Wednesday, February 19, 2014

LUN Migration on Sun Solaris 10 with Sun Cluster 3.3 -Replacing a Quorum device as well.



/dev/did/dsk/d9s2 is a quorum device and also db runs on it. It needs to be replaced with the New LUN.
The new lun has to be added to both the nodes and we need to ensure the same.

1. Check the old lun on the system.
echo |format ; powermt display dev=all; zpool status GRID

2.Check the new lun and label it.
devfsadm -Cv ; echo |format

We can run the below commands to both the nodes.
cldevice populate - update the global device namespace.
cldevice list -v - Verify that a device ID (DID) has been assigned to the disk drive.

3. Check the status of the pool
zpool status GRID

4. Attach the new lun
zpool attach -f GRID <device> <new-device>
#zpool attach -f GRID /dev/did/dsk/d9s2 <new-device> (New Device will be a DID device)


5. Detach the old lun after the mirroing gets completed. (Important -Ensure that it has completed )
zpool detach GRID /dev/did/dsk/d9s2 (OLD LUN)

6. Add a Quorum device.
scsetup , Answer the questions... then verify the quorum device by running cluster status (We should have the new device in place as a quorum disk)

VERY Important: After ensuring that the Quorum disk is in place in cluster status output.
# luxadm remove_device -F /dev/rdsk/cNtXdYsZ
# devfsadm -C
# cldevice clear

Test the Failover by running.
clrg switch -n Nodename RG

Monday, January 27, 2014

Sun Cluster 3.3 Commands

ome Sun Cluster 3.3 commands 


1. Cluster


The command to use is cluster

# cluster status: the status of the cluster nodes.
# cluster show: Displays the cluster parameters.
# cluster shutdown: shuts down the cluster.
# cluster list: list of cluster nodes configured

2. Management of a resource group


The command to use is CLRG

# CLRG list: list of resource groups.
# CLRG list RG: RG group list.
# CLRG status: the status of resource groups.
# CLRG status RG: RG group status.
# CLRG remaster +: reinstate the resources groups to their preferred nodes.
# CLRG switch-n NODE RG: RG group moves to the node NODE.
# CLRG evacuate-n NODE RG: Move all resources NODE node to the other node.
# CLRG offline RG: stop all resources in the group.
# CLRG online RG: starts all resources in the group.
# RG CLRG restart: restarts all the resources of the group.
# CLRG manage RG: Password group under the control of the cluster.
# CLRG unmanage RG: pass the group beyond the control of the cluster.
# CLRG show RG-v: displays the properties of RG Group.

3. Resource Management


The command to use is clrs

# clrs list: list all resources.
# clrs list: list all the resources of the group RG.
# clrs status: the status of all resources.
# RS clrs status: the status of the RS resources.
# clrs RG-g status: the status of all resources group RG.
# clrs disable RS: RS resource stops.
# clrs enable RS: RS resource starts.
# clrs monitor RS: starts monitoring the resource RS.
# clrs unmonitor RS: stop monitoring the resource RS.
# clrs clear-f STOP_FAILED RS: removes the flag STOP_FAILED RS resource.
# clrs show-v RS: displays the properties of the resource RS.
# clrs set-p property = value RS: Changes the value of the property by value proprité

To add a value to a property already defined, it is possible to use + = instead of =

4. Management Logical Hosts


The command to use is clrslh

# clrslh list: list all logical hosts.
# clrslh status: the status of all logical hosts.
# clrslh disable LH: LH stop resource.
# clrslh enable LH: LH starts resources.
# clrslh monitor LH: LH starts monitoring.
# clrslh unmonitor LH: LH stops monitoring.
# LH clrslh clear: removes the error flag of LH.
# clrslh show LH: displays the properties of LH