[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-API] OVS, CloudStack, and Crashing XenServer


  • To: "'xen-api@xxxxxxxxxxxxx'" <xen-api@xxxxxxxxxxxxx>
  • From: Clayton Weise <cweise@xxxxxxxxxx>
  • Date: Fri, 6 Jul 2012 16:13:03 +0000
  • Accept-language: en-US
  • Delivery-date: Mon, 09 Jul 2012 15:40:13 +0000
  • List-id: User and development list for XCP and XAPI <xen-api.lists.xen.org>
  • Thread-index: Ac1bkK06Zj4iKTJURl22bwtLjQCC6g==
  • Thread-topic: OVS, CloudStack, and Crashing XenServer

I have recently encountered a reproducible and rather catastrophic issue with 
OVS and XenServer caused by CloudStack.  For CloudStack, during the "snapshot" 
process with CloudStack a snapshot is created on the XS host, then it's copied 
off to an NFS share as a .vhd file via a specialized VM called the secondary 
storage VM (SSVM).

If the SSVM and guest VM that is being snapshotted are on the same host about 
4-5 GB into the copy process ovs-vswitchd consumes all available CPU on the 
host to the point that it becomes unusable and all management interfaces 
(including our 10gbit iSCSI NICs) become unresponsive.  I've waited over 30 
minutes for this problem to correct itself but it never seems to.  After 
switching from OVS to bridged networking mode I have been able to successfully 
copy 100GB snapshots without issue.

Any insight into this would be appreciated.  I would be happy to reproduce this 
problem and provide any logs available.

_______________________________________________
Xen-api mailing list
Xen-api@xxxxxxxxxxxxx
http://lists.xen.org/cgi-bin/mailman/listinfo/xen-api


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.