15:03:22 <anteaya> #startmeeting third-party
15:03:23 <openstack> Meeting started Mon Aug  3 15:03:22 2015 UTC and is due to finish in 60 minutes.  The chair is anteaya. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:03:24 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:03:26 <openstack> The meeting name has been set to 'third_party'
15:03:38 <anteaya> #chair asselin
15:03:39 <openstack> Current chairs: anteaya asselin
15:03:45 <anteaya> hello
15:03:55 <anteaya> <- laggy wifi
15:03:58 <asselin> hi
15:04:01 <eantyshev> hello!
15:04:17 <anteaya> asselin: do step in if I disappear
15:04:27 <asselin> sure
15:04:35 <anteaya> thanks
15:04:51 <anteaya> what would folks like to discuss today?
15:05:28 <hodos> hi
15:05:54 <akerr> I noticed the CI scoreboard website went down over the weekend.  Did it move or is it broken?
15:05:55 <anteaya> anyone with any questions?
15:06:06 <anteaya> akerr: what link?
15:06:17 <akerr> http://ec2-54-67-102-119.us-west-1.compute.amazonaws.com:5000/?project=openstack%2Fcinder&user=&timeframe=72
15:06:42 <asselin> that's hosted by patrickeast
15:06:51 <akerr> asselin: Ok, I'll ping him about it
15:06:54 <asselin> akerr, you can host your own fyi
15:07:27 <asselin> #link scoreboard source http://git.openstack.org/cgit/stackforge/third-party-ci-tools/tree/monitoring/scoreboard
15:07:37 <akerr> asselin: Ah, I noticed it looking more and more like standard infra branding and wasn't sure if it finally got legitimized
15:08:00 <eantyshev> please, clarify that issue with merge failure reports from: https://review.openstack.org/#/c/206513
15:08:38 <eantyshev> perhaps, we should discuss it here
15:09:39 <asselin> just looked at the patch.
15:10:00 <anteaya> akerr: well I would like infra to host a copy of patrick's dashboard
15:10:16 <anteaya> anyone know how that spec is coming along?
15:10:24 <asselin> from what i found, 3rd party merge failures can also occur if there's a network issue getting the source patch
15:10:55 <anteaya> eantyshev: there is nothing a dev can do if a third party ci has a merge failure
15:11:14 <asselin> anteaya, last I heard, there were 'issues' regarding hosting a 'temporary' solution
15:11:15 <mmedvede> krtaylor: around? the spec question above
15:11:20 <anteaya> so reporting that a third party ci has a merge failure is noise for the dev
15:11:43 <anteaya> asselin: why is it considered temporary?
15:11:46 <eantyshev> anteaya: What if it is a valid merge failure?
15:12:02 <akerr> its perfectly possible and legitimate for a 3rd party CI to have a merge failure when Jenkins did not.  If the 3rd party CI waits for jenkins to vote +1 then the time between Jenkins beginnings its jobs and it finally voting could allow the base repo to be updated by conflicting merges
15:12:04 <asselin> anteaya, I recall the spec said so
15:12:15 <anteaya> and yes we won't go through the effort of hosting something if it is temporary
15:12:28 <anteaya> asselin: can we remove the wrokd temporary?
15:12:42 <eantyshev> akerr: ++, same I wrote in review 206513
15:13:12 <anteaya> valid merge failure, okay what does the dev do?
15:13:20 <akerr> anteaya: they'll have to rebase
15:13:25 <asselin> akerr, eantyshev perhaps we need to add the merge-check to 3rd party as well
15:13:25 <anteaya> there is nothing the dev is able to do
15:13:32 <akerr> anteaya: In this case if jenkins were to recheck it would also find the merge failure
15:13:34 <anteaya> on a merged patch?
15:13:54 <mmedvede> eantyshev: on valid merge failure, upstream zuul would report the same, so no point of every third-party CI reporting the failure too
15:13:57 <anteaya> no, we won't be asking devs to rebase for third paerty ci sytems
15:14:01 <akerr> anteaya: this is prior to the final merge, it's during the check pipeline
15:14:20 <anteaya> if jenkins needs them to rebase, jenkins will tell them
15:14:43 <anteaya> devs won't be rebaseing for any other repo other thank the one in gerrit
15:14:47 <akerr> anteaya: This is not a merge conflic with the 3rd party ci code, it's a conflic with the base project repo.  its just that the 3rd party ci hit it first because it ran later
15:14:55 <eantyshev> mmedvede: not exactly, there could be some time between these checks
15:15:00 <anteaya> s/thank/than
15:15:09 * krtaylor joins late
15:15:35 <krtaylor> #link https://review.openstack.org/#/c/194437/
15:15:41 <anteaya> devs get information for their actions from jenkins, not from third party ci systems
15:15:51 <krtaylor> anteaya, the spec for hosting a dashboard  ^^^
15:16:13 <akerr> I'm indifferent on whether 3rd party ci's should comment on merge-failures though.  Eventually jenkins will hit it too, so I could see the argument for silencing them
15:18:37 <krtaylor> re: hosting monitoring dashboard - no issues except some confusion on big solution spec being moved, but we can discuss that tomorrow in the working group meeting
15:19:16 <anteaya> well I can't see other infra folks being willing to do the work to host a service if it is temporary
15:19:49 <anteaya> so word it how you like but don't expect much support from infra until that word is removed
15:20:44 <krtaylor> it was jeblair that suggested it
15:21:13 <anteaya> well great then he must be willing to do the work
15:21:37 <krtaylor> to get something working immediately, then we can roll out an end-all-be-all solution
15:22:25 <krtaylor> we can do the work once the spec is approved, there are other good examples
15:22:45 <krtaylor> that spec just needs more reviews
15:26:05 <anteaya> anything more here?
15:28:12 <anteaya> anyone have anything more on 206513?
15:29:38 <anteaya> anyone have anything else they would like to discuss today?
15:31:01 <anteaya> anyone have a reason why I shouldn't close this meeting?
15:31:06 <asselin> is anyone else using the new 3rd party ci scripts?
15:31:43 <anteaya> asselin: link to what scripts you mean?
15:32:26 <asselin> #link common-ci scripts: http://git.openstack.org/cgit/openstack-infra/puppet-openstackci/tree/
15:32:40 <anteaya> thank you
15:34:03 <asselin> silence = yes or no?
15:34:54 <anteaya> I talk silence to mean not anyone currently staring at their irc client
15:35:00 <anteaya> s/talk/take
15:35:19 <asselin> fair enough, then we can end meeting :)
15:35:24 <mmedvede> asselin: I answered before - we use it, but not the most recent version
15:35:31 <anteaya> is anyone considering it but hasn't yet?
15:35:45 <pots1> yes, i'd like to use it for a new CI setup
15:35:56 <anteaya> mmedvede: oh osrry I missed your answer before
15:36:12 <mmedvede> anteaya: it has bee a few weeks ago :)
15:36:18 <anteaya> pots1: thanks, can you share feedback with asselin when you do
15:36:32 <anteaya> mmedvede: ah I forget anything longer than 5 minutes ago
15:36:49 <asselin> mmedvede, thanks for the reminder. glad to hear it
15:36:49 <pots1> will do
15:36:49 <anteaya> mmedvede: any feeback on the scripts?
15:37:03 <anteaya> pots1: thank you
15:37:28 <pots1> has there been any consideration of trying to provide pre-built images?
15:37:29 <mmedvede> anteaya: the puppet module works, but we still can not use it "as is", minor adjustments necessary for our CI, but they can go away soon
15:37:39 <asselin> #link WIP patch for common-ci https://review.openstack.org/#/c/200330/
15:37:48 <anteaya> mmedvede: okay thank you
15:38:27 <asselin> reviews would be appreciated, at least at the level of "right or wrong direction"
15:38:38 <anteaya> mmedvede pots1 would either of you have some time to review 200330?
15:39:13 <asselin> pots1, what do you mean by pre-build images?
15:39:22 <mmedvede> anteaya: I'll find some time
15:39:29 <anteaya> mmedvede: thank you
15:39:56 <asselin> mmedvede, thanks
15:40:43 <pots1> in other words, minimal linux images with the packages installed.
15:41:23 <asselin> pots1, I think the idea is to start with base linux image and have a script that does the right thing from there
15:41:32 <pots1> e.g. the log server should be pretty simple but it took me a while to figure out what packages it depended on.  i am new to puppet and so far it has only made things harder.
15:42:27 <asselin> pots1, which documentation are you following?
15:42:55 <pots1> your github repo readme.md
15:43:30 <pots1> and anything else i can find.  as a noob i'm not finding the docs easy to follow.
15:43:58 <asselin> pots1, feedback would be very much appreciated, espceially specifics
15:44:14 <pots1> sure, may i pm you?
15:44:29 <anteaya> pots1: can you offer some patches with what you would like to see?
15:44:39 <asselin> pots1, sure
15:45:01 <pots1> will do
15:45:39 <anteaya> thank you
15:45:51 <anteaya> anything more here?
15:47:06 <asselin> not from me
15:47:14 <anteaya> thanks asselin
15:47:27 <anteaya> any reason why I shouldn't close the meeting?
15:47:53 <pots1> asselin anteaya: thanks
15:48:11 <anteaya> pots1: thanks for speaking up
15:48:23 <anteaya> well let's wrap pu then
15:48:44 <anteaya> thanks to everyone for your kind attendance and participation
15:48:48 <anteaya> see you next week
15:48:54 <anteaya> #endmeeting