This content has been marked as final. Show 6 replies
I really would go with the "Oracle Solaris Cluster Concepts Guide" (http://docs.oracle.com/cd/E18728_01/html/821-2682/index.html) first. It gives you a clear understanding of how the cluster is supposed to work.
There are a couple of how-to's on http://www.oracle.com/technetwork/server-storage/solaris-cluster/documentation/cluster-how-to-1389544.html . I think they are helpful too.
To start deploying HA services, I would go with
a) configuring an HA file system
b) configuring an HA IP address
c) configuring a simple HA service based on GDS (see: http://docs.oracle.com/cd/E18728_01/html/821-2848/gds-1.html)
For b) just do a
clrg create test-rg; clrslh create -g test-rg <name of IP address); clrg online -M test-rg
Then you should have an HA IP address that you can switch between nodes.
BTW: The documentation I quote is for OSC 3.3 5/11. If you are on S11 and OSC 4.0 please use http://docs.oracle.com/cd/E23623_01/index.html
I am interested in how I can get the shared file system setup using iscsi mounter ZFS pools.
I have found a number of resources but nothing is 100 clear.
Any pointers to help provide some context and assistance would be appreciated.
Could you just clarify your question a little?
1. Are you talking about an external zpool that is providing the iSCSI devices
2. By shared file system, do you mean a failover file system (mounted on one clustrer node at a time) or a global file system (mounted on all cluster nodes at once)
This is for a zpool that is mounted on an iscsi target server, I have connected the iscsi initiators but I am having issues with setting this up as the shared file system that is mounted by each node of the cluster when it is active.
Any pointers would be great.
ZFS is not a shared, global file system. If you want to use it as file system that is mounted on all active cluster nodes at the same time this won't work. This is not depending on the usage of iSCSI or not. ZFS just has not been designed to be a global FS.
However, if you are trying to create a FFS >Failover FileSystem, you would use the HA-Storage(+) Resource and refer to the "Zpool" property. Use ZFS/Zpool command syntax to set ZFS/Zpool properties (mount point, zoned, etc). There are restrictions regarding using a disk with ZFS as a quorum device. Refer to the documentation or look at Tim Read's books. HTH, Michael
Edited by: MichaelHWebb on Jun 28, 2012 7:35 AM