mmvdisk command
Manages IBM Spectrum Scale RAID node classes, servers, recovery groups, vdisk sets, file systems, pdisks, and vdisks.
Synopsis
mmvdisk nodeclass Verb [Parameters]
mmvdisk server Verb [Parameters]
mmvdisk recoverygroup Verb [Parameters]
mmvdisk vdiskset Verb [Parameters]
mmvdisk filesystem Verb [Parameters]
mmvdisk pdisk Verb [Parameters]
mmvdisk vdisk Verb [Parameters]
Availability
Available on all IBM Spectrum Scale editions.
Description
The mmvdisk command is an integrated command suite and management methodology for IBM Spectrum Scale RAID. It greatly simplifies IBM Spectrum Scale RAID administration, and encourages and enforces consistent best practices with regard to server, recovery group, vdisk NSD, and file system configuration.
- The paired recovery groups of Elastic Storage Server, where a pair of servers divide a common set of disk enclosures into a pair of recovery groups, with each server taking exclusive primary responsibility for one recovery group of the pair.
- The scale-out recovery groups of IBM Spectrum® Scale Erasure Code Edition, where a set of 4 to 32 identically equipped servers each take an equal share of responsibility in a single collective recovery group.
- The shared recovery groups of IBM Elastic Storage® System 3000, where a single recovery group is defined on a common disk enclosure shared by two servers.
New IBM Spectrum Scale clusters can use mmvdisk from the beginning to manage any IBM Spectrum Scale RAID recovery groups. The server and recovery group pairs of existing Elastic Storage Server installations can be converted to enable mmvdisk management. All existing non-mmvdisk recovery groups in a cluster must be converted to mmvdisk management before mmvdisk can be used to create new recovery groups in the cluster. The scale-out recovery groups of IBM Spectrum Scale Erasure Code Edition and the shared recovery groups of IBM Elastic Storage System 3000 are required to be managed by mmvdisk. If a cluster containing non-mmvdisk paired recovery groups wishes to add scale-out or shared recovery groups, the existing paired recovery groups must first be converted to mmvdisk management.
The mmvdisk command and the legacy IBM Spectrum Scale RAID administration commands are incompatible. When all recovery groups in a cluster are managed by the mmvdisk command, the legacy IBM Spectrum Scale RAID commands can not be used to create recovery groups or vdisks. Other legacy IBM Spectrum Scale RAID commands are similarly restricted. For example, if a recovery group is managed by mmvdisk, pdisk replacement must be performed using mmvdisk pdisk replace, since mmchcarrier will refuse to run against an mmvdisk recovery group. Conversely, mmvdisk pdisk replace will refuse to run against an existing recovery group that has not been converted to mmvdisk management, and mmchcarrier must be used with the unconverted recovery group.
The mmvdisk command structures IBM Spectrum Scale RAID around node classes, servers, recovery groups, vdisk sets, and file systems.
An mmvdisk node class contains the set of servers associated with a recovery group or recovery group pair. The IBM Spectrum Scale RAID configuration parameters for the servers are maintained under the node class. The node class also acts as a building block identifier for common server and disk capacity in the associated recovery group or recovery group pair.
The vdisks in a recovery group are sub-divided into log groups. A log group is a collection of vdisks that share a common RAID transaction log. In the case of a recovery group pair, there is just one log group per recovery group, and in this case a paired recovery group is trivially equivalent to its one and only log group. Each of the two paired recovery group servers has exclusive primary responsibility for the vdisks of one trivial log group. In the case of scale-out or shared recovery groups, the recovery group is sub-divided into multiple equal log groups, and each server is responsible for the vdisks of two log groups. Log groups act to equitably sub-divide and balance vdisks among the servers of a recovery group or recovery group pair.
The management of potentially large numbers of log group vdisks is simplified by treating the file system vdisk NSDs in the log groups of recovery groups collectively as members of vdisk sets. A vdisk set is a collection of vdisk NSDs with identical attributes from one or more recovery groups. Each log group of a recovery group included in a vdisk set contributes one member vdisk NSD. This means that when a vdisk set is defined using paired recovery groups, there will be one member vdisk NSD from each paired recovery group. When a vdisk set is defined using scale-out or shared recovery groups, there will be one member vdisk NSD from each log group of the recovery group, which is equivalent to two vdisk NSDs per server of the scale-out or shared recovery group.
A vdisk set must be managed as a unit: All member vdisk NSDs of a vdisk set must belong to the same mmvdisk file system. Defining a vdisk set across multiple recovery groups helps to ensure that a file system is balanced using identical vdisk NSDs from the recovery groups in the vdisk set.
An mmvdisk file system is an IBM Spectrum Scale RAID file system constructed from one or more vdisk sets. It is possible for an mmvdisk file system to contain non-vdisk NSDs, but the vdisk NSDs must all be members of vdisk sets, and non-vdisk and vdisk NSDs can not reside in the same file system storage pool.
In addition to the management structure provided by node classes, servers, recovery groups, vdisk sets, and file systems, mmvdisk also provides an interface to list and replace recovery group pdisks and to list individual vdisks.
Each mmvdisk sub-command has its own manual page with command syntax, information, and examples.
The mmvdisk nodeclass command manages mmvdisk recovery group server node classes.
The mmvdisk server command manages mmvdisk recovery group servers.
The mmvdisk recoverygroup command manages mmvdisk recovery groups.
The mmvdisk vdiskset command manages mmvdisk vdisk sets.
The mmvdisk filesystem command manages mmvdisk file systems.
The mmvdisk pdisk command manages mmvdisk recovery group pdisks.
The mmvdisk vdisk command lists individual recovery group vdisks.
Parameters
An mmvdisk command has the form mmvdisk Noun Verb [Parameters]. For example, mmvdisk nodeclass list, without any parameters, will list all the mmvdisk node classes in the cluster.
- nodeclass: Recovery group server node classes
- server: Recovery group servers within node classes
- recoverygroup: Recovery groups
- vdiskset: Vdisk sets across recovery groups
- filesystem: IBM Spectrum Scale file systems built from vdisk sets
- pdisk: Recovery group pdisks, for listing and maintenance
- vdisk: Individual vdisks for listing only
The verbs are functions to be performed specific to a noun, and will always include list. Other possible verbs are: create, add, delete, and change.
After the noun and the verb, additional parameters might be required or optional, can occur in any order, and are always introduced with option flags. The vast majority of option flags are of the form --option-name and might or might not take following arguments.
- -N
- This behaves like the traditional node specification.
- -v
- The traditional "verify" option to make sure objects are not already in use.
- -p
- The traditional "permanently damaged" option to delete inaccessible objects.
- -L
- The traditional "very long" detailed listing.
- -Y
- Provides colon-delimited raw output.
- mmvdisk nc can be used in place of mmvdisk nodeclass.
- mmvdisk rg can be used in place of mmvdisk recoverygroup.
- mmvdisk vs can be used in place of mmvdisk vdiskset.
- mmvdisk fs can be used in place of mmvdisk filesystem.
- --nc can be used in place of --node-class.
- --rg can be used in place of --recovery-group.
- --da can be used in place of --declustered-array.
- --vs can be used in place of --vdisk-set.
- --fs can be used in place of --file-system.
- --bs can be used in place of --block-size.
- --ss can be used in place of --set-size.
- --usage can be used in place of --nsd-usage.
- --pool can be used in place of --storage-pool.
- --ft can be used in place of --fault-tolerance.
- --lg can be used in place of --log-group.
Depending on the context, when it is possible to list multiple comma-separated objects as the argument to an option flag, the keyword all can often be used as shorthand for listing all objects of a type. For example, --vs all can be used to specify all vdisk sets. The individual command usage statements will indicate this with a --vdisk-set {all | VsName [,VsName...]} syntax diagram.
When a command will delete (or undefine, in the case of vdisk sets) an object permanently, mmvdisk will warn the user by prompting for confirmation. In these cases, the prompt can be bypassed and answered in the affirmative with the --confirm option.
mmvdisk will also warn the user and prompt for confirmation before performing potentially long-running commands. There is no option to bypass the prompt in the case of long-running commands.
Exit status
- 0
- Successful completion.
- nonzero
- A failure has occurred.
Security
You must have root authority to run the mmvdisk command.
The node on which the command is issued must be able to execute remote shell commands on any other node in the cluster without the use of a password and without producing any extraneous messages.
Examples
- To see all mmvdisk node classes in the
cluster:
mmvdisk nodeclass list
The system displays output similar to the following:
node class recovery groups ---------- --------------- ESS01 ESS01L, ESS01R ESS02 ESS02L, ESS02R
- To see all recovery group servers in the
cluster:
mmvdisk server list
The system displays output similar to the following:
node number server node class recovery groups remarks ------ -------------------------------- ---------- --------------- ------- 1 ess01io1.gpfs.net ESS01 ESS01L, ESS01R 2 ess01io2.gpfs.net ESS01 ESS01L, ESS01R 3 ess02io1.gpfs.net ESS02 ESS02L, ESS02R 4 ess02io2.gpfs.net ESS02 ESS02L, ESS02R
- To see all recovery groups in the
cluster:
mmvdisk recoverygroup list
The system displays output similar to the following:
needs user recovery group active current or master server service vdisks remarks -------------- ------- -------------------------------- ------- ------ ------- ESS01L yes ess01io1.gpfs.net no 2 ESS01R yes ess01io2.gpfs.net no 2 ESS02L yes ess02io1.gpfs.net no 2 ESS02R yes ess02io2.gpfs.net no 2
- To see all vdisk sets in the cluster:
mmvdisk vdiskset list
The system displays output similar to the following:
vdisk set created file system recovery groups ---------------- ------- ----------- --------------- VS1 yes FS1 ESS01L, ESS01R, ESS02L, ESS02R VS2 yes FS2 ESS01L, ESS01R, ESS02L, ESS02R
- To see all file systems in the cluster:
mmvdisk filesystem list
The system displays output similar to the following:
file system vdisk sets ----------- ---------- FS1 VS1 FS2 VS2
See also
See also the following IBM Spectrum Scale RAID: Administration topics:
Location
/usr/lpp/mmfs/bin