1 gnt-node(8) Ganeti | Version @GANETI_VERSION@
2 =============================================
7 gnt-node - Node administration
12 **gnt-node** {command} [arguments...]
17 The **gnt-node** is used for managing the (physical) nodes in the
26 | **add** [--readd] [-s *secondary\_ip*] [-g *nodegroup*]
27 | [--master-capable=``yes|no``] [--vm-capable=``yes|no``]
28 | [--node-parameters *ndparams*]
31 Adds the given node to the cluster.
33 This command is used to join a new node to the cluster. You will
34 have to provide the password for root of the node to be able to add
35 the node in the cluster. The command needs to be run on the Ganeti
38 Note that the command is potentially destructive, as it will
39 forcibly join the specified host the cluster, not paying attention
40 to its current status (it could be already in a cluster, etc.)
42 The ``-s`` is used in dual-home clusters and specifies the new node's
43 IP in the secondary network. See the discussion in **gnt-cluster**(8)
46 In case you're readding a node after hardware failure, you can use
47 the ``--readd`` parameter. In this case, you don't need to pass the
48 secondary IP again, it will reused from the cluster. Also, the
49 drained and offline flags of the node will be cleared before
52 The ``-g`` is used to add the new node into a specific node group,
53 specified by UUID or name. If only one node group exists you can
54 skip this option, otherwise it's mandatory.
56 The ``vm_capable``, ``master_capable`` and ``ndparams`` options are
57 described in **ganeti**(7), and are used to set the properties of the
62 # gnt-node add node5.example.com
63 # gnt-node add -s 192.0.2.5 node5.example.com
64 # gnt-node add -g group2 -s 192.0.2.9 node9.group2.example.com
70 **add-tags** [--from *file*] {*nodename*} {*tag*...}
72 Add tags to the given node. If any of the tags contains invalid
73 characters, the entire operation will abort.
75 If the ``--from`` option is given, the list of tags will be
76 extended with the contents of that file (each line becomes a tag).
77 In this case, there is not need to pass tags on the command line
78 (if you do, both sources will be used). A file name of - will be
84 **evacuate** [-f] [--early-release] [--iallocator *NAME* \|
85 --new-secondary *destination\_node*] {*node*...}
87 This command will move all secondary instances away from the given
88 node(s). It works only for instances having a drbd disk template.
90 The new location for the instances can be specified in two ways:
92 - as a single node for all instances, via the ``--new-secondary``
95 - or via the ``--iallocator`` option, giving a script name as
96 parameter, so each instance will be in turn placed on the (per the
100 The ``--early-release`` changes the code so that the old storage on
101 node being evacuated is removed early (before the resync is
102 completed) and the internal Ganeti locks are also released for both
103 the current secondary and the new secondary, thus allowing more
104 parallelism in the cluster operation. This should be used only when
105 recovering from a disk failure on the current secondary (thus the
106 old storage is already broken) or when the storage on the primary
107 node is known to be fine (thus we won't need the old storage for
112 # gnt-node evacuate -I dumb node3.example.com
118 **failover** [-f] [--ignore-consistency] {*node*}
120 This command will fail over all instances having the given node as
121 primary to their secondary nodes. This works only for instances having
122 a drbd disk template.
124 Normally the failover will check the consistency of the disks before
125 failing over the instance. If you are trying to migrate instances off
126 a dead node, this will fail. Use the ``--ignore-consistency`` option
131 # gnt-node failover node1.example.com
139 Show detailed information about the nodes in the cluster. If you
140 don't give any arguments, all nodes will be shows, otherwise the
141 output will be restricted to the given names.
147 | [--no-headers] [--separator=*SEPARATOR*]
148 | [--units=*UNITS*] [-o *[+]FIELD,...*]
152 Lists the nodes in the cluster.
154 The ``--no-headers`` option will skip the initial header line. The
155 ``--separator`` option takes an argument which denotes what will be
156 used between the output fields. Both these options are to help
159 The units used to display the numeric values in the output varies,
160 depending on the options given. By default, the values will be
161 formatted in the most appropriate unit. If the ``--separator``
162 option is given, then the values are shown in mebibytes to allow
163 parsing by scripts. In both cases, the ``--units`` option can be
164 used to enforce a given output unit.
166 By default, the query of nodes will be done in parallel with any
167 running jobs. This might give inconsistent results for the free
168 disk/memory. The ``--sync`` can be used to grab locks for all the
169 nodes and ensure consistent view of the cluster (but this might
170 stall the query for a long time).
172 Passing the ``--roman`` option gnt-node list will try to output
173 some of its fields in a latin-friendly way. This is not the default
174 for backwards compatibility.
176 The ``-o`` option takes a comma-separated list of output fields.
177 The available fields and their meaning are:
185 the number of instances having this node as primary
188 the list of instances having this node as primary, comma separated
191 the number of instances having this node as a secondary node
194 the list of instances having this node as a secondary node, comma
198 the primary ip of this node (used for cluster communication)
201 the secondary ip of this node (used for data replication in dual-ip
202 clusters, see gnt-cluster(8)
205 total disk space in the volume group used for instance disk
209 available disk space in the volume group
212 total memory on the physical node
215 the memory used by the node itself
218 memory available for instance allocations
221 the node bootid value; this is a linux specific feature that
222 assigns a new UUID to the node at each boot and can be use to
223 detect node reboots (by tracking changes in this value)
226 comma-separated list of the node's tags
229 the so called 'serial number' of the node; this is a numeric field
230 that is incremented each time the node is modified, and it can be
231 used to detect modifications
234 the creation time of the node; note that this field contains spaces
235 and as such it's harder to parse
237 if this attribute is not present (e.g. when upgrading from older
238 versions), then "N/A" will be shown instead
241 the last modification time of the node; note that this field
242 contains spaces and as such it's harder to parse
244 if this attribute is not present (e.g. when upgrading from older
245 versions), then "N/A" will be shown instead
248 Show the UUID of the node (generated automatically by Ganeti)
251 the toal number of logical processors
254 the number of NUMA domains on the node, if the hypervisor can
255 export this information
258 the number of physical CPU sockets, if the hypervisor can export
262 whether the node is a master candidate or not
265 whether the node is drained or not; the cluster still communicates
266 with drained nodes but excludes them from allocation operations
269 whether the node is offline or not; if offline, the cluster does
270 not communicate with offline nodes; useful for nodes that are not
271 reachable in order to avoid delays
274 A condensed version of the node flags; this field will output a
275 one-character field, with the following possible values:
277 - *M* for the master node
279 - *C* for a master candidate
281 - *R* for a regular node
283 - *D* for a drained node
285 - *O* for an offline node
288 whether the node can become a master candidate
291 whether the node can host instances
294 the name of the node's group, if known (the query is done without
295 locking, so data consistency is not guaranteed)
298 the UUID of the node's group
301 If the value of the option starts with the character ``+``, the new
302 fields will be added to the default list. This allows to quickly
303 see the default list plus a few other fields, instead of retyping
304 the entire list of fields.
306 Note that some of this fields are known from the configuration of
307 the cluster (e.g. name, pinst, sinst, pip, sip and thus the master
308 does not need to contact the node for this data (making the listing
309 fast if only fields from this set are selected), whereas the other
310 fields are "live" fields and we need to make a query to the cluster
313 Depending on the virtualization type and implementation details,
314 the mtotal, mnode and mfree may have slighly varying meanings. For
315 example, some solutions share the node memory with the pool of
316 memory used for instances (KVM), whereas others have separate
317 memory for the node and for the instances (Xen).
319 If no node names are given, then all nodes are queried. Otherwise,
320 only the given nodes will be listed.
325 **list-tags** {*nodename*}
327 List the tags of the given node.
332 **migrate** [-f] [--non-live] [--migration-mode=live\|non-live]
335 This command will migrate all instances having the given node as
336 primary to their secondary nodes. This works only for instances
337 having a drbd disk template.
339 As for the **gnt-instance migrate** command, the options
340 ``--no-live`` and ``--migration-mode`` can be given to influence
345 # gnt-node migrate node1.example.com
351 | **modify** [-f] [--submit]
352 | [--master-candidate=``yes|no``] [--drained=``yes|no``] [--offline=``yes|no``]
353 | [--master-capable=``yes|no``] [--vm-capable=``yes|no``] [--auto-promote]
354 | [-s *secondary_ip*]
355 | [--node-parameters *ndparams*]
358 This command changes the role of the node. Each options takes
359 either a literal yes or no, and only one option should be given as
360 yes. The meaning of the roles and flags are described in the
361 manpage **ganeti**(7).
363 In case a node is demoted from the master candidate role, the
364 operation will be refused unless you pass the ``--auto-promote``
365 option. This option will cause the operation to lock all cluster nodes
366 (thus it will not be able to run in parallel with most other jobs),
367 but it allows automated maintenance of the cluster candidate pool. If
368 locking all cluster node is too expensive, another option is to
369 promote manually another node to master candidate before demoting the
372 Example (setting a node offline, which will demote it from master
373 candidate role if is in that role)::
375 # gnt-node modify --offline=yes node1.example.com
377 The ``-s`` can be used to change the node's secondary ip. No drbd
378 instances can be running on the node, while this operation is
381 Example (setting the node back to online and master candidate)::
383 # gnt-node modify --offline=no --master-candidate=yes node1.example.com
389 **remove** {*nodename*}
391 Removes a node from the cluster. Instances must be removed or
392 migrated to another cluster before.
396 # gnt-node remove node5.example.com
402 **remove-tags** [--from *file*] {*nodename*} {*tag*...}
404 Remove tags from the given node. If any of the tags are not
405 existing on the node, the entire operation will abort.
407 If the ``--from`` option is given, the list of tags to be removed will
408 be extended with the contents of that file (each line becomes a tag).
409 In this case, there is not need to pass tags on the command line (if
410 you do, tags from both sources will be removed). A file name of - will
411 be interpreted as stdin.
416 | **volumes** [--no-headers] [--human-readable]
417 | [--separator=*SEPARATOR*] [--output=*FIELDS*]
420 Lists all logical volumes and their physical disks from the node(s)
423 The ``--no-headers`` option will skip the initial header line. The
424 ``--separator`` option takes an argument which denotes what will be
425 used between the output fields. Both these options are to help
428 The units used to display the numeric values in the output varies,
429 depending on the options given. By default, the values will be
430 formatted in the most appropriate unit. If the ``--separator``
431 option is given, then the values are shown in mebibytes to allow
432 parsing by scripts. In both cases, the ``--units`` option can be
433 used to enforce a given output unit.
435 The ``-o`` option takes a comma-separated list of output fields.
436 The available fields and their meaning are:
439 the node name on which the volume exists
442 the physical drive (on which the LVM physical volume lives)
445 the volume group name
448 the logical volume name
451 the logical volume size
454 The name of the instance to which this volume belongs, or (in case
455 it's an orphan volume) the character "-"
460 # gnt-node volumes node5.example.com
461 Node PhysDev VG Name Size Instance
462 node1.example.com /dev/hdc1 xenvg instance1.example.com-sda_11000.meta 128 instance1.example.com
463 node1.example.com /dev/hdc1 xenvg instance1.example.com-sda_11001.data 256 instance1.example.com
469 | **list-storage** [--no-headers] [--human-readable]
470 | [--separator=*SEPARATOR*] [--storage-type=*STORAGE\_TYPE*]
471 | [--output=*FIELDS*]
474 Lists the available storage units and their details for the given
477 The ``--no-headers`` option will skip the initial header line. The
478 ``--separator`` option takes an argument which denotes what will be
479 used between the output fields. Both these options are to help
482 The units used to display the numeric values in the output varies,
483 depending on the options given. By default, the values will be
484 formatted in the most appropriate unit. If the ``--separator``
485 option is given, then the values are shown in mebibytes to allow
486 parsing by scripts. In both cases, the ``--units`` option can be
487 used to enforce a given output unit.
489 The ``--storage-type`` option can be used to choose a storage unit
490 type. Possible choices are lvm-pv, lvm-vg or file.
492 The ``-o`` option takes a comma-separated list of output fields.
493 The available fields and their meaning are:
496 the node name on which the volume exists
499 the type of the storage unit (currently just what is passed in via
503 the path/identifier of the storage unit
506 total size of the unit; for the file type see a note below
509 used space in the unit; for the file type see a note below
515 whether we the unit is available for allocation (only lvm-pv can
516 change this setting, the other types always report true)
519 Note that for the "file" type, the total disk space might not equal
520 to the sum of used and free, due to the method Ganeti uses to
521 compute each of them. The total and free values are computed as the
522 total and free space values for the filesystem to which the
523 directory belongs, but the used space is computed from the used
524 space under that directory *only*, which might not be necessarily
525 the root of the filesystem, and as such there could be files
526 outside the file storage directory using disk space and causing a
527 mismatch in the values.
531 node1# gnt-node list-storage node2
532 Node Type Name Size Used Free Allocatable
533 node2 lvm-pv /dev/sda7 673.8G 1.5G 672.3G Y
534 node2 lvm-pv /dev/sdb1 698.6G 0M 698.6G Y
540 **modify-storage** [``--allocatable=yes|no``]
541 {*node*} {*storage-type*} {*volume-name*}
543 Modifies storage volumes on a node. Only LVM physical volumes can
544 be modified at the moment. They have a storage type of "lvm-pv".
548 # gnt-node modify-storage --allocatable no node5.example.com lvm-pv /dev/sdb1
554 **repair-storage** [--ignore-consistency] {*node*} {*storage-type*}
557 Repairs a storage volume on a node. Only LVM volume groups can be
558 repaired at this time. They have the storage type "lvm-vg".
560 On LVM volume groups, **repair-storage** runs "vgreduce
565 **Caution:** Running this command can lead to data loss. Use it with
568 The ``--ignore-consistency`` option will ignore any inconsistent
569 disks (on the nodes paired with this one). Use of this option is
570 most likely to lead to data-loss.
574 # gnt-node repair-storage node5.example.com lvm-vg xenvg
580 **powercycle** [``--yes``] [``--force``] {*node*}
582 This commands (tries to) forcefully reboot a node. It is a command
583 that can be used if the node environemnt is broken, such that the
584 admin can no longer login over ssh, but the Ganeti node daemon is
587 Note that this command is not guaranteed to work; it depends on the
588 hypervisor how effective is the reboot attempt. For Linux, this
589 command require that the kernel option CONFIG\_MAGIC\_SYSRQ is
592 The ``--yes`` option can be used to skip confirmation, while the
593 ``--force`` option is needed if the target node is the master