Virtual Machine will not fail!

[foofighter26]

Hi,

We are currently testing Symantec Storage Foundation to mirror data across sites in conjunction with Microsoft Clustering. We are testing that if we lose the C: drive of the active node the cluster fails over to the passive node. The problem we are having is that when the C: drive of the active node is lost the virtual machine starts queuing all commands etc (I assume on the HBA or in memory) and then does not fail and when you reconnect the C: the queued commands are all executed as normal. This is an issue as the commands will queue for some time until the memory has run out of space, so we need it to fail as soon as it loses its C: drive?

When we test this with ESX 3.5 vm's it fails instantly so I assume its something to do with the new multipathing in VSphere or something like that?

Thanks,

Paul

[RParker]

> so I assume its something to do with the new multipathing in VSphere or something like that?

Or something like that. So specify that the path is fixed, or kill the other path so you can fail the VM. I bet I can make it fail...

[foofighter26]

The policy is set to fixed. And I can make them fail to be fair but I want it to fail on its own when it loses C: drive connectivit.

Paul

[malaysiavm]

maybe you can consider FT rather than OS clustering, just a suggestion

Craig

vExpert 2009

Malaysia VMware Communities -

[foofighter26]

Hi,

I did not think FT was supported cross-site?

Paul

[malaysiavm]

what I meant is for same site and enable FT.

Craig

vExpert 2009

Malaysia VMware Communities -