So what should happen now is that when the host comes back online, the VDIs should get unlocked without any further intervention. Thus waiting for your host to reboot is the preferred solution.
If the host is permanently dead, or you want to get things back online more quickly, you'll need to:
1. Execute an 'xe vm-reset-powerstate' cli command (as was the case in 0.5)
2. Execute the script:
/opt/xensource/sm/resetvdis.py <host uuid> <sr uuid> [master]
where the arguments are the host that has gone away, the sr in which the VDIs reside, and whether the host is the SR master (not the pool master)
In almost every instance where this is useful (for a user to execute), the 'master' field will not be set, as the times when the script is useful is when your locked VDIs are on
shared storage, and in the case where the pool master is the host that died, the designate-new-master process will cause the script to be executed anyway.
Hope this helps!
Jon
On 17 Feb 2011, at 10:42, Chris Percol wrote:
Hi,
Just wanted to post an update on how we fixed 'vdi not available' and 'sr backend failure' issues after pulling the power on two of our xcp hosts several seconds after initiating a shutdown. We are usng an iSCSI san for storage (qsan).
###Fix VDI Problems after Power Failure###
#Steps that fixed 'vdi not available', sr backend failure
console
xe pbd-list sr-uuid=<uuid>
xe pbd-unplug uuid=<uuid>
Xen Center
Storage Forget
Storage > New Storage
xsconsole
Restore Virtual Machine Metadata
Thanks for previous help on this issue.
Chris
On Fri, Jan 21, 2011 at 1:18 PM, Chris Petrolino
<cpetrolino@xxxxxxxxx> wrote:
Thanks guys I'll give these a shot as soon as I get into the office!
Kind Regards,
Christopher James Petrolino
Hi,
I've worked around this issue a couple of times - These are my notes:
-
-
-
xe sr-forget uuid=c208e85a-b5ea-8b89-62ff-52ebbf5263f8
xe sr-introduce uuid=c208e85a-b5ea-8b89-62ff-52ebbf5263f8 type=ext name-label='Local storage' content-type=user
-
xe pbd-create sr-uuid=c208e85a-b5ea-8b89-62ff-52ebbf5263f8 device-config:device=/dev/disk-by-id/scsi-360345678123456781300046d062de2ee-part3 host-uuid=a600493f-d142-4785-8618-d00dc5069b1e
xe pbd-plug uuid=dc103449-ef50-6a05-c58b-995de0314edd
vgrename VG_XenStorage-c208e85a-b5ea-8b89-62ff-52ebbf5263f8 XSLocalEXT-c208e85a-b5ea-8b89-62ff-52ebbf5263f8
-
xe pbd-plug uuid=dc103449-ef50-6a05-c58b-995de0314edd
restore metadata or guess which VDI's are connected to VMs
Hey guys I have seen a couple other folks post about this happening
but haven't seen if anyone has successfully figured out the cause and
or solution to this problem.
One of our network guys did an accidental plug pulls on one of our XCP
1.0 beta boxes. I have a couple VDI on this box " a few that are on
shared storage, and a few that are on local storage. Every time I try
to start any of them I get "the vdi is unavailable" or on open xen
manager I get - Error parameters: , The VDI is not available
[opterr=VDI fc77b366-950b-49be-90ce-2a466cf73502 already attached RW],
Any advice?
Also if this isn't the right list can anyone point me to the appropriate list?
thanks so much
_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users
