Description

FSS : 

Flexible Storage Sharing (FSS) enables network sharing of local storage, cluster wide. The local storage can be in the form of Direct Attached Storage (DAS) or internal disk drives. Network shared storage is enabled by using a network interconnect between the nodes of a VCS cluster.

FSS is supported on clusters containing up to 8 nodes with CVM protocol versions 140 and above.

The local disks become visible to the cluster nodes when we start the CVM and are unplugged when we terminate the CVM port (v). This port is a part of the CVM service group and is initiated and terminated with the CVM Online/Offline.

The active systems in the cluster treat this CVM offline on one node as a disk disconnect and flag it for recovery. 

After the system has been restored, the volumes will be automatically recovered in the background. It may take time depending on their size and number of Diskgroups. Except for the VVR SRL volumes, fast resync will help it sync quickly. 

Note: SRL fast resync feature is enabled from 8.0.2 onward.

There is a potential for cluster inconsistency if the other systems are rebooted before the recovery is completed.

Veritas recommends the changes below to overcome the recovery in scheduled reboots.

One time Change:

1) Enable CVMDeportOnOffline for all the CVMVolDG resources.

Example:
# haconf -makerw
# hares -modify CVMDeportOnOffline  1
# haconf -dump -makero 

2) Install the latest VRTScavf package.

3) Keep the CVMVolDG outside of CVM Service groups. 

Steps to follow for the scheduled cluster reboots.

1) Offline all the service groups in all the nodes except CVM.
    Note: Applications, CFSMounts will be offline and shared diskgroups will be deported during this time. 
2) Run hastop -all 
3) Reboot the cluster nodes. 

Note: While rebooting the nodes one after another, ensure the plexes are recovered, prior to rebooting the other node.