inf' is recognized.
A directory with resource limits. Possible resources are cpu, fsize,
-data, stack, core, rss, nproc, nofile, memlock, as and locks. This
+data, stack, core, rss, nproc, nofile, memlock, as, locks, msgqueue,
+nsock, openfd, anon, shmem, semary, nsems and dentry. This
configuration will be honored for kernel 2.6 only.
-
+
A file which contains the hard- and soft-limit of the given resource
in the first line. The special keyword 'inf' is recognized.
-
+
-A file which contains the hard- of the given resource in the first
+A file which contains the hard-limit of the given resource in the first
line. The special keyword 'inf' is recognized.
-
+
-A file which contains the soft- of the given resource in the first
+A file which contains the soft-limit of the given resource in the first
line. The special keyword 'inf' is recognized.
-A file which contains the guaranted minimum of the given resource in
+A file which contains the guaranteed minimum of the given resource in
the first line. The special keyword 'inf' is recognized.
@@ -1263,8 +1494,12 @@ tools and can *not* be modified.
-
-
+
+
+Note that all entries are required for the disk limit to be applied.
+
+
+ 'x' is an arbitrary name, replace it with e.g. integersThe directory to which the limit should be applied
@@ -1280,7 +1515,22 @@ tools and can *not* be modified.
-
+
+
+ 'x' is an arbitrary name, replace it with e.g. integers
+
+ The sysctl setting
+
+
+ The value
+
+
+
+
+
+
+Puts the guest in a cpuset. Required entries are name, cpus and mems.
+ The name of the cpuset for this vserver
@@ -1300,5 +1550,116 @@ tools and can *not* be modified.
When this file exists, the cpuset will be assumed to exist already
+
+
+
+ The initial amount of tokens to put in the bucket
+
+
+ The minimum amount of tokens required to unhold the context
+
+
+ The bucket's size
+
+
+ Amount of tokens to add each interval
+
+
+ The interval between refills of the bucket
+
+
+ Bias added to priorities calculated within the guest (result is clamped to -20/+19)
+
+
+ Amount of tokens to add each interval2 when advancing idle time
+
+
+ The interval between refills of the bucket when advancing idle time
+
+
+ When this file exists, advancing idle time is activated
+
+
+ This directory contains per-CPU and/or per-bucket specific settings. Remember to set the cpu-id file. All CPUs inherit the global settings.
+
+ The initial amount of tokens to put in the bucket
+
+
+ The minimum amount of tokens required to unhold the context
+
+
+ The bucket's size
+
+
+ Amount of tokens to add each interval
+
+
+ The interval between refills of the bucket
+
+
+ Bias added to priorities calculated within the guest (result is clamped to -20/+19)
+
+
+ Amount of tokens to add each interval2 when advancing idle time
+
+
+ The interval between refills of the bucket when advancing idle time
+
+
+ When this file exists, advancing idle time is activated
+
+
+ The CPU to apply these settings to
+
+
+ The bucket to apply these settings to
+
+
+
+
+
+
+ Enables pid virtualization for this guest
+
+
+ Enable network virtualization for this guest
+
+
+ Contains a mask of spaces to clone/enter
+
+
+
+
+
+ The IO scheduling class to use for this guest (see ionice(1)).
+
+
+ The IO scheduling priority to use for this guest (see ionice(1)).
+
+
+
+
+
+This directory contains cgroup settings to be applied to this guest.
+See your kernel documentation for what settings are valid with your
+configuration.
+
+
+
+If this file exists, the guest will be put in a cgroup named after the
+contents of this file. The default is to name the cgroup the same thing as the
+guest, unless
+.defaults/cgroup/name says
+otherwise.
+
+
+
+
+
+If this file exists,
+.defaults/cgroup will be ignored
+for this guest.
+
+