On 3/10/2020 8:26 AM, Jamo Luhrsen wrote:

On 3/9/20 10:57 PM, Robert Varga wrote:

On 10/03/2020 00:36, Jamo Luhrsen wrote:

On 3/7/20 5:10 PM, Robert Varga wrote:

On 07/03/2020 14:19, Robert Varga wrote:

Combined patch series for downstreams to adopt these versions in
stable/sodium lives here:

https://git.opendaylight.org/gerrit/q/topic:"mri-sodium-sr3"

And has an multipatch build going on here:

https://jenkins.opendaylight.org/releng/job/integration-multipatch-test-sodium/107/

... which passed. How can I check CSIT on this thing again? :)

what I would have done is take the distribution produced in the
dist-check job and pass
it to our distribution-test CSIT parent job, but looks like there was a
failure so no distribution
was created:

https://jenkins.opendaylight.org/releng/job/controller-distribution-check-sodium/908/console

Unless I'm missing something, since you said it passed. ?

Wrong job, dude :)

https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/integration-multipatch-test-sodium/107/console.log.gz:

my bad. but what is the dist-check job you see when you go here? it's red:
https://jenkins.opendaylight.org/releng/job/integration-multipatch-test-sodium/107/

Bundle uploaded to
https://nexus.opendaylight.org/content/repositories/opendaylight.snapshot/org/opendaylight/integration/controller/karaf/0.11.3-SNAPSHOT/karaf-0.11.3-20200307.184159-9.zip

I started int-dist-test-sodium #509 ... in a couple of hours once #508
finishes up...

ok, #509 is done. I looked through all red and yellow csit and two jobs stood out as
something to dig in to.

a: https://jenkins.opendaylight.org/releng/job/netconf-csit-1node-callhome-only-sodium/504/
b: https://jenkins.opendaylight.org/releng/job/ovsdb-csit-3node-upstream-clustering-only-sodium/504/

both had a lot of failures in an otherwise stable history.

for a: I noticed the karaf.log was pretty messy with lots of ERRORs, but it's the same in
a passing job too. The high level failure is 404 when trying to see if the callhome device
ended up in operational. I suppose the problem could be with the tool, which is
netopeer via docker (I think). But, since this job is usually 100% pass, I had to point it
out.

for b: it's usually 100% as well, but there was a lot of trouble when trying to bring
back up a controller that was downed. Looks like aaa ran in to trouble and never
started which killed the show from that point onward. In the karaf.log below, you
will eventually find "Unable to start container for blueprint bundle org.opendaylight.aaa.cert"

https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/ovsdb-csit-3node-upstream-clustering-only-sodium/504/odl_1/odl1_karaf.log.gz

No idea if either of those would be related to your work, and everything
else seems to be in line with what we expect.

Thanks,
JamO

ping me when it's done and I'll glance through to see if anything seems off.

Thanks,
JamO

Thanks,
Robert