Cluster Commands

 

Status commands

 

Checking the cluster quorum status: Displays everything including DID information (but not DID status. That requires cldev status) transport status, cluster status, quorum, devices, node status, resource types registered, resource groups configured.

 

# cluster show                       

 

Checking quorum votes and quorum device only:

 

# clq(uorum) status

 

List all global devices

 

# cldev list -v

 

List cluster transport configuration and status only:

 

# clintr status

 

List disk group status and show preferenced hosts and failback status:

 

# cldg show

 

Show disk groups and current primary node

 

# cldg status

 

Show all registered resource types and properties:

 

# clrt show

 

Show all configured resource groups, and resources in the group, or a specified resource group and its properties

 

# crlg show <resource group>

 

Show resource group status

 

# clrg status

 

Show all configured resources

 

# clrs list -v

 

Show status of configured resources

 

# clrs status

 

 

Show which disks are in a VxVM disk group

 

# vxprint (on the node that is primary for the disk group)

 

Show which disks are in all SVM disk group, along with metadevices and soft partitions

 

# metastat -a

 

Management commands

 

Shut down cluster and bring all nodes to the PROM:

 

#cluster shutdown -y -g0

 

Register changes to a disk group

 

# cldg sync <disk group>

 

Bring a resource group online

 

# clrg online -M <resource group name>

 

Configuration commands

 

Perform initial configuration

 

# scinstall

 

Perform later installation, including selecting or changing a private network address or configuring a cluster or adding or removing a node:

 

# clsetup

 

Register a resource type other than SharedAddress or LogicalHostname

 

# clrt register <resource type>

 

Create an instance of a resource group and set properties

 

# clrg create -n <node1,node2…> -p <property type>=<property of group> <other properties> <name of group>

 

Create and configure an instance of any resource other a logical hostname or shared address resource and add to a resource group:

 

clrs create -t <resource type> -g <resourcee group> -p <property name=value> <other properties> <resource name>

 

Create an instance of a logical host name resource

 

#clrslh create -g <resource group> <logical hostname>

 

Create an instance of a shared address resource

 

# clrssa create -g <resource group> <shared address name>

 

 

Undo commands

 

Offline a resource group

 

# clrg offline <resource group>

 

Disabling a resource

 

# clrs disable <resource name>

 

Removing a resource

 

# clrs delete <resource name>

 

Unconfiguring a resource group: This can only be done when all included resources have been deleted, and when all dependent resource groups have been deleted. In order: 1. offline the resource group 2. disable all resources in the group 3. delete all resources in the group 4. delete the resource group.

 

# clrg delete <resource group>

 

Removing a SVM disk group:

 

One both nodes, check that /etc/nsswitch.conf has the entry

 

ipnodes: files

 

You cannot have ANYTHING else on this line.

 

Then on the node that is primary for the disk group:

 

(if none is primary then:

 

# cldg switch -n <node-to-work-from>  <disksetname>

 

i.e.

 

# cldg switch -n rams nfsds)

 

Locate the names of the disks in the diskset:

 

# metaset -s nfsds

 

Remove those disks from the diskset:

 

# metaset -s nfsds -d /dev/did/dsk/d6

 

When the last disk is removed, use the -f option to force the removal of the disk:

 

# metaset -s nfsds -f -d /dev/did/dsk/d7

 

Delete the disk set from all nodes:

 

# metaset -s nfsds -f -d -h <node1 node2…>

 

i.e.

 

# metaset -s nfsds -f -d -h rams falcons