On Wed, 2011-06-08 at 20:16 -0700, Tim Vaillancourt wrote:
> Fantastic. I really appreciate your reply, and that clarifies things
> wonderfully.
>
> We've recently increased the VM count on the Xen servers we have this
> issue on, but we (still) run the duplicity backups from the Dom0 on one
> VM at a time backup load theoretically the same as before at any one
> time), so I'm thinking now the increased VM count (and thus IO) causes
> this to happen while duplicity is running at the same time in the dom0.
>
> That is really too bad cause IO-wise we aren't even pushing the disks
> that hard, but I guess we are asking it to do a lot of simultaneous
> things at once.
>
> Again, thank you for the reply. By chance, do you have any links to the
> mainstream bug you mentioned that is already open?
There wasn't any bug report; I just compared the code between 2.6.26 and
current development. My reply to you also went to the aacraid
developers and the SCSI mailing list.
Please reply-to-all.
Ben.
--
Ben Hutchings
Lowery's Law:
If it jams, force it. If it breaks, it needed replacing anyway.
Attachment:
signature.asc
Description: This is a digitally signed message part