I am really not sure.  I know that I destroyed the pool when the storage
guys asked me to.  Then I created a new pool.  Once I did that it looked
like all the host came back without me having to add them.

I wondered if the pool really never got destroyed.  The storage was
acting so bad that I wondered if it tried but did not really do it.  If
I remember correctly, I restarted the master because things got hung so
when it came up, I tried again to delete the pool and it did it.  Within
10 seconds I had created a new pool and the hosts we already in it.

Sorry, I misunderstood your previous post and posted to devel.  :-(

Juan

+

-----Original Message-----
From: Mike McClurg [mailto:[email protected]] 
Sent: Friday, November 09, 2012 4:15 AM
To: Juan Lorenzana
Cc: Camron Lent; [email protected]
Subject: Re: [Xen-API] Can't Start any VM's from NFS Repository

On 09/11/12 10:57, Juan Lorenzana wrote:
> Mike,
> 
> Thanks.  We actually still have the original VMs.  Believe it or not, 
> when we destroyed the Pool, the VM information went with the master.  
> So when we recreated, they transferred over.

Yes, the master stores all metadata for the pool. Perhaps I'm confused
what you mean by "destroyed the pool". It sounded like you just wiped
all the hosts and started fresh.

> So I still have the VM and their UIDs as they were in the previous
pool.
> 
> 
> For whatever reason, I just can not see what we think was the running 
> .vhd file when we go to the storage repository in Xen Center.  I have 
> some 200+ files there but only about 100 of them show up when I look 
> through the XenCenter software.

I'm pretty sure that XenCenter wouldn't show you individual VHD files
anyway, just the logical VDIs. If you really want to see the VHD files,
you can look in dom0 at /var/run/sr-mount/<sr-uuid>, but I'm not sure
how useful it will be to see what VHD files there are.

> I will post to [email protected].  That was a good idea.

No, don't do that. That list is for hypervisor development; they don't
care/know about XCP there, and will just sent you back here.

Mike

_______________________________________________
Xen-api mailing list
[email protected]
http://lists.xen.org/cgi-bin/mailman/listinfo/xen-api

Reply via email to