Single Point Of Failure For MSCS Clusters

[thattommyhall2]

Am I right in saying that the LUN containing any RDM files is a SPOF in the MSCS setup as as per the MSCS howto pdf?

[vmmeup]

It could be. Depends on your SAN and how it is configured. Typically you will have redundant HBA's, Switches, and controllers along with a RAID 1, or 5 configuration for the spindels, but yes if your Array is tanked then you loose the LUN, you could cluster arrays and take remove that point of failure.

Sid Smith-----

VCP, VTSP, CCNA, CCA(Xen Server), MCTS Hyper-V & SCVMM08

• Don't forget to award points for correct and helpful answers.;-)

[thattommyhall2]

I guess you could use any LUN mirroring the SAN provides too, we have redundant SPs, HBAs and switching for the SAN, just struck me the LUN is SPOF in our current setup.

Cheers.

Tom

[kastlr]

Hi,

don't forget the filesystem and application data.

Depending on the failure condition, it might happen that the HW survives without an issue but the filesystem and/or application data need extra maintenance tasks.

Hope this helps a bit.

Greetings from Germany. (CET)