[ovs-discuss] ovs get stuck when running traffic from VM to VM on same compute
Daniele Di Proietto
diproiettod at ovn.org
Tue Apr 26 17:28:43 UTC 2016
2016-04-26 9:08 GMT-07:00 Traynor, Kevin <kevin.traynor at intel.com>:
> > -----Original Message-----
> > From: discuss [mailto:discuss-bounces at openvswitch.org] On Behalf Of
> > Kochba, Alon
> > Sent: Tuesday, April 26, 2016 4:38 PM
> > To: Ben Pfaff <blp at ovn.org>; Yi Ba <yby.developer at yahoo.com>
> > Cc: bugs at openvswitch.org
> > Subject: Re: [ovs-discuss] ovs get stuck when running traffic from VM
> > to VM on same compute
> >
> > Hi Ben,
> >
> > Could you point us to the commit that fixed this issue?
> > We already tried patching with this commit which seemed relevant, but
> > the issue still recreated -
> > https://github.com/openvswitch/ovs/commit/f519a72d9a3708fbc5f796f176e7
> > c8bd3dcfb738
> >
> > We will retry with your suggestion of using the 2.5 branch code, but
> > we might want to backport the specific fix unless there is a 2.5.1
> > release including it.
> > If the commit linked above is the one you were thinking of, please
> > note a small difference - in the commit the rcu is blocked waiting for
> > vhost_thread to quiesce, while in our case rcu is blocked waiting for
> > pmd to quiesce.
>
> It sounds similar to the problem that this commit fixed. If so the fix
> is applied to master and 2.5 branches.
>
>
> https://github.com/openvswitch/ovs/commit/61c4e39460a7db3be7262a3b2af767a84167a9d8
>
Could you try applying the above commit and see if it fixes the problem?
If you manage to reproduce the problem, could you get a backtrace of the
blocked thread (pmd101 in this case)?
Thanks,
Daniele
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://openvswitch.org/pipermail/ovs-discuss/attachments/20160426/fd680b88/attachment-0002.html>
More information about the discuss
mailing list