| *** ooolpbot has joined #tripleo | 00:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 00:10 |
|---|---|---|
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 00:10 |
| *** ooolpbot has quit IRC | 00:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 00:10 |
| *** MaxPC has quit IRC | 00:16 | |
| *** Goneri has quit IRC | 00:16 | |
| *** lblanchard has quit IRC | 00:21 | |
| *** lblanchard has joined #tripleo | 00:33 | |
| *** MaxPC has joined #tripleo | 00:35 | |
| *** Goneri has joined #tripleo | 00:35 | |
| *** irdkehn has joined #tripleo | 01:02 | |
| *** irdkehn is now known as dkehn | 01:05 | |
| *** ooolpbot has joined #tripleo | 01:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 01:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 01:10 |
| *** ooolpbot has quit IRC | 01:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 01:10 |
| *** MaxPC has quit IRC | 01:36 | |
| *** r-mibu has quit IRC | 02:09 | |
| *** ooolpbot has joined #tripleo | 02:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 02:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 02:10 |
| *** ooolpbot has quit IRC | 02:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 02:10 |
| *** r-mibu has joined #tripleo | 02:14 | |
| *** ebalduf has joined #tripleo | 02:23 | |
| *** lblanchard has quit IRC | 02:23 | |
| *** ebalduf has quit IRC | 02:26 | |
| *** MaxPC has joined #tripleo | 02:34 | |
| *** ebalduf has joined #tripleo | 02:41 | |
| *** yuanying_ has quit IRC | 02:51 | |
| *** MaxPC has quit IRC | 02:52 | |
| *** ccamacho has quit IRC | 02:55 | |
| *** ccamacho has joined #tripleo | 03:02 | |
| *** ooolpbot has joined #tripleo | 03:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 03:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 03:10 |
| *** ooolpbot has quit IRC | 03:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 03:10 |
| *** dsariel has quit IRC | 03:32 | |
| *** morazi has quit IRC | 03:44 | |
| *** yuanying has joined #tripleo | 03:46 | |
| *** coolsvap has joined #tripleo | 04:07 | |
| *** ooolpbot has joined #tripleo | 04:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 04:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 04:10 |
| *** ooolpbot has quit IRC | 04:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 04:10 |
| *** links has joined #tripleo | 04:41 | |
| *** saneax_AFK is now known as saneax | 04:49 | |
| *** oshvartz has quit IRC | 04:50 | |
| *** rcernin has joined #tripleo | 05:03 | |
| *** ooolpbot has joined #tripleo | 05:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 05:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 05:10 |
| *** ooolpbot has quit IRC | 05:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 05:10 |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: yum-minimal: strip locale archive https://review.openstack.org/304399 | 05:13 |
| *** ramishra has joined #tripleo | 05:21 | |
| *** ramishra has quit IRC | 05:28 | |
| *** ramishra has joined #tripleo | 05:30 | |
| *** oshvartz has joined #tripleo | 05:50 | |
| *** jprovazn has joined #tripleo | 05:50 | |
| *** ramishra has quit IRC | 05:52 | |
| *** ramishra has joined #tripleo | 05:55 | |
| *** florianf has joined #tripleo | 05:59 | |
| *** shadower has joined #tripleo | 06:08 | |
| *** ooolpbot has joined #tripleo | 06:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 06:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 06:10 |
| *** ooolpbot has quit IRC | 06:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 06:10 |
| *** shivrao has joined #tripleo | 06:23 | |
| *** jaosorior has joined #tripleo | 06:24 | |
| *** leanderthal|afk is now known as leanderthal | 06:26 | |
| *** jtomasek has joined #tripleo | 06:33 | |
| *** bvandenh has quit IRC | 06:49 | |
| *** dsariel has joined #tripleo | 07:02 | |
| *** liverpooler has joined #tripleo | 07:03 | |
| *** bvandenh has joined #tripleo | 07:06 | |
| *** shivrao has quit IRC | 07:08 | |
| *** ooolpbot has joined #tripleo | 07:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 07:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 07:10 |
| *** ooolpbot has quit IRC | 07:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 07:10 |
| *** jaosorior has quit IRC | 07:11 | |
| *** dtantsur|afk is now known as dtantsur | 07:14 | |
| *** hewbrocca-afk is now known as hewbrocca | 07:15 | |
| *** pcaruana has joined #tripleo | 07:15 | |
| *** mbound has joined #tripleo | 07:17 | |
| *** ccamacho has quit IRC | 07:18 | |
| *** ebalduf has quit IRC | 07:19 | |
| *** bvandenh has quit IRC | 07:20 | |
| *** ccamacho has joined #tripleo | 07:22 | |
| *** ohamada has joined #tripleo | 07:25 | |
| *** veteran has joined #tripleo | 07:25 | |
| *** veteran has quit IRC | 07:25 | |
| *** tremble has joined #tripleo | 07:31 | |
| *** tremble has joined #tripleo | 07:31 | |
| *** ifarkas has joined #tripleo | 07:32 | |
| *** shadower has quit IRC | 07:32 | |
| *** thomas_ has joined #tripleo | 07:35 | |
| *** thomas_ is now known as shadower | 07:35 | |
| *** bvandenh has joined #tripleo | 07:39 | |
| *** mikelk has joined #tripleo | 07:44 | |
| *** stendulker has joined #tripleo | 07:44 | |
| *** aufi has joined #tripleo | 07:49 | |
| *** jaosorior has joined #tripleo | 07:55 | |
| *** jaosorior has quit IRC | 07:56 | |
| *** numans_ has joined #tripleo | 07:57 | |
| *** shardy has joined #tripleo | 08:06 | |
| *** athomas has joined #tripleo | 08:07 | |
| *** ooolpbot has joined #tripleo | 08:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 08:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 08:10 |
| *** ooolpbot has quit IRC | 08:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 08:10 |
| *** hjensas has quit IRC | 08:11 | |
| *** sshnaidm has quit IRC | 08:12 | |
| *** mcornea has joined #tripleo | 08:21 | |
| *** jistr has joined #tripleo | 08:21 | |
| *** gfidente has joined #tripleo | 08:31 | |
| *** bvandenh has quit IRC | 08:37 | |
| *** bvandenh has joined #tripleo | 08:40 | |
| *** derekh has joined #tripleo | 08:40 | |
| *** ohamada has quit IRC | 08:41 | |
| *** ohamada has joined #tripleo | 08:41 | |
| derekh | So infra have blocked access to the jenkins servers http://git.openstack.org/cgit/openstack-infra/system-config/commit/?id=6d99934ff0e9a66531327c86c23e2cd88d5d2ccc | 08:42 |
| derekh | as a result our status page no longer works ;-( | 08:42 |
| derekh | anybody have any ideas for an alternative? querying gerrit maybe | 08:43 |
| *** ohamada has quit IRC | 08:44 | |
| *** ohamada has joined #tripleo | 08:45 | |
| *** hjensas has joined #tripleo | 08:48 | |
| *** hjensas has quit IRC | 08:48 | |
| *** hjensas has joined #tripleo | 08:48 | |
| *** paramite has joined #tripleo | 08:48 | |
| *** jpena|off is now known as jpena | 08:49 | |
| *** hjensas has quit IRC | 08:50 | |
| *** hjensas has joined #tripleo | 08:50 | |
| *** ohamada has quit IRC | 08:51 | |
| *** mgould has joined #tripleo | 08:51 | |
| *** ohamada has joined #tripleo | 08:51 | |
| *** zoliXXLz is now known as zoli|BRB | 08:52 | |
| jistr | derekh: an alternative might be zuul's smtp reporter http://docs.openstack.org/infra/zuul/reporters.html | 08:59 |
| *** electrofelix has joined #tripleo | 09:01 | |
| derekh | jistr: could be an option, although querying gerrit might be easier then collecting and parsing emails | 09:03 |
| *** tesseract has joined #tripleo | 09:03 | |
| *** tesseract is now known as Guest30394 | 09:03 | |
| *** jistr has quit IRC | 09:05 | |
| *** ooolpbot has joined #tripleo | 09:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 09:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 09:10 |
| *** ooolpbot has quit IRC | 09:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 09:10 |
| openstackgerrit | Jiri Stransky proposed openstack/tripleo-heat-templates: Fix ControllerExtraConfig parameter name https://review.openstack.org/306522 | 09:11 |
| *** jistr has joined #tripleo | 09:12 | |
| ccamacho | derekh, I did some gerrit queries before, let me check what I can get | 09:15 |
| *** olap has joined #tripleo | 09:15 | |
| derekh | ccamacho: ok | 09:15 |
| *** mcornea has quit IRC | 09:17 | |
| *** mcornea has joined #tripleo | 09:19 | |
| *** akrivoka has joined #tripleo | 09:23 | |
| *** sshnaidm has joined #tripleo | 09:29 | |
| *** zoli|BRB is now known as zoliXXL | 09:29 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add ceph profiles https://review.openstack.org/304675 | 09:52 |
| *** athomas has quit IRC | 09:54 | |
| ccamacho | guys, quick question (good morning) | 09:54 |
| ccamacho | the data from this site http://tripleo.org/cistatus.html where is hosted? | 09:55 |
| *** ohamada has quit IRC | 09:56 | |
| *** ohamada has joined #tripleo | 09:56 | |
| *** tosky has joined #tripleo | 09:57 | |
| *** athomas has joined #tripleo | 10:04 | |
| *** coolsvap has quit IRC | 10:05 | |
| *** athomas has quit IRC | 10:06 | |
| derekh | ccamacho: its a VM dprince spun up somewhere, looking for the database? | 10:08 |
| derekh | ccamacho: this is the code that generates the report http://git.openstack.org/cgit/openstack-infra/tripleo-ci/tree/scripts/tripleo-jobs.py | 10:08 |
| *** ooolpbot has joined #tripleo | 10:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 10:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 10:10 |
| *** ooolpbot has quit IRC | 10:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 10:10 |
| ccamacho | derekh++ that's what I needed, I will make the changes on that file by querying gerrit instead | 10:10 |
| ccamacho | thanks! | 10:10 |
| *** athomas has joined #tripleo | 10:10 | |
| *** myoung has joined #tripleo | 10:11 | |
| derekh | ccamacho: ok, let me first ask on infra to see if the changes to the jenkins rules are perminant | 10:13 |
| ccamacho | np I will un-dust some some pygerrit scripts from my HD in meantime | 10:14 |
| *** olap has quit IRC | 10:21 | |
| derekh | ccamacho: could be a few weeks http://lists.openstack.org/pipermail/openstack-dev/2016-April/092482.html | 10:22 |
| ccamacho | derekh, "temporary loss of functionality" I see, well if needed just let me know. | 10:23 |
| *** olap has joined #tripleo | 10:23 | |
| derekh | ccamacho: will do, will see later today if anybody has other ideas | 10:24 |
| dtantsur | hey folks, shardy. could you give some response to the backport request http://lists.openstack.org/pipermail/openstack-dev/2016-April/091523.html please? | 10:28 |
| shardy | dtantsur: sure, looking | 10:41 |
| shardy | dtantsur: +1, I think we can arguably consider it a bug, e.g choosing the wrong/different device after upgrade is a bug from a user perspective | 10:43 |
| shardy | I'll reply to the ML now | 10:43 |
| *** zoliXXL is now known as zoli|lunch | 10:44 | |
| dtantsur | thnx | 10:44 |
| openstackgerrit | Miles Gould proposed openstack/tripleo-quickstart: Typo fix https://review.openstack.org/307086 | 10:46 |
| *** michchap_ has joined #tripleo | 11:01 | |
| *** SlickN1k has joined #tripleo | 11:01 | |
| *** marcusvrn__ has joined #tripleo | 11:02 | |
| *** stendulker_ has joined #tripleo | 11:02 | |
| *** pino|work_ has joined #tripleo | 11:02 | |
| *** rhefner_ has joined #tripleo | 11:03 | |
| *** igorbelikov_ has joined #tripleo | 11:04 | |
| *** beekneemech has joined #tripleo | 11:04 | |
| *** krotscheck_ has joined #tripleo | 11:04 | |
| *** CaptTofu_ has joined #tripleo | 11:04 | |
| *** tristanC_ has joined #tripleo | 11:04 | |
| *** rodrigods has quit IRC | 11:04 | |
| *** jaosorior has joined #tripleo | 11:05 | |
| *** rodrigods has joined #tripleo | 11:05 | |
| *** cinerama` has joined #tripleo | 11:05 | |
| *** Ng has quit IRC | 11:05 | |
| *** rhefner has quit IRC | 11:05 | |
| *** marcusvrn_ has quit IRC | 11:05 | |
| *** SlickNik has quit IRC | 11:05 | |
| *** tdasilva has quit IRC | 11:05 | |
| *** igorbelikov has quit IRC | 11:05 | |
| *** CaptTofu has quit IRC | 11:05 | |
| *** therve has quit IRC | 11:05 | |
| *** spredzy has quit IRC | 11:06 | |
| *** telnoratti has quit IRC | 11:06 | |
| *** krotscheck has quit IRC | 11:06 | |
| *** dtantsur has quit IRC | 11:06 | |
| *** ianw has quit IRC | 11:06 | |
| *** cinerama has quit IRC | 11:06 | |
| *** pino|work has quit IRC | 11:06 | |
| *** bnemec has quit IRC | 11:06 | |
| *** tristanC has quit IRC | 11:06 | |
| *** snecklifter has quit IRC | 11:06 | |
| *** mburned has quit IRC | 11:06 | |
| *** michchap has quit IRC | 11:06 | |
| *** SlickN1k is now known as SlickNik | 11:06 | |
| *** stendulker has quit IRC | 11:06 | |
| *** snecklifter has joined #tripleo | 11:06 | |
| *** pino|work_ is now known as pino|work | 11:06 | |
| *** rhefner_ is now known as rhefner | 11:06 | |
| *** therve has joined #tripleo | 11:06 | |
| *** krotscheck_ is now known as krotscheck | 11:06 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: [NO MERGY] Test a fake periodic job https://review.openstack.org/229789 | 11:06 |
| *** marcusvrn__ is now known as marcusvrn_ | 11:07 | |
| *** mburned has joined #tripleo | 11:07 | |
| *** igorbelikov_ is now known as igorbelikov | 11:07 | |
| *** ianw has joined #tripleo | 11:08 | |
| *** stendulker_ has quit IRC | 11:09 | |
| *** spredzy has joined #tripleo | 11:09 | |
| *** dtantsur has joined #tripleo | 11:09 | |
| *** tosky_ has joined #tripleo | 11:10 | |
| *** ooolpbot has joined #tripleo | 11:10 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 11:10 |
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1571090 | 11:10 |
| *** ooolpbot has quit IRC | 11:10 | |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Triaged] | 11:10 |
| *** pblaho has quit IRC | 11:10 | |
| *** CaptTofu_ has quit IRC | 11:10 | |
| *** Ng has joined #tripleo | 11:11 | |
| *** ChanServ sets mode: +v Ng | 11:11 | |
| *** pblaho has joined #tripleo | 11:11 | |
| *** tosky has quit IRC | 11:12 | |
| *** rcernin has quit IRC | 11:12 | |
| *** olap has quit IRC | 11:12 | |
| *** egafford has joined #tripleo | 11:13 | |
| *** olap has joined #tripleo | 11:14 | |
| *** telnoratti has joined #tripleo | 11:18 | |
| *** tdasilva has joined #tripleo | 11:18 | |
| *** rhallisey has joined #tripleo | 11:22 | |
| *** rcernin has joined #tripleo | 11:23 | |
| *** coolsvap has joined #tripleo | 11:23 | |
| *** CaptTofu_ has joined #tripleo | 11:24 | |
| *** dprince has joined #tripleo | 11:28 | |
| *** MaxPC has joined #tripleo | 11:31 | |
| *** liverpooler has quit IRC | 11:42 | |
| *** ccamacho is now known as ccamacho|lunch | 11:43 | |
| *** lucasagomes is now known as lucas-hungry | 11:43 | |
| jaosorior | How long does it take for a package that should be available in RDO to reflect in our CI? | 11:51 |
| jaosorior | The CI still seems broken, although the fix has been released https://bugs.launchpad.net/tripleo/+bug/1571090 | 11:51 |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Fix released] | 11:51 |
| *** athomas has quit IRC | 11:52 | |
| *** weshay has joined #tripleo | 11:56 | |
| *** mbound has quit IRC | 12:00 | |
| *** pkovar has joined #tripleo | 12:04 | |
| *** trown|outtypewww is now known as trown | 12:09 | |
| trown | jaosorior: it should be immediate, but there might be some caching? derekh ^ | 12:09 |
| trown | derekh: I confirmed that the bad galera package is removed from the deps repo, do we need to do anything in tripleoci? | 12:10 |
| *** apetrich has joined #tripleo | 12:14 | |
| jaosorior | trown: Alright, will wait then, thanks dude | 12:15 |
| *** ryansb has left #tripleo | 12:18 | |
| *** saneax is now known as saneax_AFK | 12:29 | |
| *** aufi has quit IRC | 12:29 | |
| *** julim has joined #tripleo | 12:30 | |
| *** mbound has joined #tripleo | 12:30 | |
| *** coolsvap has quit IRC | 12:30 | |
| *** aufi has joined #tripleo | 12:31 | |
| *** rbrady has joined #tripleo | 12:33 | |
| *** jpena is now known as jpena|lunch | 12:36 | |
| *** dsariel has quit IRC | 12:42 | |
| *** lucas-hungry is now known as lucasagomes | 12:44 | |
| *** hewbrocca is now known as hewbrocca-afk | 12:44 | |
| *** jdob has joined #tripleo | 12:46 | |
| sshnaidm | does anybody know why http://tripleo.org/cistatus.html is stuck on 04-15 ? | 12:48 |
| trown | sshnaidm: we no longer have access to jenkins :( | 12:49 |
| sshnaidm | trown, oh, that's why jenkins page doesn't open | 12:49 |
| *** rlandy has joined #tripleo | 12:49 | |
| sshnaidm | trown, but this is from zuul, isn't it? | 12:49 |
| trown | sshnaidm: nope it polls jenkins | 12:50 |
| trown | https://github.com/openstack-infra/tripleo-ci/blob/master/scripts/tripleo-jobs.py#L49 | 12:50 |
| sshnaidm | trown, maybe it's possible to use https://zuul.openstack.org/ | 12:52 |
| sshnaidm | trown, is it forever? or just temporary block? | 12:52 |
| trown | sshnaidm: seems at least a few weeks http://lists.openstack.org/pipermail/openstack-dev/2016-April/091523.html | 12:53 |
| trown | so any ideas are welcome I think :) | 12:53 |
| shardy | https://review.openstack.org/#/c/289082/ looks OK to land if anyone has a moment | 12:53 |
| shardy | related to fixing the containers CI job | 12:53 |
| *** morazi has joined #tripleo | 12:54 | |
| sshnaidm | trown, sorry, isn't it wrong link? it's about ironic | 12:55 |
| *** pradk has joined #tripleo | 12:55 | |
| trown | sshnaidm: lol, ya sorry wrong link http://lists.openstack.org/pipermail/openstack-dev/2016-April/092482.html | 12:55 |
| *** ccamacho|lunch is now known as ccamacho | 12:55 | |
| *** zoli|lunch is now known as zoli_brb | 12:56 | |
| sshnaidm | trown, yeah.. it's about this csrf problem :( | 12:56 |
| *** dprince has quit IRC | 12:57 | |
| sshnaidm | though I hardly understand why it should take a few weeks(!), it's about a few days work.. | 12:58 |
| *** apetrich has quit IRC | 12:59 | |
| openstackgerrit | Merged openstack/tripleo-common: Missing python-ipaddr in heat-agents image https://review.openstack.org/289082 | 12:59 |
| *** tzumainn has joined #tripleo | 13:02 | |
| *** olap has quit IRC | 13:05 | |
| *** olap has joined #tripleo | 13:06 | |
| *** trozet has joined #tripleo | 13:11 | |
| *** trozet has quit IRC | 13:14 | |
| *** trozet has joined #tripleo | 13:15 | |
| derekh | trown: there is caching but repodata should be bypassed for the cache, what repo are we wondering about ? | 13:16 |
| trown | derekh: the delorean-deps.repo | 13:16 |
| trown | it might be resolved, waiting to see an HA pass to know for sure though | 13:17 |
| trown | which is trickier without our status page | 13:17 |
| openstackgerrit | mathieu bultel proposed openstack/tripleo-quickstart: Set default tht parameters for net-iso and remove hardcoded values https://review.openstack.org/305119 | 13:17 |
| tremble | sshnaidm: "it's about a few days work" <- that comment makes so many assumptions which rarely hold true in the real world. People tend to get rather annoyed when you turn on a security feature and break random little things all over the place. | 13:18 |
| *** dtantsur is now known as dtantsur|brb | 13:19 | |
| trown | matbu: ah I see now, ansible was not including quotes there | 13:19 |
| matbu | trown: yes | 13:19 |
| matbu | trown: the patchset #5 should fixed it | 13:20 |
| *** tosky_ is now known as tosky | 13:23 | |
| derekh | trown: accoring to the proxy server that last download of the galera 25.3.12 package was just over 3 hours ago | 13:24 |
| *** egafford has quit IRC | 13:24 | |
| trown | derekh: that is about when it was fixed, so we should be good | 13:24 |
| derekh | trown: ok | 13:25 |
| jaosorior | yay | 13:26 |
| *** thrash has quit IRC | 13:29 | |
| trown | ahhhh.... so close https://review.openstack.org/229789 | 13:30 |
| *** shadower is now known as shadower_ | 13:31 | |
| *** shadower_ is now known as shadower | 13:32 | |
| *** shadower is now known as shadower_ | 13:32 | |
| *** paramite is now known as paramite|afk | 13:32 | |
| *** shadower_ is now known as shadower | 13:33 | |
| *** thrash has joined #tripleo | 13:33 | |
| *** thrash has joined #tripleo | 13:33 | |
| derekh | trown: it should promote | 13:35 |
| trown | derekh: because it passed all 3 on the same PS? | 13:35 |
| trown | even though they are on different rechecks? | 13:36 |
| derekh | trown: the last 2 runs were with the same repo https://review.openstack.org/#/c/229789/41/toci_gate_test.sh | 13:36 |
| trown | cool | 13:36 |
| derekh | trown: once the mirror server registers a pass for each job for the same repo, it should promote | 13:36 |
| derekh | trown: cron job is due to trigger in 4 minutes | 13:36 |
| *** Goneri has quit IRC | 13:37 | |
| *** numans_ has quit IRC | 13:38 | |
| *** egafford has joined #tripleo | 13:40 | |
| adarazs | derekh: last week I was quite busy, does it still make sense to look into the disk utilization as we discussed it on the previous Friday? | 13:40 |
| *** lblanchard has joined #tripleo | 13:40 | |
| *** dustins has joined #tripleo | 13:42 | |
| derekh | trown: updated http://trunk.rdoproject.org/centos7/current-tripleo/ | 13:44 |
| *** zoli_brb is now known as zoliXXL | 13:44 | |
| trown | derekh: nice | 13:44 |
| trown | derekh: did we get cached IPA? | 13:45 |
| openstackgerrit | Florian Fuchs proposed openstack/tripleo-ui: Check for ongoing deployment when opening DeploymentConfiguration https://review.openstack.org/307192 | 13:45 |
| derekh | trown: yup | 13:45 |
| trown | awesome | 13:45 |
| derekh | trown: just should start using it now, if I didn't screw something up | 13:46 |
| derekh | adarazs: looks like we're going ahead with the SSD's anyway so it probably wont change anything | 13:47 |
| adarazs | derekh: all right :) | 13:48 |
| openstackgerrit | Florian Fuchs proposed openstack/tripleo-ui: Check for ongoing deployment when opening DeploymentConfiguration https://review.openstack.org/307192 | 13:48 |
| *** Goneri has joined #tripleo | 13:49 | |
| *** jpena|lunch is now known as jpena | 13:49 | |
| *** saneax_AFK is now known as saneax | 13:51 | |
| derekh | trown: actually sorry, that was wronge, I left the patch to use the IPA image in WIP so we could test befor merging, testing now https://review.openstack.org/#/c/301699/2 | 13:52 |
| derekh | also if anybody has a chance to look at this patch to start caching the overcloud image , https://review.openstack.org/#/c/306499/ | 13:53 |
| *** derekh is now known as derekh_afk | 13:53 | |
| *** psanchez has quit IRC | 13:55 | |
| *** psanchez has joined #tripleo | 13:56 | |
| *** dkehn_ has joined #tripleo | 13:57 | |
| *** links has quit IRC | 13:57 | |
| *** paramite|afk is now known as paramite | 13:59 | |
| *** ebalduf has joined #tripleo | 14:16 | |
| jaosorior | trown: Hey dude, any idea if there's something wrong with http://artifacts.ci.centos.org/artifacts/rdo/images/mitaka/delorean/stable/undercloud.qcow2 ? Been trying to fetch the image through tripleo-quickstart but it's failing :/ | 14:21 |
| trown | jaosorior: hmm. no issue for me, but we have a new mirrored image location, mind trying that with '-u http://buildlogs.centos.org/centos/7/cloud/x86_64/tripleo_images/mitaka/delorean/undercloud.qcow2' | 14:23 |
| *** aufi has quit IRC | 14:24 | |
| trown | jaosorior: it is on my list to patch that into tripleo-quickstart as the default | 14:24 |
| jaosorior | trown: Will do | 14:24 |
| trown | my internet connection originates from the same zipcode as the artifacts server, so I rarely have issues (and can rarely reproduce others issues) | 14:25 |
| jaosorior | trown: Funky... cancelled the other run | 14:27 |
| jaosorior | ran it witht he command you provided | 14:27 |
| jaosorior | and there is no undercloud set up (not even in libvirt) | 14:27 |
| jaosorior | trown: Where is the undercloud image downlaoded to? | 14:28 |
| jaosorior | it seems to have even failed the checksum check | 14:28 |
| *** olap has quit IRC | 14:30 | |
| *** dsariel has joined #tripleo | 14:31 | |
| trown | jaosorior: hmm, I will do a test with the buildlogs url | 14:31 |
| jaosorior | trown: Sure. I have no clue what's up, cause if I do manually wget with that url, in the target machine, it actually starts building it | 14:31 |
| jaosorior | I mean, fetching it | 14:32 |
| jaosorior | maybe I could do that, just download the image in the cache folder | 14:32 |
| jaosorior | trown: Is the stack user recreated every time? | 14:33 |
| trown | jaosorior: ya the "-u" argument can take a 'file://' url, so if you download it to anywhere world-readable on the host you can pass that location as url | 14:33 |
| trown | jaosorior: ya... need to fix that https://bugs.launchpad.net/tripleo-quickstart/+bug/1570289 | 14:33 |
| openstack | Launchpad bug 1570289 in tripleo-quickstart "Use a more explicit method for teardown" [High,Triaged] | 14:33 |
| jaosorior | I see | 14:34 |
| *** jdob has quit IRC | 14:36 | |
| *** jdob has joined #tripleo | 14:36 | |
| *** saneax is now known as saneax_AFK | 14:41 | |
| *** dmacpher has joined #tripleo | 14:41 | |
| *** saneax_AFK is now known as saneax | 14:42 | |
| *** dmsimard has quit IRC | 14:42 | |
| *** dmsimard has joined #tripleo | 14:43 | |
| *** jdob has quit IRC | 14:48 | |
| *** jdob has joined #tripleo | 14:48 | |
| *** rbrady has quit IRC | 14:48 | |
| *** rbrady has joined #tripleo | 14:49 | |
| *** sshnaidm has quit IRC | 14:58 | |
| *** oshvartz has quit IRC | 15:04 | |
| *** leanderthal is now known as leanderthal|afk | 15:06 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-tripleoclient: Updated from global requirements https://review.openstack.org/300044 | 15:07 |
| *** cinerama` is now known as cinerama | 15:07 | |
| *** dtantsur|brb is now known as dtantsur | 15:07 | |
| *** julim has quit IRC | 15:11 | |
| *** hjensas has quit IRC | 15:13 | |
| *** julim has joined #tripleo | 15:15 | |
| *** links has joined #tripleo | 15:19 | |
| *** shadower has quit IRC | 15:27 | |
| *** dtrainor has joined #tripleo | 15:30 | |
| *** olap has joined #tripleo | 15:31 | |
| *** dtrainor has quit IRC | 15:32 | |
| *** dtrainor has joined #tripleo | 15:33 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/python-tripleoclient: Check Keystone init before running it https://review.openstack.org/306333 | 15:35 |
| *** shadower has joined #tripleo | 15:36 | |
| *** aufi has joined #tripleo | 15:38 | |
| *** egafford has quit IRC | 15:38 | |
| *** egafford has joined #tripleo | 15:38 | |
| *** tremble has quit IRC | 15:39 | |
| shardy | EmilienM: Hi, I'm looking for some help with an undercloud puppet-keystone issue when you get a moment | 15:41 |
| shardy | ref https://bugs.launchpad.net/tripleo/+bug/1571708 | 15:41 |
| openstack | Launchpad bug 1571708 in tripleo "undercloud user role assignments deleted by undercloud install" [Undecided,New] | 15:41 |
| shardy | Which I think is related to https://bugzilla.redhat.com/show_bug.cgi?id=1325475 | 15:43 |
| openstack | bugzilla.redhat.com bug 1325475 in rhel-osp-director "rhel-osp-director: upgrade 7.3->8.0, that follows update 7.2->7.3, times out due to os-collect-config auth failure" [High,New] - Assigned to therve | 15:43 |
| EmilienM | shardy: looking | 15:43 |
| shardy | EmilienM: basically we delete all user role assignments when re-running the undercloud install | 15:43 |
| shardy | I'm wondering how we stop that, as it appears it strips the _member_ role for older environments, which breaks heat | 15:43 |
| *** cwolferh has joined #tripleo | 15:45 | |
| EmilienM | shardy: I suspect something else but puppet (python-tripleoclient?) manages keystone resources (roles, assignments, etc). | 15:46 |
| pkovar | am i blind or are the triple docs not featured on http://docs.openstack.org/developer/openstack-projects.html ? | 15:46 |
| EmilienM | shardy: if that's the case, it makes sense Puppet will clean it up | 15:46 |
| pkovar | * tripleo | 15:46 |
| EmilienM | shardy: we declare keystone::roles::admin class, which takes care of admin resources | 15:46 |
| jaosorior | EmilienM: Currently it does for the overcloud deployment, but I believe shardy was referring to the undercloud install | 15:46 |
| EmilienM | jaosorior: yeah, I mean for undercloud | 15:47 |
| jaosorior | which python-tripleoclient shouldn't be doing anything for | 15:47 |
| shardy | EmilienM: since https://review.openstack.org/#/q/Iad619d3b10e8329234bdfb3fc738ed130e26fe3f it looks like we manage keystone via puppet on the undercloud | 15:47 |
| jaosorior | regarding keystone endpoints and roles | 15:47 |
| * shardy checks this includes roles | 15:47 | |
| *** alhernan has joined #tripleo | 15:47 | |
| EmilienM | right, but are we sure puppet is the only one thing managing admin resources? | 15:47 |
| shardy | Personally I don't think it's reasonable to overwrite operator additions when re-asserting service configuration like this | 15:47 |
| EmilienM | all providers in puppet-keystone are idempotent | 15:47 |
| EmilienM | this is the role: https://github.com/openstack/puppet-keystone/blob/master/manifests/roles/admin.pp | 15:48 |
| *** dprince has joined #tripleo | 15:50 | |
| EmilienM | shardy: I disagree, if we want puppet to manage a resource, nothing has to be manual. | 15:50 |
| EmilienM | shardy: it's the purpose of automation with puppet (or ansible or chef): to not do manual actions. Any manual actions with those tools, will break things | 15:50 |
| *** jobewan has joined #tripleo | 15:51 | |
| EmilienM | it's like modifying nova.conf by hand, puppet will override it at next run | 15:51 |
| shardy | EmilienM: Yeah, but this is a pitfall of managing stuff like keystone users via puppet | 15:51 |
| shardy | it's not service configuration, it's content of those services | 15:51 |
| EmilienM | shardy: so don't manage them via puppet | 15:51 |
| shardy | EmilienM: Sure, but we are managing them via puppet, and that means everything breaks on upgrade | 15:52 |
| shardy | so, we either figure out how to fix that, or don't manage the users via puppet anymore | 15:52 |
| *** jobewan has quit IRC | 15:52 | |
| EmilienM | why does it break on upgrade? | 15:52 |
| beagles | pkovar: you aren't blind - the only mention of tripleo at all is os-cloud-config... what's up with that? shardy, know you're busy with keystone issue, but when you get a sec - is there a reason for it not being there, or is it just "we haven't added yet"? | 15:52 |
| EmilienM | if there is a change in the resource, let's make the change happen in the puppet manifest | 15:52 |
| *** paramite has quit IRC | 15:52 | |
| shardy | EmilienM: because we remove the _member_ role from all users on the undercloud | 15:52 |
| shardy | which breaks heat, because we then try to delegate a non-existent role assignment via a trust | 15:53 |
| *** jobewan has joined #tripleo | 15:53 | |
| *** sshnaidm has joined #tripleo | 15:53 | |
| shardy | I'm trying to figure out exactly where/why we delete those role assignments | 15:53 |
| EmilienM | shardy: I can patch undercloud to fix that | 15:53 |
| EmilienM | with a Puppet collector | 15:53 |
| jaosorior | isn't the _member_ role quite standard in openstack deployments? Why would it be deleted? O_o\ | 15:53 |
| *** chem has joined #tripleo | 15:55 | |
| EmilienM | jaosorior: let me look that, there are different things right now | 15:55 |
| shardy | jaosorior: It's no longer used as part of the move to keystone v3 ref https://bugs.launchpad.net/keystone/+bug/1366133 | 15:55 |
| openstack | Launchpad bug 1366133 in OpenStack Identity (keystone) "User create via v3 API doesn't add _member_ role in default project" [Undecided,Won't fix] | 15:55 |
| EmilienM | chem: jaosorior is asking why puppet-keystone does not add _member_ role to admin user | 15:55 |
| shardy | jaosorior: however we have an issue in that we have overclouds created prior to the move to v3 | 15:55 |
| shardy | hence _member_ stil exists in the trusts stored inside the undercloud heat | 15:55 |
| EmilienM | ok, so you want the admin user with _member_ role, right? | 15:55 |
| shardy | EmilienM: Yeah, the admin user had the _member_ role before upgrade, and removing it breaks the delegation made to heat | 15:56 |
| shardy | for brand new deployments, not having it would be fine | 15:56 |
| *** mikelk has quit IRC | 15:56 | |
| shardy | but we cannot remove any existing roles from any user on upgrade | 15:56 |
| EmilienM | shardy: because we don't use keystone v3 right? | 15:57 |
| shardy | because heat by default delegates all roles from the user to heat | 15:57 |
| EmilienM | ok got it | 15:57 |
| EmilienM | shardy: let me 2 min | 15:57 |
| shardy | EmilienM: No, the problem is that even after moving everything to v3, heat still has a trust which says "delegate _member_ and admin roles to user "heat" | 15:57 |
| shardy | EmilienM: thanks! | 15:57 |
| EmilienM | shardy: we need to fix that | 15:58 |
| EmilienM | shardy: what params set that? | 15:58 |
| EmilienM | right now, we set https://github.com/openstack/puppet-heat/blob/master/manifests/keystone/domain.pp#L94 | 15:58 |
| chem | shardy: EmilienM jaosorior: we could just add the missing bit in puppet manifest | 15:58 |
| EmilienM | by defeault | 15:58 |
| EmilienM | maybe tripleo overrides it? | 15:58 |
| EmilienM | chem: I was thinking at patching instack-undercloud with a Collector | 15:59 |
| shardy | EmilienM: will that remove all roles except "admin" ? | 15:59 |
| shardy | or role assignments rather | 15:59 |
| chem | EmilienM: do not get the "collector' | 16:00 |
| chem | part of your sentence | 16:00 |
| EmilienM | the problem is that I don't find where _member_ was added before | 16:00 |
| *** alhernan has left #tripleo | 16:00 | |
| *** alhernan has joined #tripleo | 16:00 | |
| pkovar | beagles: shardy: i can ask the docs team to add the tripleo guide there... | 16:00 |
| *** links has quit IRC | 16:01 | |
| pkovar | assuming they control the process | 16:01 |
| shardy | https://review.openstack.org/#/c/116884/ shows it not being created by puppet-keystone | 16:01 |
| chem | jaosorior: the '_member_' was used because before openstack cli juno an user could not exist without having a role, hence the member creation. This limitation has been removed since then | 16:01 |
| shardy | because it was created as part of the keystone DB bootstrap | 16:01 |
| chem | jaosorior: sorry openstack cli kilo not juno | 16:02 |
| jaosorior | chem: I see | 16:02 |
| shardy | For new deployments it's fine not to create it, but we can't remove user role assignments on upgrades | 16:02 |
| chem | jaosorior: EmilienM now, if we still need it we just add keystone_role {'_member': ensure=> present} somewhere | 16:02 |
| chem | and assign the required users to it | 16:03 |
| shardy | chem: interestingly we don't delete the role, only the role assignments | 16:03 |
| shardy | also it's important that the role must not be deleted and re-created with a different ID | 16:03 |
| shardy | e.g ref that patch I just linked | 16:03 |
| chem | shardy: hum ... how comes, where does it happen ? | 16:03 |
| chem | shardy: looking | 16:04 |
| shardy | according to that patch it's provided by a DB migration | 16:04 |
| shardy | but for any trust delegation to work, the ID's of roles (and assignments to users) must never change | 16:04 |
| shardy | because delegation is managed by ID, not name | 16:04 |
| *** Guest30394 has quit IRC | 16:05 | |
| chem | shardy: well, nothing in the puppet-keystone will remove an existing role, but the assignement may be change, depending on the manifest | 16:05 |
| shardy | chem: Ok, thanks, but it does ensure that role assignments for each user match the manifest exactly? | 16:06 |
| shardy | that's what we're seeing, the role is left alone, but all users get the role assignment removed | 16:07 |
| chem | shardy: yes, I guess that where the problem is, as member is no longer assumed | 16:07 |
| chem | shardy: so just recreating the assignement would do it no ? | 16:07 |
| EmilienM | shardy: why heat needs _member_? How can we configure it? | 16:08 |
| shardy | chem: Yup that would work around it provided the role ID hasnt changed (and it hasn't) | 16:08 |
| EmilienM | chem: I'm working on that | 16:08 |
| *** mbound has quit IRC | 16:08 | |
| jaosorior | any idea why the stable/liberty is failing even before starting the undercloud build? | 16:08 |
| shardy | EmilienM: it doesn't need _member_, it needs the roles it delegated to still exist, and be assigned to the user | 16:08 |
| shardy | so when you create a heat stack, if you have roles "x, y, z" you must still have these roles when something tries to delegate via the trust | 16:09 |
| chem | shardy: EmilienM this is legacy practice, as the member role was created no matter what, it has been used in a lot of places | 16:09 |
| shardy | which happens e.g when agents inside the VMs signal back to heat | 16:09 |
| EmilienM | ok | 16:09 |
| shardy | I suspect that unconditionally removing _member_ will also break some interactions with the v2 API | 16:09 |
| shardy | which I know is deprecated, but may cause issues for other users | 16:10 |
| chem | shardy: not really, but you would bump into problem like the one for heat, where permission assumbe use to be part of _member_ | 16:10 |
| chem | assume | 16:10 |
| chem | shardy: there is really nothing special about _member_ :) | 16:11 |
| shardy | chem: OK, I thought _member_ was used to grant access to a project in v2, assuming no other roles assignments are added to the same project, won't removing _member_ deny access to the project? | 16:11 |
| *** dsariel has quit IRC | 16:11 | |
| openstackgerrit | Merged openstack/python-tripleoclient: Throw exception if undercloud password file is missing https://review.openstack.org/294229 | 16:11 |
| *** saneax is now known as saneax_AFK | 16:12 | |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: Make sure we assign _member_ role to admin user https://review.openstack.org/307352 | 16:12 |
| EmilienM | shardy, chem, jaosorior: please review ^ | 16:12 |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: Make sure we assign _member_ role to admin user https://review.openstack.org/307352 | 16:13 |
| chem | shardy: it will if there are no other role for the user in the project. But you could have setup on non-default policy where each user has a meaningful (for your organisation) named role (like, "dev", "manager", ....) | 16:13 |
| chem | EmilienM: looking | 16:13 |
| chem | EmilienM: jaosorior are we sure admin has only those two roles ? | 16:13 |
| shardy | EmilienM: thanks! | 16:13 |
| chem | shardy: ^ | 16:14 |
| shardy | is it possible to make it only assert the role assignment if the _member_role exists? | 16:14 |
| shardy | chem: Well, unless an operator decides to create some additional ones like I just did (which will also be erased by the update) | 16:14 |
| chem | shardy: I would say that it would be complicated, puppet being what it is. | 16:15 |
| shardy | chem: https://bugzilla.redhat.com/show_bug.cgi?id=1325475 shows there are two roles delegated on this old deployment | 16:17 |
| openstack | bugzilla.redhat.com bug 1325475 in rhel-osp-director "rhel-osp-director: upgrade 7.3->8.0, that follows update 7.2->7.3, times out due to os-collect-config auth failure" [High,New] - Assigned to therve | 16:17 |
| shardy | _member_ and admin | 16:17 |
| shardy | on a new deployment, it will only be admin | 16:17 |
| chem | shardy: oki, so it's good | 16:17 |
| chem | shardy: maybe let the user know that if the admin role has been modified after deployment the upgrade will reset them to admin, _member_ ? | 16:17 |
| chem | shardy: in the upgrade's readme ? | 16:17 |
| shardy | chem: Yeah, it'd be a good thing to document | 16:17 |
| EmilienM | I'm out for lunch, feel free to comment in gerrit. | 16:17 |
| shardy | EmilienM: ack, thanks! | 16:17 |
| *** trown is now known as trown|lunch | 16:18 | |
| zoliXXL | see you tomorrow | 16:19 |
| *** ohamada has quit IRC | 16:21 | |
| *** pcaruana has quit IRC | 16:21 | |
| *** saneax_AFK is now known as saneax | 16:22 | |
| *** zoliXXL is now known as zoli_gone-proxy | 16:23 | |
| *** mgould has quit IRC | 16:25 | |
| *** beekneemech changes topic to "TripleO | CI status: http://tripleo.org/cistatus.html | Docs: http://tripleo.org/" | 16:26 | |
| *** dsariel has joined #tripleo | 16:27 | |
| dprince | gfidente: hey, so I'm blocked on a Heat bug ATM. So if we want to land the MySQL VIP change against master I would rather just accept it sooner if we can get it passing. | 16:27 |
| dprince | gfidente: so land your patch first and then I rebase on it? | 16:28 |
| gfidente | dprince, honestly, it consistently failed the ha and nonha jobs and consistently passed upgrades | 16:28 |
| gfidente | I tested it locally and it worked | 16:28 |
| gfidente | but I am suspicious | 16:28 |
| gfidente | it looks like the deployment fails just after the overcloud goes in CREATE_COMPLETE | 16:28 |
| dprince | gfidente: sounds like a slacker bit of work to me man :) | 16:28 |
| *** tosky has quit IRC | 16:29 | |
| dprince | gfidente: I think the idea is sane. We can get it passing | 16:29 |
| dprince | gfidente: just saying... we may as well go for it first I think | 16:29 |
| dprince | gfidente: also, the HA job is failing? | 16:30 |
| dprince | gfidente: looks like the HA jobs was failing all weekend... | 16:31 |
| dprince | jistr: ? | 16:31 |
| jistr | hmm i thought i saw it pass today | 16:32 |
| jistr | oh that was on mitaka though | 16:33 |
| dprince | jistr: yeah, master is consistently failing | 16:33 |
| dprince | jistr: like some sort of Galera issue perhaps? | 16:33 |
| dprince | jistr: http://logs.openstack.org/18/279218/14/check-tripleo/gate-tripleo-ci-f22-ha/ecba98a/console.html.gz | 16:33 |
| jistr | ah yea i saw that LP bug. i thought it was being un-done in RDO? | 16:33 |
| dprince | jistr: sorry, I may have missed it this morning. I was offline a bit | 16:34 |
| jistr | dprince: https://bugs.launchpad.net/tripleo/+bug/1571090 | 16:34 |
| openstack | Launchpad bug 1571090 in tripleo "clustercheck always failing on ha job" [Critical,Fix released] | 16:34 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-tripleoclient: Updated from global requirements https://review.openstack.org/268528 | 16:35 |
| dprince | jistr: it doesn't look resolved yet though. Like it still fails | 16:35 |
| dprince | trown|lunch: ^^^ any ideas on that? | 16:35 |
| *** tosky has joined #tripleo | 16:35 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-common: Updated from global requirements https://review.openstack.org/300629 | 16:36 |
| derekh_afk | dprince: FYI we no longer have access to jenkins, so report isn't updating http://lists.openstack.org/pipermail/openstack-dev/2016-April/092482.html | 16:36 |
| dprince | derekh_afk: oh... | 16:37 |
| dprince | derekh_afk: is that why my monday is confused :/ | 16:37 |
| jaosorior | Well, I've just seen some CI runs succeeding, but with very varied success rate | 16:37 |
| *** ccamacho has quit IRC | 16:38 | |
| therve | dprince, Updated https://review.openstack.org/#/c/307160/ with the hopefully actual fix :) | 16:38 |
| dprince | derekh_afk: can we re-write our report to use gerrit queries only | 16:38 |
| therve | Was waiting on a build result which won't happen today apparently | 16:38 |
| jaosorior | and for some reason stable/liberty seems preeeetty red | 16:38 |
| dprince | therve: nice, I will fire a recheck to test it in TripleO CI | 16:38 |
| derekh_afk | dprince: also, I was deleting unused floating IP's from the tripleo rack all weekend to keep thing moving, but its all ok now again, mordred sorted it out | 16:38 |
| dprince | therve: appreciate you looking into that one | 16:38 |
| therve | No problem! | 16:39 |
| derekh_afk | dprince: I was thinging the same, we wouldn't get the jobs until they finished | 16:39 |
| dprince | derekh_afk: yeah, so those nodepool changes were our issue then? | 16:39 |
| derekh_afk | dprince: but its better then nothing | 16:39 |
| *** jistr has quit IRC | 16:39 | |
| dprince | derekh_afk: yeah, we could update the metrics to post mid-job (on the fly). But still, I think scraping the gerrit logs is probably a better plan for our report | 16:40 |
| derekh_afk | dprince: not sure exactly what changes were the issue but its now fixed on the nodpool/shade side | 16:40 |
| dprince | derekh_afk: cool. That partial revert I emailed you was what I thought looked suspicious | 16:40 |
| openstackgerrit | Keith Schincke proposed openstack/python-tripleoclient: Add support for CephRgwKey generation https://review.openstack.org/296002 | 16:41 |
| dprince | derekh_afk: http://git.openstack.org/cgit/openstack-infra/nodepool/commit/?id=b38290421483c19428910c31fe4c918229c14d86 | 16:41 |
| derekh_afk | dprince: if we're going to go down the gerrit query route then ccamacho said he'd looking into it | 16:41 |
| *** derekh_afk is now known as derekh | 16:41 | |
| dprince | derekh: so others aren't confused we might should just take our reports down then | 16:42 |
| dprince | derekh: want me to post a patch to disable them? | 16:42 |
| derekh | dprince: agreed, go for it | 16:42 |
| dprince | derekh: one more thing while I've got you | 16:43 |
| *** rcernin has quit IRC | 16:43 | |
| derekh | dprince: yup | 16:43 |
| dprince | derekh: related to the metrics stuff.... I think we ought to make that fail-safe. In that if metrics posting fails the job would still pass | 16:43 |
| mordred | derekh: yay! | 16:43 |
| dprince | derekh: as is if the stats posting fails I think it would fail the job | 16:44 |
| derekh | dprince: makes sense | 16:44 |
| *** mcornea has quit IRC | 16:44 | |
| dprince | derekh: so just a little cleanup I thought after we landed it | 16:44 |
| dprince | derekh: I would hate to see jobs fail due to metrics not posting, etc. | 16:44 |
| derekh | dprince: yup | 16:45 |
| *** ccamacho has joined #tripleo | 16:46 | |
| openstackgerrit | Keith Schincke proposed openstack/tripleo-heat-templates: Add support for deploying RGW on controllers. https://review.openstack.org/289027 | 16:47 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: composable neutron dhcp service https://review.openstack.org/303386 | 16:48 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: composable neutron metadata service https://review.openstack.org/303618 | 16:48 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: composable neutron l3 service https://review.openstack.org/303562 | 16:48 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: composable glance services https://review.openstack.org/237370 | 16:48 |
| *** derekh has quit IRC | 16:48 | |
| openstackgerrit | Keith Schincke proposed openstack/python-tripleoclient: Add support for CephRgwKey generation https://review.openstack.org/296002 | 16:52 |
| chem | shardy: so I think there is a way to keep all existing role and adding only the '_member_' to the admin role. Removing the need to modify the upgrade documention | 16:56 |
| *** ccamacho has quit IRC | 17:00 | |
| *** psanchez has quit IRC | 17:01 | |
| *** psanchez has joined #tripleo | 17:01 | |
| openstackgerrit | Dan Prince proposed openstack-infra/tripleo-ci: Ignore bash errors when posting metrics https://review.openstack.org/307374 | 17:01 |
| openstackgerrit | Dan Prince proposed openstack-infra/tripleo-ci: Make the reviewday config file pluggable https://review.openstack.org/307377 | 17:03 |
| shardy | chem: Ok - note this must leave the _member_ role assignment when _member_ exists, but not try to add it when that role is missing | 17:06 |
| *** aufi has quit IRC | 17:06 | |
| shardy | e.g new/recent deployments won't have that role | 17:07 |
| chem | shardy: this will do | 17:07 |
| *** alhernan has quit IRC | 17:07 | |
| *** jpena is now known as jpena|off | 17:07 | |
| chem | shardy: but I'm going to check anyway :) | 17:07 |
| shardy | chem: ack, thanks - I'm currently trying to test but undercloud install overwrites my locally updated instack-undercloud RPM | 17:08 |
| openstackgerrit | Ryan Hallisey proposed openstack-infra/tripleo-ci: Allow the container job to run again https://review.openstack.org/288915 | 17:08 |
| chem | shardy: sorry, I misread, this won't do | 17:10 |
| chem | shardy: I will fail with "No role with a name or ID of '_member_' exists" | 17:10 |
| shardy | chem: Ok, we need it to skip attempting the assignment in that case | 17:10 |
| chem | shardy: Oki, I'll think of something ... | 17:11 |
| *** trown|lunch is now known as trown | 17:13 | |
| *** rook has joined #tripleo | 17:14 | |
| chem | shardy: the best we can do with puppet (Emilien corrects me if I'm wrong) is to test if a resource is defined in the catalog, not that a resource exists on the system. | 17:17 |
| shardy | chem: can you detect the current state of the role assignments via the puppet model, e.g what roles are assigned? | 17:18 |
| shardy | an alternative would be to detect it outside of puppet I guess and pass a roles list in via hiera | 17:18 |
| chem | shardy: yep, that's basically the only option. Puppet cannot take decision at "runtime" depending on how the current system is, you hardcode the final state of the system, that's it that's all | 17:20 |
| chem | shardy: so is something like add the role during upgrade, but not during creation easy to do ? | 17:21 |
| shardy | chem: well it's just a case of listing the roles and appending _member_ if it exists | 17:21 |
| shardy | so I guess we could do that from within the instack-undercloud python code | 17:22 |
| chem | shardy: and transform it to a hiera variable that would be accessible from https://review.openstack.org/#/c/307352/2/elements/puppet-stack-config/puppet-stack-config.pp | 17:22 |
| chem | shardy: so that we can do if ($upgrade) then ensure member role; fi | 17:23 |
| chem | shardy: sorry "if hiera($upgrade, false) ..." | 17:23 |
| shardy | chem: No, we can'd do a conditional on upgrade, because not all upgrades have this problem | 17:23 |
| shardy | only those from sufficiently old versions | 17:23 |
| shardy | it'll be something like append a possibly empty list of additional roles | 17:24 |
| shardy | let me hack out a patch, sec | 17:24 |
| chem | shardy: so could we have "hiera($admin_roles)" ? so that if it include '_member_' then we make sure it has ? | 17:25 |
| *** yamahata has joined #tripleo | 17:29 | |
| *** colonwq_ is now known as colonwq | 17:30 | |
| EmilienM | chem: right | 17:31 |
| EmilienM | chem: or maybe we can have an Exec + bash script :-/ | 17:31 |
| chem | EmilienM: hum ... I didn't think of it ... but how would you make such script idempotent ? | 17:32 |
| chem | EmilienM: it would have to run at each puppet run | 17:32 |
| colonwq | I am trying to update a template to pass the internal API IP address of into a puppet module. What is the correct parameter? | 17:32 |
| EmilienM | chem: right, bad idea | 17:32 |
| chem | EmilienM: but maybe idempotent is not required here as it's run only once during image creation (if I understand this right ) | 17:32 |
| EmilienM | chem: I think a bash script could work, regarding shardy's comment, we need specific logic | 17:34 |
| *** hjensas has joined #tripleo | 17:34 | |
| EmilienM | chem: but the problem is how to handtle this collector | 17:34 |
| chem | EmilienM: and provided we don't care about idempotency | 17:34 |
| *** gfidente has quit IRC | 17:35 | |
| *** tosky has quit IRC | 17:36 | |
| chem | EmilienM: that's really, really tricky (not possible?) to do with puppet | 17:36 |
| *** dprince has quit IRC | 17:36 | |
| chem | EmilienM: the only way to do this I think is using a dedicated fact, but having a fact that describe the state of the system, goes to the keystone db each time it's run, looks like a very bad idea | 17:37 |
| EmilienM | I thought about a fact :) | 17:38 |
| EmilienM | but it's really weird | 17:38 |
| chem | EmilienM: better have the python code provides us with a hiera variable like "does_member_role_exists" ? | 17:39 |
| EmilienM | why can't we tell heat to live without _member_? | 17:39 |
| EmilienM | what thing am I missing? | 17:39 |
| chem | EmilienM: I think it can, we just have to adjust its membership | 17:40 |
| shardy | EmilienM: because we already delegated that role to heat when the overcloud was created | 17:40 |
| EmilienM | shardy: can't we update that? | 17:40 |
| shardy | EmilienM: No, the delegation by default exists for the lifetime of the stack | 17:41 |
| shardy | we could change that, but that's how it currently works | 17:41 |
| EmilienM | ok no problem, just trying to understand how it works | 17:41 |
| shardy | http://hardysteven.blogspot.co.uk/2014/04/heat-auth-model-updates-part-1-trusts.html | 17:42 |
| shardy | that provides some background FYI | 17:42 |
| chem | shardy: EmilienM so our best bet I think is to have access to an hiera variable that can tell puppet if the "_member_" role exists | 17:42 |
| chem | shardy:thanks | 17:42 |
| EmilienM | chem: an fact could work, it would check "if openstack role list | grep _member_ then true else false" and use the fact in undercloud manifest | 17:42 |
| *** ifarkas has quit IRC | 17:44 | |
| chem | EmilienM: note that this will be run each time puppet run, that openstack may, or may not be installed at the time, that it somehow has to get its permission from somewhere and that '/root/openrc' may not exists at the time it run, and finally it's bad puppet practice (unfortunatly ... ?) | 17:45 |
| chem | EmilienM: dynamically modifying the catalog depending on the state of the system is not what puppet is designed for | 17:46 |
| *** Marga_ has quit IRC | 17:46 | |
| EmilienM | right, it's really hard to solve | 17:46 |
| chem | EmilienM: the only "clean" way to do it would be a type/provider but that seems like overkill | 17:46 |
| EmilienM | maybe we can just continue to use _member_ | 17:47 |
| chem | EmilienM:then keystone_role { '_member_': ensure => present } would have to be added | 17:47 |
| shardy | EmilienM: I don't think we want to add it to recent deployments where that role does not exist | 17:47 |
| chem | EmilienM: but shardy said that newer install do not have it | 17:47 |
| shardy | folks will raise bugs saying it's wrong for keystone v3 etc, and probably it will cause confusion IMO | 17:48 |
| chem | shardy: Emilien so do you think we can have this external hiera variable ? | 17:48 |
| *** dmacpher is now known as dmacpher-afk | 17:49 | |
| *** dtantsur is now known as dtantsur|afk | 17:49 | |
| shardy | chem: yes, I'm writing an instack-undercloud patch which we can try | 17:49 |
| EmilienM | there is one option, very easy | 17:49 |
| chem | shardy:ack | 17:49 |
| chem | EmilienM: ? | 17:49 |
| EmilienM | we add a param in undercloud.conf | 17:49 |
| EmilienM | like: keystone_member_role | 17:50 |
| EmilienM | we set it with Hiera | 17:50 |
| EmilienM | and we use it as a boolean | 17:50 |
| EmilienM | and we set it to True for backward compatibility (or False for new deployments?) | 17:50 |
| shardy | EmilienM: that's basically what I'm doing, but the user shouldn't have to care about this IMO | 17:51 |
| EmilienM | right | 17:51 |
| *** lucasagomes is now known as lucas-dinner | 17:51 | |
| *** dprince has joined #tripleo | 17:52 | |
| *** shivrao has joined #tripleo | 17:52 | |
| EmilienM | shardy: you take over my patch? | 17:52 |
| shardy | EmilienM: sure, I'll squash my patch into yours and we can review | 17:53 |
| shardy | thanks to you and chem for your help with this :) | 17:53 |
| EmilienM | cool | 17:53 |
| *** electrofelix has quit IRC | 17:59 | |
| *** mkovacik_ has quit IRC | 18:00 | |
| EmilienM | the check/gate/post queue is huge, fyi: ovh cloud has been dropped from nodepool | 18:02 |
| *** jaosorior has quit IRC | 18:03 | |
| *** rhallisey has quit IRC | 18:08 | |
| *** pkovar has quit IRC | 18:09 | |
| *** yamahata has quit IRC | 18:11 | |
| *** Marga_ has joined #tripleo | 18:13 | |
| *** Marga_ has quit IRC | 18:13 | |
| *** Marga_ has joined #tripleo | 18:15 | |
| *** Marga__ has joined #tripleo | 18:17 | |
| *** Marga__ has quit IRC | 18:18 | |
| *** Marga_ has quit IRC | 18:19 | |
| *** Marga_ has joined #tripleo | 18:23 | |
| *** rhallisey has joined #tripleo | 18:26 | |
| *** Marga_ has quit IRC | 18:28 | |
| *** mbound has joined #tripleo | 18:28 | |
| *** Marga_ has joined #tripleo | 18:29 | |
| *** sambetts is now known as sambetts|afk | 18:32 | |
| *** rbrady has quit IRC | 18:32 | |
| openstackgerrit | Merged openstack/tripleo-quickstart: Set default tht parameters for net-iso and remove hardcoded values https://review.openstack.org/305119 | 18:33 |
| colonwq | EmilienM, the long queue only happened because I submitted patches for review... and its Monday | 18:35 |
| slagle | EmilienM: i'm looking to move more stuff to puppet in instack-undercloud, what's the right puppet module to use to manage iptables? puppetlabs-firewall? | 18:37 |
| EmilienM | yes | 18:38 |
| EmilienM | definitly | 18:38 |
| EmilienM | slagle: let me show you how we do it in overcloud | 18:38 |
| EmilienM | we have this class: https://github.com/openstack/puppet-tripleo/blob/master/manifests/firewall.pp to whether or not enable pre/post rules | 18:39 |
| EmilienM | pre/post are defined here https://github.com/openstack/puppet-tripleo/tree/master/manifests/firewall | 18:39 |
| EmilienM | and this define will create rules for you: https://github.com/openstack/puppet-tripleo/blob/master/manifests/firewall/rule.pp | 18:39 |
| EmilienM | colonwq: ahah | 18:39 |
| *** Marga_ has quit IRC | 18:39 | |
| *** rbrady has joined #tripleo | 18:40 | |
| slagle | EmilienM: k, thanks. i wonder if i can reuse this for the undercloud, i'll investigate | 18:40 |
| EmilienM | slagle: I think so, it would be great to have the same bits | 18:40 |
| slagle | EmilienM: yea, unfortunately the rules are set in tht though | 18:41 |
| slagle | in the hieradata files | 18:41 |
| slagle | that should probably be moved to puppet-tripleo | 18:41 |
| EmilienM | slagle: we could have them in the undercloud hiera file | 18:41 |
| slagle | yea, it's just a lot to copy/paste | 18:42 |
| slagle | and when you add a new service to under/over cloud you have to update 2 places | 18:42 |
| EmilienM | right | 18:42 |
| slagle | not a big deal, i'll see if i can make it better | 18:42 |
| *** Marga_ has joined #tripleo | 18:48 | |
| *** Marga_ has quit IRC | 18:48 | |
| *** Marga_ has joined #tripleo | 18:49 | |
| *** Marga_ has quit IRC | 18:49 | |
| *** Marga_ has joined #tripleo | 18:50 | |
| *** Marga_ has quit IRC | 18:50 | |
| *** Marga_ has joined #tripleo | 18:51 | |
| *** dbecker has quit IRC | 18:59 | |
| *** dbecker has joined #tripleo | 19:05 | |
| *** Marga_ has quit IRC | 19:07 | |
| *** Marga_ has joined #tripleo | 19:07 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: images/build role has moved to its own repo https://review.openstack.org/307422 | 19:08 |
| *** florianf has quit IRC | 19:08 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: CI jjb is moved to rdo-infra/ci-config repo https://review.openstack.org/307425 | 19:11 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: images/build role has moved to its own repo https://review.openstack.org/307422 | 19:13 |
| *** eil397 has joined #tripleo | 19:13 | |
| *** eil397 has left #tripleo | 19:13 | |
| openstackgerrit | Merged openstack/diskimage-builder: Updated from global requirements https://review.openstack.org/303416 | 19:16 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Nodes assignment part 2 https://review.openstack.org/307427 | 19:16 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Nodes assignment part 3 https://review.openstack.org/307428 | 19:16 |
| openstackgerrit | Merged openstack/diskimage-builder: Fix add-apt-repository package for precise https://review.openstack.org/294760 | 19:17 |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: run nova-api with mod_proxy_uwsgi httpd module https://review.openstack.org/307431 | 19:23 |
| *** weshay has quit IRC | 19:29 | |
| *** jprovazn has quit IRC | 19:31 | |
| openstackgerrit | Merged openstack/diskimage-builder: Skip gentoo test https://review.openstack.org/302516 | 19:31 |
| openstackgerrit | Merged openstack/diskimage-builder: Add qcow2 generation for better test coverage https://review.openstack.org/302517 | 19:32 |
| openstackgerrit | Merged openstack/diskimage-builder: Fix disk usage report https://review.openstack.org/302509 | 19:36 |
| openstackgerrit | Merged openstack/diskimage-builder: dib-run-parts: make cp to target root more robust https://review.openstack.org/219453 | 19:37 |
| openstackgerrit | Merged openstack/diskimage-builder: simple-init: Fix path for /etc/ssh test https://review.openstack.org/305498 | 19:37 |
| *** rhallisey_ has joined #tripleo | 19:47 | |
| *** rhallisey has quit IRC | 19:50 | |
| *** bvandenh has quit IRC | 19:58 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Reload haproxy after injecting certs w/o pcmk too https://review.openstack.org/301840 | 20:11 |
| beekneemech | Simple patch ready for +2a: https://review.openstack.org/#/c/306533 | 20:12 |
| *** beekneemech is now known as bneme | 20:12 | |
| *** bneme is now known as bnemec | 20:12 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Redirect instack image build output to file https://review.openstack.org/306533 | 20:14 |
| slagle | trown beat me to it | 20:15 |
| *** dprince has quit IRC | 20:16 | |
| trown | :) | 20:18 |
| openstackgerrit | James Slagle proposed openstack/python-tripleoclient: Use enable-packages-install https://review.openstack.org/307462 | 20:18 |
| trown | beekneemech is my favorite voltron piece that was rarely in any episodes | 20:19 |
| bnemec | trown: lol | 20:20 |
| bnemec | And thanks | 20:20 |
| trown | but clearly essential to have a good knee | 20:20 |
| * bnemec has been failing to casual nick friday lately | 20:20 | |
| *** morazi has quit IRC | 20:20 | |
| *** pkovar has joined #tripleo | 20:21 | |
| openstackgerrit | James Slagle proposed openstack/instack-undercloud: Use enable-packages-install element https://review.openstack.org/263829 | 20:22 |
| *** rhallisey_ is now known as rhallisey | 20:23 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/puppet-tripleo: Add redis profile https://review.openstack.org/304754 | 20:26 |
| openstackgerrit | Merged openstack/puppet-tripleo: Disable ip_nonlocal_bind (rely on the HAProxy 'transparent' option) https://review.openstack.org/267138 | 20:27 |
| openstackgerrit | Merged openstack/instack-undercloud: Remove centos-cloud-repo element https://review.openstack.org/301851 | 20:28 |
| *** yamahata has joined #tripleo | 20:33 | |
| *** egafford has quit IRC | 20:37 | |
| stevebaker | trown: hey, tripleo-quickstart OVB is progressing but I can't currently use the built undercloud images because they are a little too virt specific (no cloud-init, possibly tweaked default network interface) | 20:39 |
| *** MaxPC has quit IRC | 20:40 | |
| *** rickflare has quit IRC | 20:40 | |
| *** rickflare has joined #tripleo | 20:41 | |
| trown | stevebaker: hmm, I removed cloud-init because the image took forever to boot otherwise since it was trying to find non-existant metadata server | 20:41 |
| stevebaker | trown: one option would be to configure cloud-init for config-drive only during the image build | 20:42 |
| *** eggmaster has quit IRC | 20:43 | |
| trown | stevebaker: if that would solve the OVB case, I am all for it | 20:43 |
| trown | or even if it helped get it closer | 20:43 |
| stevebaker | trown: but I wonder if there is a way of doing that during virt boot, like injecting the config-drive data so cloud-init doesn't try to reach the metadata server | 20:44 |
| trown | I bet larsks would know ^ | 20:44 |
| *** lucas-dinner has quit IRC | 20:44 | |
| stevebaker | trown: is undercloud built by modifying the overcloud image? | 20:46 |
| *** dsariel has quit IRC | 20:46 | |
| trown | stevebaker, kind of, they both use the same base image that has all of the pacakages pre-installed | 20:46 |
| stevebaker | trown: ah, ok | 20:47 |
| trown | for the overcloud that base image is just passed to DIB | 20:47 |
| trown | and for the undercloud there are a small amount of conversion steps | 20:47 |
| trown | it saves installing 300 packages twice | 20:47 |
| stevebaker | trown: fair enough | 20:48 |
| larsks | stevebaker: trown: you can certain just attach a config-drive to your libvirt guest to make cloud-init happy. | 20:49 |
| larsks | I generally use "cloud" images with libvirt, because that makes it super easy to inject an ssh key... | 20:50 |
| trown | nice, so we should probably stop removing cloud-init and just add a step to attach a config-drive to the libvirt/setup role | 20:51 |
| *** lucasagomes has joined #tripleo | 20:53 | |
| trown | stevebaker: we dont do anything wrt network interfaces https://github.com/redhat-openstack/ansible-role-tripleo-image-build/blob/master/templates/undercloud-convert-default.sh.j2 | 20:53 |
| larsks | In case you wonder what a config drive needs to look like, https://github.com/larsks/virt-utils/blob/master/create-config-drive is what I use locally. | 20:54 |
| stevebaker | trown, bnemec: btw I'm going to propose that the tripleo-quickstart OVB roles live in the openstack-virtual-baremetal repo rather than create yet another one https://github.com/steveb/openstack-virtual-baremetal/commit/3dd7921168a32cf46b1b10c57007c5ed7e40a9f1 | 20:54 |
| stevebaker | trown: yeah, I wasn't sure about custom networking, its probably just cloud-init | 20:54 |
| *** lblanchard has quit IRC | 20:55 | |
| trown | stevebaker: the downside to doing it that way is that we can't use ansible-galaxy, but I am not sure how useful that is | 20:56 |
| stevebaker | trown: I'm adding it via galaxy and it seems to work | 20:56 |
| openstackgerrit | Steve Baker proposed openstack/tripleo-quickstart: Flag to enable OpenStack Virtual Baremetal instead of libvirt https://review.openstack.org/304954 | 20:57 |
| stevebaker | trown: ^, although I'm not sure if you're planning on registering these repos on the galaxy website | 20:58 |
| *** weshay has joined #tripleo | 20:58 | |
| trown | stevebaker: ya not sure... not in the short term, but what does https://review.openstack.org/#/c/304954/3/ansible-role-requirements.yml look like when using the OVB repo instead of a role repo? | 20:59 |
| openstackgerrit | Steven Hardy proposed openstack/instack-undercloud: Work around deletion of _member_ role assignments on upgrade https://review.openstack.org/307352 | 20:59 |
| stevebaker | trown: It looks exactly like that! | 20:59 |
| shardy | EmilienM, chem: ^^ | 21:00 |
| stevebaker | trown: its just that the role ends up being a path https://review.openstack.org/#/c/304954/3/playbooks/quickstart.yml | 21:00 |
| trown | stevebaker: I am confused... that points to https://review.openstack.org/#/c/304954/3/ansible-role-requirements.yml | 21:00 |
| shardy | I've not had time to test the entire patch, but I tested the _member_role_exists part and wired in the hiera boolean | 21:00 |
| shardy | I'm dropping off for the day now, if anyone can review/test that would be great | 21:00 |
| stevebaker | trown: oh duh, sorry | 21:01 |
| trown | stevebaker: whoops wrong buffer. that points to https://github.com/steveb/ansible-role-tripleo-quickstart-ovb.git | 21:01 |
| *** lucasagomes has quit IRC | 21:03 | |
| *** rhallisey has quit IRC | 21:04 | |
| *** jobewan is now known as jobewan_away | 21:05 | |
| trown | larsks: I think we could do some of the steps in libvirt/setup/undercloud with a config drive instead | 21:05 |
| *** dsariel has joined #tripleo | 21:06 | |
| larsks | trown: Yeah, probably, but I'm not sure we want to: we don't get the same feedback in the event of failures that we do with virt-customize. What steps were you thinking about? | 21:07 |
| stevebaker | trown: hmm, maybe galaxy doesn't work with roles in subdirectories | 21:08 |
| trown | larsks: ah right, key injection, and hostname setting | 21:10 |
| *** shardy has quit IRC | 21:11 | |
| trown | larsks: hostname setting is done in the image now, which is fine but if we are passing a config-drive to make cloud-init happy anyways might as well do it there | 21:11 |
| trown | makes the image even more generic | 21:11 |
| trown | could also move selinux config there from the image build | 21:12 |
| *** lucasagomes has joined #tripleo | 21:12 | |
| larsks | trown: I am kind of nervous about splitting our config so that we have "some things in config drive" and "some things in playbooks". | 21:12 |
| trown | stevebaker: ya, I think it needs to have meta/main.yml at the top level, but that is not a deal breaker, pypi seems like it would work just as well | 21:12 |
| stevebaker | trown: ok I'll have a play, I don't want to pollute the root ovb dirs too much | 21:13 |
| *** thrash is now known as thrash|g0ne | 21:14 | |
| trown | stevebaker: pypi just feels a bit strange because it is not really a python project, it would be a bit like distrubuting a puppet module as a ruby gem | 21:15 |
| stevebaker | trown: we've already committed worse pypi sins (*cough* dib) | 21:16 |
| trown | larsks: k, we can pass an empty config drive to make cloud-init happy? | 21:16 |
| larsks | trown: well, not *empty*, but with the minimal config as shown in that script I linked earlier... | 21:17 |
| trown | stevebaker: lol, ya but there is no BASH distribution platform... or if there is I dont want to know about it | 21:17 |
| *** ccamacho has joined #tripleo | 21:18 | |
| trown | larsks: cool, I will check it out, thanks! | 21:18 |
| trown | gotta run, have a good night all | 21:18 |
| stevebaker | trown: thanks \o | 21:18 |
| *** trown is now known as trown|outtypewww | 21:18 | |
| *** dbecker has quit IRC | 21:19 | |
| *** julim has quit IRC | 21:20 | |
| *** jobewan_away has quit IRC | 21:27 | |
| *** akrivoka has quit IRC | 21:29 | |
| bnemec | stevebaker: At least dib has _some_ python in it. dib-utils is the really horrible one - it's just one bash script, yet it's on pypi. :-) | 21:32 |
| *** dustins has quit IRC | 21:33 | |
| *** jobewan has joined #tripleo | 21:35 | |
| stevebaker | bnemec: I don't think having galaxy roles in the ovb repo will work - lots of little git repos is fine though | 21:35 |
| *** shadower has quit IRC | 22:01 | |
| *** ccamacho has quit IRC | 22:05 | |
| *** ccamacho has joined #tripleo | 22:11 | |
| *** mbound has quit IRC | 22:12 | |
| *** mbound has joined #tripleo | 22:15 | |
| *** zoli_gone-proxy is now known as zoliXXL | 22:17 | |
| openstackgerrit | Merged openstack/instack-undercloud: Add a reload of keepalived https://review.openstack.org/305478 | 22:19 |
| *** ayoung has quit IRC | 22:27 | |
| *** yamahata has quit IRC | 22:27 | |
| *** zoliXXL is now known as zoli_gone-proxy | 22:30 | |
| *** yamahata has joined #tripleo | 22:34 | |
| *** lblanchard has joined #tripleo | 22:38 | |
| *** MaxPC has joined #tripleo | 22:45 | |
| *** pkovar has quit IRC | 22:53 | |
| *** jobewan has quit IRC | 22:54 | |
| openstackgerrit | Merged openstack/diskimage-builder: Prioritize venv python on host https://review.openstack.org/285885 | 23:17 |
| *** mbound has quit IRC | 23:20 | |
| *** Marga__ has joined #tripleo | 23:25 | |
| *** Marga_ has quit IRC | 23:28 | |
| *** panda has quit IRC | 23:41 | |
| *** panda has joined #tripleo | 23:41 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!