13:01:32 #startmeeting rpm_packaging 13:01:33 Meeting started Thu Aug 18 13:01:32 2016 UTC and is due to finish in 60 minutes. The chair is toabctl. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:01:34 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:01:37 The meeting name has been set to 'rpm_packaging' 13:01:41 ping dirk toabctl IgorYozhikov number80 jruzicka 13:02:03 #chair dirk number80 IgorYozhikov jruzicka 13:02:04 Current chairs: IgorYozhikov dirk jruzicka number80 toabctl 13:02:46 so anybody here for a meeting? 13:03:12 please add you topics as usual to the agenda on the etherpad 13:03:14 #link https://etherpad.openstack.org/p/openstack-rpm-packaging 13:03:44 and please say hello if somebody is here. otherwise I'll jus end the meeting in some minutes again 13:04:22 o/ 13:04:48 o/ :) 13:06:00 o/ 13:06:01 ok. let's start 13:06:05 #topic https://etherpad.openstack.org/p/openstack-rpm-packaging-python3-renderspec 13:06:08 ^^ IgorYozhikov 13:06:46 yes, I posted some of my thoughts 13:06:49 tbh I'm not sure if we need to discuss it if we are just 3 persons. I would at least have an opinion from dirk and number80 13:07:35 toabctl, i'm fine with this, lets move this to further meeting 13:07:59 ok 13:08:04 Just want to make you know about my thoughts about py3 13:08:21 and what could we do 13:08:21 IgorYozhikov, we can just move the topic to the next meeting, ok? 13:08:25 sure 13:08:32 #topic osc-lib - bump to 1.0.2 13:08:54 o/ 13:08:57 I just saw that 1.0.2 was merged to U-C 13:09:19 Yeah 13:09:20 as === 1.0.2 - might be we can update https://review.openstack.org/#/c/355335/1/openstack/osc-lib/osc-lib.spec.j2? 13:09:34 IgorYozhikov, ok. I'll update the changeset 13:09:55 and while investigating fails in our CI I figured out that with MOS version of cliff - it passed fine 13:10:14 and failing when using cliff built from rpm-packaging 13:10:29 so I'm now in the middle of investigaion process 13:10:52 ok 13:10:56 #topic pending reviews https://review.openstack.org/#/q/project:openstack/rpm-packaging+status:open 13:11:01 when I'll find a root cause- I'll propose corresponding changes 13:11:14 IgorYozhikov, yes. sounds good 13:11:24 anything special in the current review list? 13:11:27 https://review.openstack.org/#/c/343335/ 13:12:10 ack 13:12:11 looks like it stuck ^ 13:13:21 Well, I don't mind using xargs but it doesn't play well with parameter expansion 13:13:25 dirk had a comment there. should we wait? I would also be fine with merging it. 13:13:55 well, we can wait 13:14:21 it's not critical, just wanted to run tools faster :) 13:14:25 :) 13:14:45 number80, there are some changeset with a -1. can you update them? or should somebody else takeover? 13:15:58 I'm going to check all other failures and update || backport missed depends 13:16:05 in our repos 13:16:24 number80: I wouldn't mind taking over some of your reviews if you're short on time 13:16:41 that should fix issues like DEBUG util.py:417: Error: No Package found for python-pifpaf >= 0.10.0 13:16:42 DEBUG util.py:417: Error: No Package found for python-pika_pool >= 0.1.3 13:17:17 O_o pifpaf 13:18:32 and here about naming - python-pika_pool in logs 13:18:47 and package is python-pika-pool-0.1.3-1.el7~mos1.src.rpm 13:19:06 jpena: I'd focus on uploading new packages 13:19:41 it's still manageable 13:19:42 does it means that I need to fix the name? 13:19:49 jpena, there are plenty of updates needed :) 13:19:51 #link http://toabctl.de/openstack/rpm-packaging-status-newton.html 13:20:13 I can take cliff 13:20:30 toabctl: ok, I'll start with some of them 13:20:31 because of working on it anyway 13:21:18 IgorYozhikov, the name is pika-pool . what requires pika_pool ? 13:21:56 toabctl, https://review.openstack.org/#/c/350429/6 13:22:21 https://packaging-ci.fuel-infra.org/job/master-rpm-packaging-build-centos7/544/artifact/artifacts/oslo.messaging.spec/*view*/ 13:22:56 guess it is from pymod2pkg transformation rule 13:23:17 oslo.messaging 13:23:47 IgorYozhikov, yes. It's from pymod2pkg . the oslo.messaging spec.j2 has the correct name 13:24:12 I know, i'm about resulting spec 13:24:54 does it means that it is require to update pymod2pkg? 13:25:20 IgorYozhikov, it's a RDO rule. not sure why you have different pkg names 13:25:48 IgorYozhikov: either we need to add MOS specific mapping or look at it. 13:26:14 If I remember pika_pool was the upstream name at some point and it got accepted as-is in Fedora 13:26:26 number80, I think that I need to fix our name to be close to rdo/fedora naming 13:26:35 and add provides /obsoletes 13:26:35 I guess, we can fix that package name 13:26:41 for compatibility 13:26:53 looks like jpena is the fedora maintainer ^ 13:27:48 number80: do you suggest doing the change in fedora? 13:28:57 jpena, I think that number80 just asking what is the right name here for fedora || centos package 13:29:20 jpena: upstream name is now pika-pool, maybe we should rename it 13:29:50 number80: mmm you're right. Ok, it's an easy change 13:30:20 jpena: just add me as a reviewer, and I'll approve it 13:30:42 we just need Provides/Obsoletes: python-pika_pool 13:31:01 number80, jpena - I add additional provides to mos pika-pool package to handle both variants 13:32:12 yeah, but rename would be ok in this case 13:32:28 * number80 suggests to discuss this on our channel post-meeting 13:32:36 so, and after that we could update pymod2pkg? 13:32:41 yes 13:32:42 ok 13:32:55 jruzicka, I commented on https://review.openstack.org/#/c/347447 but in general lgtm. I'm fine with adding this change. but we need to think about how we want to continue with renderspec... 13:33:05 yes 13:33:16 any other change that needs to be discussed? 13:33:54 https://review.openstack.org/#/c/339458/ 13:34:59 IgorYozhikov, ah. the sed from the spec.j2 removes the line https://github.com/openstack/taskflow/blob/master/taskflow/types/graph.py#L21 13:35:08 that leads to the import error and the build failure. 13:35:17 number80, can we just remove the sed? is that really needed? 13:36:44 * number80 looks 13:37:24 we can remove it, but it pulls quite a lot of deps 13:37:47 was like 600 to 800 Mb for a one-liner function 13:38:21 hm 13:39:16 I tried to convince upstream removing it but though it's unused, they want to keep it forever ... 13:39:20 whoa.... 13:39:49 whatever, let's remove it 13:41:23 ok 13:41:33 anything else in the queue? 13:42:14 toabctl, no from my side 13:42:26 toabctl, how do we want to continue with renderspec? 13:42:36 I mean, what are the options? 13:42:41 going to fix of missed and continue with cliff 13:43:09 #topic open discussion 13:43:14 anything else? 13:43:53 jruzicka, what do you mean ? 13:44:07 features? 13:44:09 jruzicka, ah. missed that question. sorry 13:44:48 we can discuss post-meeting 13:45:12 jruzicka, ok. let's move to #openstack-rpm-packaging 13:45:17 thanks everybody 13:45:19 #endmeeting