Hi Ben,
> [This was originally on xen-devel. I''m resending to xen-api at a
> suggestion from Goncalo Gomes, who thought this might be a better place
> to ask. My apologies if you''ve already read it.]
Thanks for the retransmit -- I was intending to reply to the original message
but forgot.
> I noticed an item on the most recent XCP meeting minutes, "vSwitch -
> needs more testing". As one of the OVS developers, I''m
curious what
> kind of additional testing is needed.
I think the sentiment was, "everyone please try it in your environment and
give us some feedback". As you say, there''s lots of explicit
testing going on so we''re confident it''s great. However
there''s always the slim possibility that a problem surfaces in some
obscure environment-- so the more end-user testing it gets the better.
Note that XCP 0.5 (should be in RC form RSN) will have a refreshed openvswitch
so that should be even better than the older snapshot in XCP 0.1.1.
> Perhaps I can get our QA guys to
> add some more tests to what they do already, or I can ask them whether
> we can give out a description of the testing we do. Currently we
> internally test XenServer, but not XCP, but I imagine that some of the
> tests are still useful.
That would be interesting. As far as networking goes, XenServer uses the same
code as XCP so the test results should carry over nicely.
Cheers,
Dave
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel