On 5/11/19 1:29 AM, Robert Varga wrote:

Hello everyone,

I ran multipatch builds on
https://git.opendaylight.org/gerrit/#/q/topic:mdsal-4.0.1 to validate
the upgrade:

https://jenkins.opendaylight.org/releng/view/integration/job/integration-multipatch-test-sodium/33/
https://jenkins.opendaylight.org/releng/view/integration/job/integration-multipatch-test-sodium/36/
https://jenkins.opendaylight.org/releng/view/integration/job/integration-multipatch-test-sodium/37/

The first two jobs ran unit tests and failed twice in elanmanager, yet
it passed in local testing -- are those tests expected to be stable?

The third job ran with -Pq and ran CSIT. From what I see there are some
unstable tests, but I have no clue whether they are regression or jitter
-- can someone take a look at them?

I looked at 6 jobs roughly at random (aaa, bgpcep, genius, netconf,
netvirt, ovsdb) and none
of the failures were out of line from what the job is normally failing
with. Nothing
stood out as worrisome to me. If we want to dig all the way to the
bottom of each
failure to be 100% sure, I'll need to find some extra time to do it.