13:02:15 <IgorYozhikov> #startmeeting rpm_packaging
13:02:16 <openstack> Meeting started Thu Dec 15 13:02:15 2016 UTC and is due to finish in 60 minutes.  The chair is IgorYozhikov. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:02:17 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:02:19 <openstack> The meeting name has been set to 'rpm_packaging'
13:02:30 <IgorYozhikov> ping toabctl, dirk, apevec, aplanas, IgorYozhikov, jpena, jruzicka, number80, kaslcrof
13:02:35 <jpena> o/
13:02:51 <IgorYozhikov> #chair IgorYozhikov jpena toabctl
13:02:52 <openstack> Current chairs: IgorYozhikov jpena toabctl
13:03:18 <toabctl> hi
13:03:36 <dirk> o/
13:03:41 * toabctl will be a bit distracted during the meeting...
13:03:45 <IgorYozhikov> #chair IgorYozhikov jpena toabctl dirk
13:03:46 <openstack> Current chairs: IgorYozhikov dirk jpena toabctl
13:03:55 <IgorYozhikov> let's spend some time on agenda
13:07:04 <IgorYozhikov> let's start
13:07:32 <IgorYozhikov> #topic https://review.openstack.org/#/c/407102/ - semver version part - let's finalize decision
13:07:47 <IgorYozhikov> toabctl, still has questions here
13:09:05 <jpena> Yep, I have questions too. I've seen that the XStatic-* packages have 4 digit versions, e.g. https://review.openstack.org/#/c/410747/8/openstack/XStatic-Magic-Search/XStatic-Magic-Search.spec.j2
13:09:07 <IgorYozhikov> I believe that here we need some additional input from RDO side
13:10:08 <IgorYozhikov> jpena, afaik py2rpmversion is used only for OS projects
13:10:33 <IgorYozhikov> since that dependencies are not affected here
13:10:54 <jpena> ok, in that case we should be fine
13:11:08 <IgorYozhikov> also dependencies doesn't use mid-milestone dev tails in versions, no?
13:11:27 <jpena> no, they don't
13:12:00 <toabctl> renderspec should work also for non-OS project. but if you are fine with the fedora solution, that is ok for me
13:12:19 <toabctl> it just feels wrong. imo we should at least document that it is not doing the right thing
13:12:28 <IgorYozhikov> ok, so as I remember - these 2 functions were developed to handle OS components version/release
13:12:45 <toabctl> IgorYozhikov, the can handle any pre/post release
13:14:03 <IgorYozhikov> toabctl, didn't get your question :(
13:14:19 <toabctl> what question?
13:14:28 <IgorYozhikov> about pre/post releases
13:14:47 <toabctl> hm. there is no question
13:15:32 <IgorYozhikov> ah, ok. and yes - I'm agree about documenting functions usage
13:17:41 <IgorYozhikov> if there are no objections and every1 agree with proposed changes - I will update documentation for py2rpmversion
13:19:17 <IgorYozhikov> otherwise - let's elaborate additional changes which should be made
13:19:40 <IgorYozhikov> your thoughts ?
13:21:57 <jpena> +1 for me
13:23:45 <dirk> next?
13:24:17 <IgorYozhikov> dirk, yes
13:24:47 <IgorYozhikov> hope that we will agree on something just a little bit later
13:24:53 <IgorYozhikov> #topic systemd macro, new common with translation into rdo & suse or something else?
13:25:23 <jpena> dirk, you mentioned the upstream systemd macros where available for SUSE, right?
13:25:31 <IgorYozhikov> I saw that commit was abandoned
13:26:16 <dirk> jpena: so those with %service_* and with %systemd_* should be available now in the suse ci, yes
13:26:30 <dirk> I'm confused which ones are the "upstream ones" and which we should be using going forward
13:27:19 <jpena> When I say upstream macros I'm referring to the ones in https://github.com/systemd/systemd/blob/master/src/core/macros.systemd.in
13:28:01 <jpena> I tried to dig a bit into the SUSE packaging, and I found they use different ones (the ones I was trying to emulate in my review)
13:29:56 <IgorYozhikov> dirk, does it means that I can revert your changes - https://review.openstack.org/#/c/382196/34/openstack/mistral/mistral.spec.j2 ?
13:30:48 <IgorYozhikov> to be more specific - https://review.openstack.org/#/c/382196/32..34/openstack/mistral/mistral.spec.j2
13:31:18 <IgorYozhikov> and systemd_post will work as in centos an in suse linux?
13:32:22 <jpena> that's what I expect
13:34:25 <IgorYozhikov> ok, I'll try to do that today and will see :)
13:35:42 <IgorYozhikov> anything else related to this topic? just want to be more || less sure that this will not block us with the rest of services.
13:37:28 <IgorYozhikov> #topic Xstatics* almost done - do we ready for horizon?
13:38:22 <IgorYozhikov> There is last of xstatics*, I believe, on review - https://review.openstack.org/#/c/410228/
13:38:54 <IgorYozhikov> Does it means that we can start work on horizon?
13:39:53 <IgorYozhikov> or there are another parts to be done before?
13:39:53 <dirk> jpena: IgorYozhikov : yep, so lets go with %systemd_ macros
13:40:11 <dirk> #agreed for service packaging please use the %systemd_* macros from https://github.com/systemd/systemd/blob/master/src/core/macros.systemd.in
13:40:24 <IgorYozhikov> dirk, yey
13:40:27 <IgorYozhikov> thanx
13:41:12 <dirk> IgorYozhikov: I am not sure regarding horizon, but its always worth putting up a review and see where it fails..
13:41:29 <dirk> I was actually planning to help a bit with that but currently there are too many fires
13:42:09 <IgorYozhikov> dirk, got it
13:43:52 <IgorYozhikov> ok, may be kaslcrof or tlbr could help with horizon, it' is not to urgent but will be very nice to have it in working state
13:44:27 <IgorYozhikov> or if any1 else want to take it(horizon) - I'm fine with it
13:44:41 <tlbr> yes, we can take a look if required
13:45:18 <IgorYozhikov> great, tlbr thanx a lot
13:45:28 <tlbr> np :)
13:45:31 <IgorYozhikov> moving forward?
13:45:41 <IgorYozhikov> #topic Open floor
13:46:40 <IgorYozhikov> dirk, I believe that you want to clarify rdo gate readiness/status?
13:48:09 <IgorYozhikov> I saw commit from Tristan https://review.openstack.org/#/c/393606/
13:49:10 <jpena> this was a test to verify if some of the gerrit problems we had were fixed
13:49:39 <jpena> right now, the CI job is mostly behaving, although we still have the networkx.drawing package issue
13:49:56 <jpena> we could set the CI as non-voting for now, at least it's giving me some useful feedback
13:50:20 <dirk> yeah
13:50:26 <dirk> I would really like to see some reports in the reviews
13:50:29 <IgorYozhikov> ah, that's nice, so it's moving
13:50:47 <jpena> ok, I'll go for it tomorrow then
13:50:59 <dirk> jpena: is there a way we can have networkx.drawing issue worked around for now?
13:51:39 <jpena> dirk: we have https://github.com/rdo-packages/taskflow-distgit/blob/rpm-master/python-taskflow.spec#L92-L93 in the RDO specs
13:52:13 <dirk> jpena: so put that up as an review under a %if is_rdo condition?
13:52:31 <jpena> dirk: works for me
13:52:38 <dirk> I'd prefer imperfect but working ci over no ci any time
13:52:59 <jpena> agreed, I'll go for it
13:53:12 <dirk> is it just building the package under review like the mos ci or does it rebuidl everything like the suse ci?
13:53:23 <jpena> only the package under review
13:53:37 <dirk> k, so the issue with networkx should be fairly isolated
13:54:00 <IgorYozhikov> I also have 2 updates from my side
13:54:22 <IgorYozhikov> 1. about keystone + updated passlib
13:54:45 <IgorYozhikov> I'll update keystone PR right after b2 will be released
13:55:01 <IgorYozhikov> because b1 is not compatible with passlib 1.7.0
13:55:26 <IgorYozhikov> already spoke with keystone developers
13:55:56 <IgorYozhikov> and 2nd - finally mos-ci add newton support
13:56:30 <IgorYozhikov> and now it rebuilds whole bunch of projects resided under stable/newton branch @ rpm-packaging
13:56:33 <IgorYozhikov> https://packaging-ci.fuel-infra.org/job/newton-rpm-packaging-build-centos7/
13:57:11 <IgorYozhikov> this means that mos-ci will reacts on commits into stable/newton as already reacts for mitaka & master
13:57:36 <IgorYozhikov> all issues will be fixed in a couple of days
13:58:01 <IgorYozhikov> that's all from my side
13:58:42 <dirk> thanks IgorYozhikov  for the update!
13:58:56 <dirk> last q : another meeting next week same time same place?
13:59:06 <dirk> or is everyone already off for end of the year vacation?
13:59:08 <IgorYozhikov> wfm
13:59:28 <IgorYozhikov> we have vacation right after NY
13:59:54 <IgorYozhikov> I know that in EU & US vacations before NY
14:01:03 <IgorYozhikov> let's discuss this at our channel. we r out of time
14:01:16 <IgorYozhikov> #endmeeting