I'm not positive, but I don't think there is a k3 distro in netvirt master. 
Better if
someone can confirm that.

but, looking at the flow rules ending up on one node, things don't look right. 
there
are some drop rules I don't think should be there. See table 51 here:

https://logs.opendaylight.org/releng/jenkins092/netvirt-csit-1node-openstack-newton-nodl-v2-upstream-stateful-nitrogen/69/log.html.gz#s1-s1-s1-t8-k14-k2-k1-k1-k1-k15-k4

Thanks,
JamO

On 06/19/2017 11:06 AM, Luis Gomez wrote:
> Sam, do you still maintain K3 local distro in master? If so can you repeat 
> the test with K3 distro? this way we can confirm this is K4 (not OFP issue).
> 
>> On Jun 19, 2017, at 4:08 AM, Sam Hague <[email protected]> wrote:
>>
>> adding openflowplugin
>>
>> On Mon, Jun 19, 2017 at 7:08 AM, Sam Hague <[email protected]> wrote:
>>> Hi all,
>>>
>>> could everyone take a look at the job [1] to see if they know the
>>> problem with the tests? the nitrogen csit is running with the netvirt
>>> distribution and has a consistent 77 failures. Carbon job is working
>>> fine so this could be a karaf 4 issue or something else has changed on
>>> nitrogen outside netvirt. There are warnings coming from the openflow
>>> serializer which there is another thread about.
>>>
>>> Thanks, Sam
>>>
>>> [1] 
>>> https://logs.opendaylight.org/releng/jenkins092/netvirt-csit-1node-openstack-newton-nodl-v2-upstream-stateful-nitrogen/69/odl1_karaf.log.gz
>> _______________________________________________
>> openflowplugin-dev mailing list
>> [email protected]
>> https://lists.opendaylight.org/mailman/listinfo/openflowplugin-dev
> 
> _______________________________________________
> netvirt-dev mailing list
> [email protected]
> https://lists.opendaylight.org/mailman/listinfo/netvirt-dev
> 
_______________________________________________
openflowplugin-dev mailing list
[email protected]
https://lists.opendaylight.org/mailman/listinfo/openflowplugin-dev

Reply via email to