Ubuntu Manpage: gnt-cluster - Ganeti administration, cluster-wide

Provided by: ganeti-3.0_3.0.2-1ubuntu1_all

Name

       gnt-cluster - Ganeti administration, cluster-wide

Synopsis

       gnt-cluster {command} [arguments...]

DESCRIPTION

       The gnt-cluster is used for cluster-wide administration in the Ganeti system.

COMMANDS

ACTIVATE-MASTER-IP
activate-master-ip

Activates the master IP on the master node.

COMMAND
command [-n node-name] [-g group] [-M] {command}

Executes a command on all nodes. This command is designed for simple usage. For more complex use cases
the commands dsh(1) or cssh(1) should be used instead.

If the option -n is not given, the command will be executed on all nodes, otherwise it will be executed
only on the node(s) specified. Use the option multiple times for running it on multiple nodes, like:

# gnt-cluster command -n node1.example.com -n node2.example.com date

The -g option can be used to run a command only on a specific node group, e.g.:

# gnt-cluster command -g default date

The -M option can be used to prepend the node name to all output lines. The --failure-only option hides
successful commands, making it easier to see failures.

The command is executed serially on the selected nodes. If the master node is present in the list, the
command will be executed last on the master. Regarding the other nodes, the execution order is somewhat
alphabetic, so that node2.example.com will be earlier than node10.example.com but after node1.exam‐
ple.com.

So given the node names node1, node2, node3, node10, node11, with node3 being the master, the order will
be: node1, node2, node10, node11, node3.

The command is constructed by concatenating all other command line arguments. For example, to list the
contents of the /etc directory on all nodes, run:

# gnt-cluster command ls -l /etc

and the command which will be executed will be ls -l /etc.

COPYFILE
copyfile [--use-replication-network] [-n node-name] [-g group]
{file}

Copies a file to all or to some nodes. The argument specifies the source file (on the current system),
the -n argument specifies the target node, or nodes if the option is given multiple times. If -n is not
given at all, the file will be copied to all nodes. The -g option can be used to only select nodes in a
specific node group. Passing the --use-replication-network option will cause the copy to be done over
the replication network (only matters if the primary/secondary IPs are different). Example:

# gnt-cluster copyfile -n node1.example.com -n node2.example.com /tmp/test

This will copy the file /tmp/test from the current node to the two named nodes.

DEACTIVATE-MASTER-IP
deactivate-master-ip [--yes]

Deactivates the master IP on the master node.

This should be run only locally or on a connection to the node ip directly, as a connection to the master
ip will be broken by this operation. Because of this risk it will require user confirmation unless the
--yes option is passed.

DESTROY
destroy {--yes-do-it}

Remove all configuration files related to the cluster, so that a gnt-cluster init can be done again af‐
terwards.

Since this is a dangerous command, you are required to pass the argument --yes-do-it.

EPO
epo [--on] [--groups|--all] [--power-delay] arguments

Performs an emergency power-off on nodes given as arguments. If --groups is given, arguments are node
groups. If --all is provided, the whole cluster will be shut down.

The --on flag recovers the cluster after an emergency power-off. When powering on the cluster you can
use --power-delay to define the time in seconds (fractions allowed) waited between powering on individual
nodes.

Please note that the master node will not be turned down or up automatically. It will just be left in a
state, where you can manually perform the shutdown of that one node. If the master is in the list of af‐
fected nodes and this is not a complete cluster emergency power-off (e.g. using --all), you're required
to do a master failover to another node not affected.

GETMASTER
getmaster

Displays the current master node.

INFO
info [--roman]

Shows runtime cluster information: cluster name, architecture (32 or 64 bit), master node, node list and
instance list.

Passing the --roman option gnt-cluster info will try to print its integer fields in a latin friendly way.
This allows further diffusion of Ganeti among ancient cultures.

SHOW-ISPECS-CMD
show-ispecs-cmd

Shows the command line that can be used to recreate the cluster with the same options relative to specs
in the instance policies.

INIT
init
[{-s|--secondary-ip} secondary_ip]
[--vg-name vg-name]
[--master-netdev interface-name]
[--master-netmask netmask]
[--use-external-mip-script {yes | no}]
[{-m|--mac-prefix} mac-prefix]
[--no-etc-hosts]
[--no-ssh-init]
[--file-storage-dir dir]
[--shared-file-storage-dir dir]
[--gluster-storage-dir dir]
[--enabled-hypervisors hypervisors]
[{-H|--hypervisor-parameters} hypervisor:hv-param=*value*[,*hv-param*=*value*...]]
[{-B|--backend-parameters} be-param=*value*[,*be-param*=*value*...]]
[{-N|--nic-parameters} nic-param=*value*[,*nic-param*=*value*...]]
[{-D|--disk-parameters} disk-template:disk-param=*value*[,*disk-param*=*value*...]]
[--maintain-node-health {yes | no}]
[--uid-pool user-id pool definition]
[{-I|--default-iallocator} default instance allocator]
[--default-iallocator-params ial-param=*value*,*ial-param*=*value*]
[--primary-ip-version version]
[--prealloc-wipe-disks {yes | no}]
[--node-parameters ndparams]
[{-C|--candidate-pool-size} candidate_pool_size]
[--specs-cpu-count spec-param=*value* [,*spec-param*=*value*...]]
[--specs-disk-count spec-param=*value* [,*spec-param*=*value*...]]
[--specs-disk-size spec-param=*value* [,*spec-param*=*value*...]]
[--specs-mem-size spec-param=*value* [,*spec-param*=*value*...]]
[--specs-nic-count spec-param=*value* [,*spec-param*=*value*...]]
[--ipolicy-std-specs spec=*value* [,*spec*=*value*...]]
[--ipolicy-bounds-specs bounds_ispecs]
[--ipolicy-disk-templates template [,*template*...]]
[--ipolicy-spindle-ratio ratio]
[--ipolicy-vcpu-ratio ratio]
[--disk-state diskstate]
[--hypervisor-state hvstate]
[--drbd-usermode-helper helper]
[--enabled-disk-templates template [,*template*...]]
[--install-image image]
[--zeroing-image image]
[--compression-tools [tool, [tool]]]
[--user-shutdown {yes | no}]
[--ssh-key-type type]
[--ssh-key-bits bits]
{cluster-name}

This commands is only run once initially on the first node of the cluster. It will initialize the clus‐
ter configuration, setup the ssh-keys, start the daemons on the master node, etc. in order to have a
working one-node cluster.

Note that the cluster-name is not any random name. It has to be resolvable to an IP address using DNS,
and it is best if you give the fully-qualified domain name. This hostname must resolve to an IP address
reserved exclusively for this purpose, i.e. not already in use.

The cluster can run in two modes: single-home or dual-homed. In the first case, all traffic (both public
traffic, inter-node traffic and data replication traffic) goes over the same interface. In the dual-
homed case, the data replication traffic goes over the second network. The -s (--secondary-ip) option
here marks the cluster as dual-homed and its parameter represents this node's address on the second net‐
work. If you initialise the cluster with -s, all nodes added must have a secondary IP as well.

Note that for Ganeti it doesn't matter if the secondary network is actually a separate physical network,
or is done using tunnelling, etc. For performance reasons, it's recommended to use a separate network,
of course.

The --vg-name option will let you specify a volume group different than "xenvg" for Ganeti to use when
creating instance disks. This volume group must have the same name on all nodes. Once the cluster is
initialized this can be altered by using the modify command. Note that if the volume group name is modi‐
fied after the cluster creation and DRBD support is enabled you might have to manually modify the metavg
as well.

If you don't want to use lvm storage at all use the --enabled-disk-templates option to restrict the set
of enabled disk templates. Once the cluster is initialized you can change this setup with the modify
command.

The --master-netdev option is useful for specifying a different interface on which the master will acti‐
vate its IP address. It's important that all nodes have this interface because you'll need it for a mas‐
ter failover.

The --master-netmask option allows to specify a netmask for the master IP. The netmask must be specified
as an integer, and will be interpreted as a CIDR netmask. The default value is 32 for an IPv4 address
and 128 for an IPv6 address.

The --use-external-mip-script option allows to specify whether to use an user-supplied master IP address
setup script, whose location is @SYSCONFDIR@/ganeti/scripts/master-ip-setup. If the option value is set
to False, the default script (located at @PKGLIBDIR@/tools/master-ip-setup) will be executed.

The -m (--mac-prefix) option will let you specify a three byte prefix under which the virtual MAC ad‐
dresses of your instances will be generated. The prefix must be specified in the format XX:XX:XX and the
default is aa:00:00.

The --no-etc-hosts option allows you to initialize the cluster without modifying the /etc/hosts file.

The --no-ssh-init option allows you to initialize the cluster without creating or distributing SSH key
pairs. This also sets the cluster-wide configuration parameter modify ssh setup to False. When adding
nodes, Ganeti will consider this parameter to determine whether to create and distribute SSH key pairs on
new nodes as well.

The --file-storage-dir, --shared-file-storage-dir and --gluster-storage-dir options allow you set the di‐
rectory to use for storing the instance disk files when using respectively the file storage backend, the
shared file storage backend and the gluster storage backend. Note that these directories must be an al‐
lowed directory for file storage. Those directories are specified in the @SYSCONFDIR@/ganeti/file-stor‐
age-paths file. The file storage directory can also be a subdirectory of an allowed one. The file stor‐
age directory should be present on all nodes.

The --prealloc-wipe-disks sets a cluster wide configuration value for wiping disks prior to allocation
and size changes (gnt-instance grow-disk). This increases security on instance level as the instance
can't access untouched data from its underlying storage.

The --enabled-hypervisors option allows you to set the list of hypervisors that will be enabled for this
cluster. Instance hypervisors can only be chosen from the list of enabled hypervisors, and the first en‐
try of this list will be used by default. Currently, the following hypervisors are available:

xen-pvm
Xen PVM hypervisor

xen-hvm
Xen HVM hypervisor

kvm Linux KVM hypervisor

chroot a simple chroot manager that starts chroot based on a script at the root of the filesystem holding
the chroot

fake fake hypervisor for development/testing

Either a single hypervisor name or a comma-separated list of hypervisor names can be specified. If this
option is not specified, only the xen-pvm hypervisor is enabled by default.

The --user-shutdown option enables or disables user shutdown detection at the cluster level. User shut‐
down detection allows users to initiate instance poweroff from inside the instance, and Ganeti will re‐
port the instance status as 'USER_down' (as opposed, to 'ERROR_down') and the watcher will not restart
these instances, thus preserving their instance status. This option is disabled by default. For KVM,
the hypervisor parameter user_shutdown must also be set, either at the cluster level or on a per-instance
basis (see gnt-instance(8)).

The -H (--hypervisor-parameters) option allows you to set default hypervisor specific parameters for the
cluster. The format of this option is the name of the hypervisor, followed by a colon and a comma-sepa‐
rated list of key=value pairs. The keys available for each hypervisors are detailed in the gnt-in‐
stance(8) man page, in the add command plus the following parameters which are only configurable globally
(at cluster level):

migration_port
Valid for the Xen PVM and KVM hypervisors.

This options specifies the TCP port to use for live-migration when using the xm toolstack. The
same port should be configured on all nodes in the @XEN_CONFIG_DIR@/xend-config.sxp file, under
the key "xend-relocation-port".

migration_bandwidth
Valid for the KVM hypervisor.

This option specifies the maximum bandwidth that KVM will use for instance live migrations. The
value is in MiB/s.

This option is only effective with kvm versions >= 78 and qemu-kvm versions >= 0.10.0.

The -B (--backend-parameters) option allows you to set the default backend parameters for the cluster.
The parameter format is a comma-separated list of key=value pairs with the following supported keys:

vcpus Number of VCPUs to set for an instance by default, must be an integer, will be set to 1 if no
specified.

maxmem Maximum amount of memory to allocate for an instance by default, can be either an integer or an
integer followed by a unit (M for mebibytes and G for gibibytes are supported), will be set to
128M if not specified.

minmem Minimum amount of memory to allocate for an instance by default, can be either an integer or an
integer followed by a unit (M for mebibytes and G for gibibytes are supported), will be set to
128M if not specified.

auto_balance
Value of the auto_balance flag for instances to use by default, will be set to true if not speci‐
fied.

always_failover
Default value for the always_failover flag for instances; if not set, False is used.

The -N (--nic-parameters) option allows you to set the default network interface parameters for the clus‐
ter. The parameter format is a comma-separated list of key=value pairs with the following supported
keys:

mode The default NIC mode, one of routed, bridged or openvswitch.

link In bridged or openvswitch mode the default interface where to attach NICs. In routed mode it rep‐
resents an hypervisor-vif-script dependent value to allow different instance groups. For example
under the KVM default network script it is interpreted as a routing table number or name. Open‐
vswitch support is also hypervisor dependent and currently works for the default KVM network
script. Under Xen a custom network script must be provided.

The -D (--disk-parameters) option allows you to set the default disk template parameters at cluster lev‐
el. The format used for this option is similar to the one use by the -H option: the disk template name
must be specified first, followed by a colon and by a comma-separated list of key-value pairs. These pa‐
rameters can only be specified at cluster and node group level; the cluster-level parameter are inherited
by the node group at the moment of its creation, and can be further modified at node group level using
the gnt-group(8) command.

The following is the list of disk parameters available for the drbd template, with measurement units
specified in square brackets at the end of the description (when applicable):

resync-rate
Static re-synchronization rate. [KiB/s]

data-stripes
Number of stripes to use for data LVs.

meta-stripes
Number of stripes to use for meta LVs.

disk-barriers
What kind of barriers to disable for disks. It can either assume the value "n", meaning no barri‐
er disabled, or a non-empty string containing a subset of the characters "bfd". "b" means disable
disk barriers, "f" means disable disk flushes, "d" disables disk drains.

meta-barriers
Boolean value indicating whether the meta barriers should be disabled (True) or not (False).

metavg String containing the name of the default LVM volume group for DRBD metadata. By default, it is
set to xenvg. It can be overridden during the instance creation process by using the metavg key
of the --disk parameter.

disk-custom
String containing additional parameters to be appended to the arguments list of drbdsetup disk.

net-custom
String containing additional parameters to be appended to the arguments list of drbdsetup net.

protocol
Replication protocol for the DRBD device. Has to be either "A", "B" or "C". Refer to the DRBD
documentation for further information about the differences between the protocols.

dynamic-resync
Boolean indicating whether to use the dynamic resync speed controller or not. If enabled, c-plan-
ahead must be non-zero and all the c-* parameters will be used by DRBD. Otherwise, the value of
resync-rate will be used as a static resync speed.

c-plan-ahead
Agility factor of the dynamic resync speed controller. (the higher, the slower the algorithm will
adapt the resync speed). A value of 0 (that is the default) disables the controller. [ds]

c-fill-target
Maximum amount of in-flight resync data for the dynamic resync speed controller. [sectors]

c-delay-target
Maximum estimated peer response latency for the dynamic resync speed controller. [ds]

c-min-rate
Minimum resync speed for the dynamic resync speed controller. [KiB/s]

c-max-rate
Upper bound on resync speed for the dynamic resync speed controller. [KiB/s]

List of parameters available for the plain template:

stripes
Number of stripes to use for new LVs.

List of parameters available for the rbd template:

pool The RADOS cluster pool, inside which all rbd volumes will reside. When a new RADOS cluster is de‐
ployed, the default pool to put rbd volumes (Images in RADOS terminology) is 'rbd'.

access If 'userspace', instances will access their disks directly without going through a block device,
avoiding expensive context switches with kernel space and the potential for deadlocks
(http://tracker.ceph.com/issues/3076) in low memory scenarios.

The default value is 'kernelspace' and it disables this behaviour. This setting may only be
changed to 'userspace' if all instance disks in the affected group or cluster can be accessed in
userspace.

Attempts to use this feature without rbd support compiled in KVM result in a "no such file or di‐
rectory" error messages.

The option --maintain-node-health allows one to enable/disable automatic maintenance actions on nodes.
Currently these include automatic shutdown of instances and deactivation of DRBD devices on offline
nodes; in the future it might be extended to automatic removal of unknown LVM volumes, etc. Note that
this option is only useful if the use of ganeti-confd was enabled at compilation.

The --uid-pool option initializes the user-id pool. The user-id pool definition can contain a list of
user-ids and/or a list of user-id ranges. The parameter format is a comma-separated list of numeric
user-ids or user-id ranges. The ranges are defined by a lower and higher boundary, separated by a dash.
The boundaries are inclusive. If the --uid-pool option is not supplied, the user-id pool is initialized
to an empty list. An empty list means that the user-id pool feature is disabled.

The -I (--default-iallocator) option specifies the default instance allocator. The instance allocator
will be used for operations like instance creation, instance and node migration, etc. when no manual
override is specified. If this option is not specified and htools was not enabled at build time, the de‐
fault instance allocator will be blank, which means that relevant operations will require the administra‐
tor to manually specify either an instance allocator, or a set of nodes. If the option is not specified
but htools was enabled, the default iallocator will be hail(1) (assuming it can be found on disk). The
default iallocator can be changed later using the modify command.

The option --default-iallocator-params sets the cluster-wide iallocator parameters used by the default
iallocator only on instance allocations.

The --primary-ip-version option specifies the IP version used for the primary address. Possible values
are 4 and 6 for IPv4 and IPv6, respectively. This option is used when resolving node names and the clus‐
ter name.

The --node-parameters option allows you to set default node parameters for the cluster. Please see
ganeti(7) for more information about supported key=value pairs.

The -C (--candidate-pool-size) option specifies the candidate_pool_size cluster parameter. This is the
number of nodes that the master will try to keep as master_candidates. For more details about this role
and other node roles, see the ganeti(7).

The --specs-... and --ipolicy-... options specify the instance policy on the cluster. The --ipolicy-
bounds-specs option sets the minimum and maximum specifications for instances. The format is:
min:param=*value*,.../max:param=*value*,... and further specifications pairs can be added by using // as
a separator. The --ipolicy-std-specs option takes a list of parameter/value pairs. For both options,
param can be:

• cpu-count: number of VCPUs for an instance

• disk-count: number of disk for an instance

• disk-size: size of each disk

• memory-size: instance memory

• nic-count: number of network interface

• spindle-use: spindle usage for an instance

For the --specs-... options, each option can have three values: min, max and std, which can also be modi‐
fied on group level (except for std, which is defined once for the entire cluster). Please note, that
std values are not the same as defaults set by --beparams, but they are used for the capacity calcula‐
tions.

• --specs-cpu-count limits the number of VCPUs that can be used by an instance.

• --specs-disk-count limits the number of disks

• --specs-disk-size limits the disk size for every disk used

• --specs-mem-size limits the amount of memory available

• --specs-nic-count sets limits on the number of NICs used

The --ipolicy-spindle-ratio option takes a decimal number. The --ipolicy-disk-templates option takes a
comma-separated list of disk templates. This list of disk templates must be a subset of the list of
cluster-wide enabled disk templates (which can be set with --enabled-disk-templates).

• --ipolicy-spindle-ratio limits the instances-spindles ratio

• --ipolicy-vcpu-ratio limits the vcpu-cpu ratio

All the instance policy elements can be overridden at group level. Group level overrides can be removed
by specifying default as the value of an item.

The --drbd-usermode-helper option can be used to specify a usermode helper. Check that this string is
the one used by the DRBD kernel.

For details about how to use --hypervisor-state and --disk-state have a look at ganeti(7).

The --enabled-disk-templates option specifies a list of disk templates that can be used by instances of
the cluster. For the possible values in this list, see gnt-instance(8). Note that in contrast to the
list of disk templates in the ipolicy, this list is a hard restriction. It is not possible to create in‐
stances with disk templates that are not enabled in the cluster. It is also not possible to disable a
disk template when there are still instances using it. The first disk template in the list of enabled
disk template is the default disk template. It will be used for instance creation, if no disk template
is requested explicitly.

The --install-image option specifies the location of the OS image to use to run the OS scripts inside a
virtualized environment. This can be a file path or a URL. In the case that a file path is used, nodes
are expected to have the install image located at the given path, although that is enforced during a in‐
stance create with unsafe OS scripts operation only.

The --zeroing-image option specifies the location of the OS image to use to zero out the free space of an
instance. This can be a file path or a URL. In the case that a file path is used, nodes are expected to
have the zeroing image located at the given path, although that is enforced during a zeroing operation
only.

The --compression-tools option specifies the tools that can be used to compress the disk data of in‐
stances in transfer. The default tools are: 'gzip', 'gzip-slow', and 'gzip-fast'. For compatibility
reasons, the 'gzip' tool cannot be excluded from the list of compression tools. Ganeti knows how to use
certain tools, but does not provide them as a default as they are not commonly present: currently only
'lzop'. The user should indicate their presence by specifying them through this option. Any other cus‐
tom tool specified must have a simple executable name ('[-_a-zA-Z0-9]+'), accept input on stdin, and pro‐
duce output on stdout. The '-d' flag specifies that decompression rather than compression is taking
place. The '-h' flag must be supported as a means of testing whether the executable exists. These re‐
quirements are compatible with the gzip command line options, allowing many tools to be easily wrapped
and used.

The --ssh-key-type and --ssh-key-bits options determine the properties of the SSH keys Ganeti generates
and uses to execute commands on nodes. The supported types are currently 'dsa', 'rsa', and 'ecdsa'. The
supported bit sizes vary across keys, reflecting the options ssh-keygen(1) exposes. These are currently:

• dsa: 1024 bits

• rsa: >=768 bits

• ecdsa: 256, 384, or 521 bits

Ganeti defaults to using 2048-bit RSA keys.

MASTER-FAILOVER
master-failover [--no-voting] [--yes-do-it]
[--ignore-offline-nodes]

Failover the master role to the current node.

The --no-voting option skips the remote node agreement checks. This is dangerous, but necessary in some
cases (for example failing over the master role in a 2 node cluster with the original master down). If
the original master then comes up, it won't be able to start its master daemon because it won't have
enough votes, but so won't the new master, if the master daemon ever needs a restart. You can pass --no-
voting to ganeti-luxid and ganeti-wconfd on the new master to solve this problem, and run gnt-cluster re‐
dist-conf to make sure the cluster is consistent again.

The option --yes-do-it is used together with --no-voting, for skipping the interactive checks. This is
even more dangerous, and should only be used in conjunction with other means (e.g. a HA suite) to con‐
firm that the operation is indeed safe.

Note that in order for remote node agreement checks to work, a strict majority of nodes still needs to be
functional. To avoid situations with daemons not starting up on the new master, master-failover without
the --no-voting option verifies a healthy majority of nodes and refuses the operation otherwise.

The --ignore-offline-nodes flag ignores offline nodes when the cluster is voting on the master. Any
nodes that are offline are not counted towards the vote or towards the healthy nodes required for a ma‐
jority, as they will be brought into sync with the rest of the cluster during a node readd operation.

MASTER-PING
master-ping

Checks if the master daemon is alive.

If the master daemon is alive and can respond to a basic query (the equivalent of gnt-cluster info), then
the exit code of the command will be 0. If the master daemon is not alive (either due to a crash or be‐
cause this is not the master node), the exit code will be 1.

MODIFY
modify [--submit] [--print-jobid]
[--force]
[--vg-name vg-name]
[--enabled-hypervisors hypervisors]
[{-H|--hypervisor-parameters} hypervisor:hv-param=*value*[,*hv-param*=*value*...]]
[{-B|--backend-parameters} be-param=*value*[,*be-param*=*value*...]]
[{-N|--nic-parameters} nic-param=*value*[,*nic-param*=*value*...]]
[{-D|--disk-parameters} disk-template:disk-param=*value*[,*disk-param*=*value*...]]
[--uid-pool user-id pool definition]
[--add-uids user-id pool definition]
[--remove-uids user-id pool definition]
[{-C|--candidate-pool-size} candidate_pool_size]
[--max-running-jobs count ]
[--max-tracked-jobs count ]
[--maintain-node-health {yes | no}]
[--prealloc-wipe-disks {yes | no}]
[{-I|--default-iallocator} default instance allocator]
[--default-iallocator-params ial-param=*value*,*ial-param*=*value*]
[--reserved-lvs=*NAMES*]
[--node-parameters ndparams]
[{-m|--mac-prefix} mac-prefix]
[--master-netdev interface-name]
[--master-netmask netmask]
[--modify-etc-hosts {yes | no}]
[--use-external-mip-script {yes | no}]
[--hypervisor-state hvstate]
[--disk-state diskstate]
[--ipolicy-std-specs spec=*value* [,*spec*=*value*...]]
[--ipolicy-bounds-specs bounds_ispecs]
[--ipolicy-disk-templates template [,*template*...]]
[--ipolicy-spindle-ratio ratio]
[--ipolicy-vcpu-ratio ratio]
[--enabled-disk-templates template [,*template*...]]
[--drbd-usermode-helper helper]
[--file-storage-dir dir]
[--shared-file-storage-dir dir]
[--compression-tools [tool, [tool]]]
[--instance-communication-network network]
[--install-image image]
[--zeroing-image image]
[--user-shutdown {yes | no}]
[--enabled-data-collectors collectors]
[--data-collector-interval intervals]

Modify the options for the cluster.

The --vg-name, --enabled-hypervisors, -H (--hypervisor-parameters), -B (--backend-parameters), -D
(--disk-parameters), --nic-parameters, -C (--candidate-pool-size), --maintain-node-health, --prealloc-
wipe-disks, --uid-pool, --node-parameters, --mac-prefix, --master-netdev, --master-netmask, --use-exter‐
nal-mip-script, --drbd-usermode-helper, --file-storage-dir, --shared-file-storage-dir, --compression-
tools, and --enabled-disk-templates options are described in the init command. --master-netdev, --mas‐
ter-netmask, --use-external-mip-script, --drbd-usermode-helper, --file-storage-dir, --shared-file-stor‐
age-dir, --enabled-disk-templates, and --user-shutdown options are described in the init command.

The --modify-etc-hosts option is described by --no-etc-hosts in the init command.

The --hypervisor-state and --disk-state options are described in detail in ganeti(7).

The --max-running-jobs options allows to set limit on the number of jobs in non-finished jobs that are
not queued, i.e., the number of jobs that are in waiting or running state. The --max-tracked-jobs op‐
tions allows to set the limit on the tracked jobs. Normally, Ganeti will watch waiting and running jobs
by tracking their job file with inotify. If this limit is exceeded, however, Ganeti will back off and
only periodically pull for updates.

The --add-uids and --remove-uids options can be used to modify the user-id pool by adding/removing a list
of user-ids or user-id ranges.

The option --reserved-lvs specifies a list (comma-separated) of logical volume group names (regular ex‐
pressions) that will be ignored by the cluster verify operation. This is useful if the volume group used
for Ganeti is shared with the system for other uses. Note that it's not recommended to create and mark
as ignored logical volume names which match Ganeti's own name format (starting with UUID and then
.diskN), as this option only skips the verification, but not the actual use of the names given.

To remove all reserved logical volumes, pass in an empty argument to the option, as in --reserved-lvs= or
--reserved-lvs ''.

The -I (--default-iallocator) is described in the init command. To clear the default iallocator, just
pass an empty string ('').

The option --default-iallocator-params is described in the init command. To clear the default iallocator
parameters, just pass an empty string ('').

The --ipolicy-... options are described in the init command.

The --instance-communication-network enables instance communication by specifying the name of the Ganeti
network that should be used for instance communication. If the supplied network does not exist, Ganeti
will create a new network with the supplied name with the default parameters for instance communication.
If the supplied network exists, Ganeti will check its parameters and warn about unusual configurations,
but it will still use that network for instance communication.

The --enabled-data-collectors and --data-collector-interval options are to control the behavior of the
ganeti-mond(8). The first expects a list name=bool pairs to activate or deactivate the mentioned data
collector. The second option expects similar pairs of collector name and number of seconds specifying
the interval at which the collector shall be collected.

See gnt-cluster init for a description of --install-image and --zeroing-image.

See ganeti(7) for a description of --submit and other common options.

QUEUE
queue {drain | undrain | info}

Change job queue properties.

The drain option sets the drain flag on the job queue. No new jobs will be accepted, but jobs already in
the queue will be processed.

The undrain will unset the drain flag on the job queue. New jobs will be accepted.

The info option shows the properties of the job queue.

WATCHER
watcher {pause duration | continue | info}

Make the watcher pause or let it continue.

The pause option causes the watcher to pause for duration seconds.

The continue option will let the watcher continue.

The info option shows whether the watcher is currently paused.

REDIST-CONF
redist-conf [--submit] [--print-jobid]

This command forces a full push of configuration files from the master node to the other nodes in the
cluster. This is normally not needed, but can be run if the verify complains about configuration mis‐
matches.

See ganeti(7) for a description of --submit and other common options.

RENAME
rename [-f] {new-name}

Renames the cluster and in the process updates the master IP address to the one the new name resolves to.
At least one of either the name or the IP address must be different, otherwise the operation will be
aborted.

Note that since this command can be dangerous (especially when run over SSH), the command will require
confirmation unless run with the -f option.

RENEW-CRYPTO
renew-crypto [-f]
[--new-cluster-certificate] | [--new-node-certificates]
[--new-confd-hmac-key]
[--new-rapi-certificate] [--rapi-certificate rapi-cert]
[--new-spice-certificate | --spice-certificate spice-cert
--spice-ca-certificate spice-ca-cert]
[--new-ssh-keys] [--no-ssh-key-check]
[--new-cluster-domain-secret] [--cluster-domain-secret filename]
[--ssh-key-type type] | [--ssh-key-bits bits]

This command will stop all Ganeti daemons in the cluster and start them again once the new certificates
and keys are replicated. The option --new-confd-hmac-key can be used to regenerate the HMAC key used by
ganeti-confd(8).

The option --new-cluster-certificate will regenerate the cluster-internal server SSL certificate. The
option --new-node-certificates will generate new node SSL certificates for all nodes. Note that for the
regeneration of of the server SSL certificate will invoke a regeneration of the node certificates as
well, because node certificates are signed by the server certificate and thus have to be recreated and
signed by the new server certificate. Nodes which are offline during a renewal of the server or the node
certificates are not accessible anymore once they are marked as online again. To fix this, please readd
the node instead.

To generate a new self-signed RAPI certificate (used by ganeti-rapi(8)) specify --new-rapi-certificate.
If you want to use your own certificate, e.g. one signed by a certificate authority (CA), pass its file‐
name to --rapi-certificate.

To generate a new self-signed SPICE certificate, used for SPICE connections to the KVM hypervisor, speci‐
fy the --new-spice-certificate option. If you want to provide a certificate, pass its filename to
--spice-certificate and pass the signing CA certificate to --spice-ca-certificate.

The option --new-ssh-keys renews all SSH keys of all nodes and updates the authorized_keys files of all
nodes to contain only the (new) public keys of all master candidates. To avoid having to confirm the
fingerprint of each node use the --no-ssh-key-check option. Be aware of that this includes a security
risk as you omit verifying the machines' identities.

Finally --new-cluster-domain-secret generates a new, random cluster domain secret, and --cluster-domain-
secret reads the secret from a file. The cluster domain secret is used to sign information exchanged be‐
tween separate clusters via a third party.

The options --ssh-key-type and ssh-key-bits determine the properties of the disk types used. They are
described in more detail in the init option description.

REPAIR-DISK-SIZES
repair-disk-sizes [instance-name...]

This command checks that the recorded size of the given instance's disks matches the actual size and up‐
dates any mismatches found. This is needed if the Ganeti configuration is no longer consistent with re‐
ality, as it will impact some disk operations. If no arguments are given, all instances will be checked.
When exclusive storage is active, also spindles are updated.

Note that only active disks can be checked by this command; in case a disk cannot be activated it's ad‐
vised to use gnt-instance activate-disks --ignore-size ... to force activation without regard to the cur‐
rent size.

When all the disk sizes are consistent, the command will return no output. Otherwise it will log details
about the inconsistencies in the configuration.

UPGRADE
upgrade {--to version | --resume}

This command safely switches all nodes of the cluster to a new Ganeti version. It is a prerequisite that
the new version is already installed, albeit not activated, on all nodes; this requisite is checked be‐
fore any actions are done.

If called with the --resume option, any pending upgrade is continued, that was interrupted by a power
failure or similar on master. It will do nothing, if not run on the master node, or if no upgrade was in
progress.

VERIFY
verify [--no-nplus1-mem] [--node-group nodegroup]
[--error-codes] [{-I|--ignore-errors} errorcode]
[{-I|--ignore-errors} errorcode...]
[--verify-ssh-clutter]

Verify correctness of cluster configuration. This is safe with respect to running instances, and incurs
no downtime of the instances.

If the --no-nplus1-mem option is given, Ganeti won't check whether if it loses a node it can restart all
the instances on their secondaries (and report an error otherwise).

With --node-group, restrict the verification to those nodes and instances that live in the named group.
This will not verify global settings, but will allow to perform verification of a group while other oper‐
ations are ongoing in other groups.

The --error-codes option outputs each error in the following parseable format: ftype:ecode:edo‐
main:name:msg. These fields have the following meaning:

ftype Failure type. Can be WARNING or ERROR.

ecode Error code of the failure. See below for a list of error codes.

edomain
Can be cluster, node or instance.

name Contains the name of the item that is affected from the failure.

msg Contains a descriptive error message about the error

gnt-cluster verify will have a non-zero exit code if at least one of the failures that are found are of
type ERROR.

The --ignore-errors option can be used to change this behaviour, because it demotes the error represented
by the error code received as a parameter to a warning. The option must be repeated for each error that
should be ignored (e.g.: -I ENODEVERSION -I ENODEORPHANLV). The --error-codes option can be used to de‐
termine the error code of a given error.

Note that the verification of the configuration file consistency across master candidates can fail if
there are other concurrently running operations that modify the configuration.

The --verify-ssh-clutter option checks if more than one SSH key for the same '<user@hostname>' pair ex‐
ists in the 'authorized_keys' file. This is only checked for hostnames of nodes which belong to the
cluster. This check is optional, because there might be other systems manipulating the 'authorized_keys'
files, which would cause too many false positives otherwise.

List of error codes:

ECLUSTERCERT
Cluster certificate files verification failure

ECLUSTERCFG
Cluster configuration verification failure

ECLUSTERDANGLINGINST
Some instances have a non-existing primary node

ECLUSTERDANGLINGNODES
Some nodes belong to non-existing groups

ECLUSTERFILECHECK
Cluster configuration verification failure

EGROUPDIFFERENTPVSIZE
PVs in the group have different sizes

EINSTANCEBADNODE
Instance marked as running lives on an offline node

EINSTANCEDOWN
Instance not running on its primary node

EINSTANCEFAULTYDISK
Impossible to retrieve status for a disk

EINSTANCELAYOUT
Instance has multiple secondary nodes

EINSTANCEMISSINGCFGPARAMETER
A configuration parameter for an instance is missing

EINSTANCEMISSINGDISK
Missing volume on an instance

EINSTANCEPOLICY
Instance does not meet policy

EINSTANCESPLITGROUPS
Instance with primary and secondary nodes in different groups

EINSTANCEUNSUITABLENODE
Instance running on nodes that are not suitable for it

EINSTANCEWRONGNODE
Instance running on the wrong node

ENODEDRBD
Error parsing the DRBD status file

ENODEDRBDHELPER
Error caused by the DRBD helper

ENODEDRBDVERSION
DRBD version mismatch within a node group

ENODEFILECHECK
Error retrieving the checksum of the node files

ENODEFILESTORAGEPATHS
Detected bad file storage paths

ENODEFILESTORAGEPATHUNUSABLE
File storage path unusable

ENODEGLUSTERSTORAGEPATHUNUSABLE
Gluster storage path unusable

ENODEHOOKS
Communication failure in hooks execution

ENODEHV
Hypervisor parameters verification failure

ENODELVM
LVM-related node error

ENODEN1
Not enough memory to accommodate instance failovers

ENODENET
Network-related node error

ENODEOOBPATH
Invalid Out Of Band path

ENODEORPHANINSTANCE
Unknown intance running on a node

ENODEORPHANLV
Unknown LVM logical volume

ENODEOS
OS-related node error

ENODERPC
Error during connection to the primary node of an instance

ENODESETUP
Node setup error

ENODESHAREDFILESTORAGEPATHUNUSABLE
Shared file storage path unusable

ENODESSH
SSH-related node error

ENODETIME
Node returned invalid time

ENODEUSERSCRIPTS
User scripts not present or not executable

ENODEVERSION
Protocol version mismatch or Ganeti version mismatch

VERIFY-DISKS
verify-disks [--node-group nodegroup]

The command checks which instances have degraded DRBD disks and activates the disks of those instances.

With --node-group, restrict the verification to those nodes and instances that live in the named group.

This command is run from the ganeti-watcher tool, which also has a different, complementary algorithm for
doing this check. Together, these two should ensure that DRBD disks are kept consistent.

VERSION
version

Show the cluster version.

Tags
ADD-TAGS
add-tags [--from file] {tag...}

Add tags to the cluster. If any of the tags contains invalid characters, the entire operation will
abort.

If the --from option is given, the list of tags will be extended with the contents of that file (each
line becomes a tag). In this case, there is not need to pass tags on the command line (if you do, both
sources will be used). A file name of - will be interpreted as stdin.

LIST-TAGS
list-tags

List the tags of the cluster.

REMOVE-TAGS
remove-tags [--from file] {tag...}

Remove tags from the cluster. If any of the tags are not existing on the cluster, the entire operation
will abort.

If the --from option is given, the list of tags to be removed will be extended with the contents of that
file (each line becomes a tag). In this case, there is not need to pass tags on the command line (if you
do, tags from both sources will be removed). A file name of - will be interpreted as stdin.

SEARCH-TAGS
search-tags {pattern}

Searches the tags on all objects in the cluster (the cluster itself, the nodes and the instances) for a
given pattern. The pattern is interpreted as a regular expression and a search will be done on it (i.e.
the given pattern is not anchored to the beginning of the string; if you want that, prefix the pattern
with ^).

If no tags are matching the pattern, the exit code of the command will be one. If there is at least one
match, the exit code will be zero. Each match is listed on one line, the object and the tag separated by
a space. The cluster will be listed as /cluster, a node will be listed as /nodes/name, and an instance
as /instances/name. Example:

# gnt-cluster search-tags time
/cluster ctime:2007-09-01
/nodes/node1.example.com mtime:2007-10-04

REPORTING BUGS

       Report  bugs  to the project's issue tracker (https://github.com/ganeti/ganeti/issues) or contact the de‐
       velopers using the Ganeti mailing list.

COPYRIGHT

Redistribution and use in source and binary forms, with or without modification, are permitted provided
that the following conditions are met:

1. Redistributions of source code must retain the above copyright notice, this list of conditions and
the following disclaimer.

2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and
the following disclaimer in the documentation and/or other materials provided with the distribution.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED
WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR
ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMIT‐
ED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUP‐
TION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (IN‐
CLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
THE POSSIBILITY OF SUCH DAMAGE.