14:01:35 <fdegir> #startmeeting Cross Community CI
14:01:35 <collabot`> Meeting started Wed Oct 31 14:01:35 2018 UTC.  The chair is fdegir. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:01:35 <collabot`> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:01:35 <collabot`> The meeting name has been set to 'cross_community_ci'
14:01:39 <fdegir> lets go then
14:01:44 <fdegir> #topic Rollcall
14:02:00 <mbuil> #info Manuel Buil
14:02:15 <hwoarang> #info Markos Chandras
14:02:45 <fdegir> i suppose we are alone this time again
14:02:50 <fdegir> #topic Baremetal Status
14:03:11 <fdegir> mbuil: i'll have few questions once you talk about the latest status
14:03:59 <mbuil> fdegir: so, patches were ready. Then I rebase one of them and jenkins is giving a -1
14:04:26 <mbuil> fdegir: need to investigate the problem, fix it and then we quickly merge them before it happens again
14:05:18 <mbuil> #info We have two patches missing. One providing network config for the baremetal nodes in Ubuntu and another one in opensuse
14:05:24 <mbuil> #link https://gerrit.opnfv.org/gerrit/#/c/60799/
14:05:31 <mbuil> #link https://gerrit.opnfv.org/gerrit/#/c/63739/
14:05:52 <mbuil> #info After rebasing, jenkins is giving -1. Needs investigation
14:06:29 <mbuil> #info to complete the opensuse work, we require a rabbitmq patch from OSA: https://review.openstack.org/#/c/614355/
14:06:30 <fdegir> ok
14:06:37 <mbuil> that's it
14:06:39 <fdegir> sorry - was too fast
14:06:44 <fdegir> ok :)
14:06:50 <fdegir> so the first question i would like to ask is
14:07:12 <fdegir> what is the relation between the flavors we use and the pdf?
14:07:20 <fdegir> is it like; if it is noha, use the first 3 nodes
14:07:28 <fdegir> if it is mini, use the first 2
14:07:33 <fdegir> and for ha, use all
14:07:38 <mbuil> yes
14:07:39 <fdegir> is this correct understanding?
14:07:42 <mbuil> correct
14:07:43 <fdegir> ok
14:07:49 <fdegir> it seems i'm not that behind
14:07:55 <fdegir> the second question is about this part
14:08:07 <fdegir> https://github.com/opnfv/releng/blob/master/jjb/xci/xci-run-functest.sh#L53
14:08:15 <mbuil> fdegir: look at this: https://github.com/opnfv/releng-xci/blob/master/xci/var/idf.yml#L38-L44
14:08:16 <fdegir> maybe this is a question to both of you
14:08:37 <fdegir> now i know where i saw this
14:08:40 <fdegir> thanks for the link
14:08:51 <fdegir> so we are still dynamic when it comes to setting the flavor
14:09:07 <fdegir> without changing the pdf, we can assign roles to the nodes - which is very useful
14:09:12 <mbuil> fdegir: we don't need that anymore. I have a patch that removes that ==> https://gerrit.opnfv.org/gerrit/#/c/63737/
14:09:37 <fdegir> mbuil: but...
14:09:46 <fdegir> the link i pasted is used by verify jobs
14:09:58 <fdegir> and what your change does is fixing it for daily
14:10:07 <fdegir> i suppose you need to fix verify jobs too, isn't it?
14:10:30 <fdegir> sorry
14:10:33 <fdegir> you fix both
14:10:56 <mbuil> fdegir: yes. The reason why I added that was because we needed to know where opendaylight was deployed
14:11:05 <fdegir> yeah
14:11:08 <mbuil> fdegir: but if you remember, you gave me an alternative solution which is simpler
14:11:20 <fdegir> kind of ;)
14:11:28 <fdegir> too many things happening and i'm forgetting things
14:11:30 <mbuil> and now it is done in the prepare.... playbook of each scenario
14:11:55 <fdegir> but this answers my question since i was going to say that line is osa specific, causing warnings during k8s deployments
14:13:54 <fdegir> hwoarang: anything you want to add for baremetal or are you ok for us to move on?
14:14:12 <hwoarang> nothing on my end
14:15:07 <fdegir> ok
14:15:26 <fdegir> #topic Status of os-nosdn-osm and os-odl-sfc_osm
14:15:33 <fdegir> i'm not sure is avhvr is around
14:16:01 <OPNFV-Gerrit-Bot> Manuel Buil proposed releng-xci: [baremetal] suse support for baremetal networking  https://gerrit.opnfv.org/gerrit/63739
14:16:09 <OPNFV-Gerrit-Bot> Manuel Buil proposed releng-xci: [Baremetal] Use physical interfaces for networking  https://gerrit.opnfv.org/gerrit/60799
14:16:19 <OPNFV-Gerrit-Bot> Manuel Buil proposed releng-xci: [baremetal] suse support for baremetal networking  https://gerrit.opnfv.org/gerrit/63739
14:16:32 <mbuil> I had to rebase, sorry ^
14:16:45 <mbuil> fdegir: if he is not around, I can give a status
14:16:57 <fdegir> mbuil: please do
14:17:58 <mbuil> #info The scenario os-odl-sfc_osm is finished and it deploys successfully using OSM version 4. However, there is a bug for SFC. The SFC config is not propagated into openstack
14:18:40 <mbuil> #info Harsha talked to OSM guys and they recommended to move to master branch where the bug is fixed. Now the scenario uses master branch and the deployment fails. Harsha is investigating it
14:19:14 <mbuil> Note that Harsha will not work during November because of his exams period
14:19:49 <fdegir> yep
14:20:01 <fdegir> but i think xci doing what is supposed to do in this case
14:20:05 <fdegir> finding bugs ;)
14:20:27 <fdegir> mbuil: thanks for the update
14:20:33 <fdegir> #topic CI Status
14:20:56 <fdegir> #info The daily jobs have been failing which needs investigation
14:21:27 <fdegir> mbuil: i think the place they fail is the place where you are doing changes as well with https://gerrit.opnfv.org/gerrit/#/c/63737/
14:22:01 <fdegir> mbuil: so I will wait until after the above change is merged, attempt to run 1 os and 1 k8s daily job and then fix daily jobs if they continue to fail at that point
14:22:19 <mbuil> ok
14:23:10 <fdegir> #info Once the change https://gerrit.opnfv.org/gerrit/#/c/63737/ gets merged, the issue will be attempted to be reproduced and fixed accordingly
14:23:19 <fdegir> mbuil: this one is for you as well
14:23:46 <fdegir> so, once we get the daily jobs fixed and baremetal stuff in, i suppose we can start running daily jobs on baremetal
14:24:30 <mbuil> right. We will need the idf and pdf of the servers
14:24:47 <fdegir> i thought we had them for ericsson-pod2
14:25:01 <mbuil> we have the ericsson-pod2 pdf and idf
14:25:14 <fdegir> yes, so we can start using that for daily jobs
14:25:20 <mbuil> aaah, correct
14:25:35 <fdegir> and the question to all of us is that can you guesstimate the resource need?
14:25:47 <fdegir> for bareemetal dev work especially
14:26:01 <fdegir> sicne we are losing the pod to ci which you used for development
14:26:28 <fdegir> we were supposed to get lf-pod4 but i'm not sure if the networking issue is fixed there
14:26:32 <fdegir> need to check with bramwelt
14:26:39 <mbuil> fdegir: yes, it was fixed. We also have it
14:26:47 <fdegir> mbuil: i didn't know it
14:26:50 <fdegir> then we are good
14:26:55 <mbuil> tes
14:26:56 <mbuil> yes
14:27:35 <fdegir> #info After fixing the daily jobs and baremetal patches are merged, daily jobs will be switched to baremetal from virtual
14:27:54 <fdegir> #info The testing we do will be adjusted accordingly as well
14:28:09 <fdegir> that is all for CI stuff
14:28:15 <fdegir> anything to add?
14:29:16 <fdegir> i take this as no
14:29:21 <fdegir> #info Release Status
14:29:25 <fdegir> #undo
14:29:25 <collabot`> Removing item from minutes: <MeetBot.ircmeeting.items.Info object at 0x22a4f90>
14:29:29 <fdegir> #topic Release Status
14:29:46 <fdegir> i think we should think of tagging the repos during next week
14:29:51 <fdegir> this includes sfc as well
14:30:01 <mbuil> right
14:30:15 <fdegir> we need to think about what to document
14:30:23 <fdegir> given that we have lots of updates lately
14:30:45 <fdegir> i'll try to do that and send reviews so at least i can contribute something - which i failed to do recently
14:30:45 <mbuil> fdegir: are we going to be able to run sfc tests in order to check that everything is fine and thus do the tagging?
14:30:59 <fdegir> mbuil: that's the plan
14:31:13 <fdegir> mbuil: that's why i was looking at daily jobs
14:31:31 <fdegir> mbuil: once they turn green, we can include sfc tests for sfc scenario before we add anyhting else
14:32:23 <fdegir> mbuil: and when we get a green sfc run, that sha could be tagged which then becomes your release candidate to get opnfv-x.y.z tag
14:32:39 <fdegir> and pinned in xci repo
14:32:44 <mbuil> ok
14:33:11 <fdegir> i think that was all i wanted to talk about
14:33:17 <fdegir> anything you want to add mbuil hwoarang >
14:33:19 <fdegir> ?
14:33:44 <mbuil> nope
14:35:09 <fdegir> ok then
14:35:29 <fdegir> thanks a lot for everything you've been doing and apologies again for not being around lately
14:35:36 <fdegir> but i'm coming back
14:35:41 <fdegir> have a nice day!
14:35:43 <fdegir> #endmeeting