Wednesday, 2016-04-20

« Tuesday, 2016-04-19 Index Thursday, 2016-04-21 »
*** shivrao has quit IRC00:09
*** spredzy has quit IRC00:09
*** spredzy has joined #tripleo00:15
*** shivrao has joined #tripleo00:45
*** shivrao has quit IRC00:47
*** lblanchard has quit IRC00:54
*** dsneddon has quit IRC00:55
*** dmacpher-afk is now known as dmacpher00:57
*** ayoung has joined #tripleo01:09
*** yamahata has quit IRC01:23
*** rhallisey has quit IRC01:26
*** saneax is now known as saneax_AFK01:27
*** shivrao has joined #tripleo01:36
*** lblanchard has joined #tripleo01:46
*** lblanchard has quit IRC01:51
*** dsneddon has joined #tripleo01:54
*** r-mibu has quit IRC02:19
*** fragatin_ has quit IRC02:26
*** absubram has joined #tripleo02:28
*** r-mibu has joined #tripleo02:29
*** shivrao has quit IRC02:31
*** ebalduf has joined #tripleo02:32
*** coolsvap has joined #tripleo02:42
*** fragatina has joined #tripleo02:46
*** fragatina has quit IRC02:51
*** yuanying has quit IRC02:56
*** weshay has quit IRC03:06
*** pradk has quit IRC03:07
*** fragatina has joined #tripleo03:07
*** fragatina has quit IRC03:12
*** links has joined #tripleo03:21
*** dsariel has joined #tripleo03:21
*** pradk has joined #tripleo03:29
*** ayoung has quit IRC03:40
*** panda has quit IRC03:42
*** panda has joined #tripleo03:42
openstackgerritIan Wienand proposed openstack/diskimage-builder: Move selinux restore to end of finalise  https://review.openstack.org/30812803:45
*** rlandy has quit IRC03:48
*** yuanying has joined #tripleo03:48
openstackgerritIan Wienand proposed openstack/diskimage-builder: Move selinux restore to end of finalise  https://review.openstack.org/30812803:53
openstackgerritgreghaynes proposed openstack/diskimage-builder: Document upstream executable numbering convention  https://review.openstack.org/30813104:00
openstackgerritIan Wienand proposed openstack/diskimage-builder: Add releasenotes  https://review.openstack.org/29219304:01
openstackgerritIan Wienand proposed openstack/diskimage-builder: Add releasenotes  https://review.openstack.org/29219304:05
*** dsariel has quit IRC04:08
*** fragatina has joined #tripleo04:09
openstackgerritgreghaynes proposed openstack/diskimage-builder: Document upstream executable numbering convention  https://review.openstack.org/30813104:09
*** andreaf has quit IRC04:11
*** fragatina has quit IRC04:13
*** andreaf has joined #tripleo04:21
*** fragatina has joined #tripleo04:26
*** fragatina has quit IRC04:31
*** coolsvap is now known as coolsvap|away04:47
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci: Tempest for tripleo CI  https://review.openstack.org/29584404:49
*** yamahata has joined #tripleo05:10
*** shivrao has joined #tripleo05:11
openstackgerritMerged openstack/instack-undercloud: Enable Zaqar by default.  https://review.openstack.org/29833105:14
openstackgerritMerged openstack/tripleo-heat-templates: Nova needs the proper volumes to use Cinder  https://review.openstack.org/30159005:20
openstackgerritMerged openstack/tripleo-common: Replace deprecated LOG.warn with LOG.warning  https://review.openstack.org/29157105:20
openstackgerritOpenStack Proposal Bot proposed openstack/tripleo-common: Updated from global requirements  https://review.openstack.org/30062905:20
*** ramishra has quit IRC05:28
*** ramishra has joined #tripleo05:30
*** oshvartz has joined #tripleo05:35
*** adarazs has quit IRC05:40
*** saneax_AFK is now known as saneax05:40
*** adarazs has joined #tripleo05:40
*** aufi_ has joined #tripleo05:57
*** yuanying_ has joined #tripleo05:57
*** stendulker has joined #tripleo05:57
*** yuanying has quit IRC05:58
*** fragatina has joined #tripleo06:02
*** fragatina has quit IRC06:07
*** rcernin has joined #tripleo06:21
*** hjensas has quit IRC06:22
*** shivrao has quit IRC06:27
*** tzumainn has quit IRC06:27
*** mcornea has joined #tripleo06:31
*** mikelk has joined #tripleo06:40
*** florianf has joined #tripleo06:47
*** yuanying_ has quit IRC06:47
*** tesseract has joined #tripleo06:47
*** yuanying has joined #tripleo06:48
*** tesseract is now known as Guest6708206:48
*** rasca_ has quit IRC06:49
*** florianf1 has joined #tripleo06:50
*** rasca has joined #tripleo06:50
*** florianf has quit IRC06:50
*** florianf1 has quit IRC06:50
*** ccamacho has joined #tripleo06:53
*** florianf has joined #tripleo06:53
*** jaosorior has joined #tripleo06:57
*** tremble has joined #tripleo07:01
*** tremble has joined #tripleo07:01
*** coolsvap|away has quit IRC07:06
*** jprovazn has joined #tripleo07:06
*** bvandenh has joined #tripleo07:07
*** ebalduf has quit IRC07:11
*** shadower has joined #tripleo07:11
*** zoli_gone-proxy is now known as zoliXXL07:14
*** chmouel has joined #tripleo07:15
*** coolsvap|away has joined #tripleo07:19
*** jpena|off is now known as jpena07:22
*** ifarkas has joined #tripleo07:24
*** ohamada has joined #tripleo07:24
*** dmacpher has quit IRC07:27
*** athomas has joined #tripleo07:28
*** dmacpher has joined #tripleo07:29
*** leanderthal|afk is now known as leanderthal07:31
*** hjensas has joined #tripleo07:34
*** hjensas has quit IRC07:34
*** hjensas has joined #tripleo07:34
*** spredzy has quit IRC07:35
*** paramite has joined #tripleo07:36
*** shardy has joined #tripleo07:36
*** olap has joined #tripleo07:38
*** olap has quit IRC07:45
*** olap has joined #tripleo07:47
*** paramite is now known as paramite|afk07:47
*** yuanying has quit IRC07:49
openstackgerritMerged openstack/tripleo-common: Updated from global requirements  https://review.openstack.org/30062907:50
*** yuanying has joined #tripleo07:52
*** dsariel has joined #tripleo07:53
*** lucas-dinner is now known as lucasagomes07:55
*** gfidente has joined #tripleo07:59
*** yuanying has quit IRC08:05
openstackgerritJuan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Avoid repetition in endpoint map  https://review.openstack.org/30818208:06
*** mbound has joined #tripleo08:07
openstackgerritJuan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Avoid repetition in endpoint map  https://review.openstack.org/30818208:07
*** ccamacho has quit IRC08:09
*** coolsvap|away is now known as coolsvap08:15
*** ccamacho has joined #tripleo08:16
openstackgerritMerged openstack/tripleo-heat-templates: Increase galera sync timeout in yum_update.sh  https://review.openstack.org/27266408:18
*** jistr has joined #tripleo08:18
*** ccamacho has quit IRC08:21
*** derekh has joined #tripleo08:37
*** shardy has quit IRC08:38
*** shardy has joined #tripleo08:38
*** mcornea has quit IRC08:40
openstackgerritCyril Lopez proposed openstack/tripleo-docs: The control placement is needed more information to aviod a no valid host found  https://review.openstack.org/30118508:42
*** mcornea has joined #tripleo08:43
*** mgould has joined #tripleo08:47
derekhIf anybody has a chance could I get some eyes on this please https://review.openstack.org/#/c/306506/08:50
*** electrofelix has joined #tripleo08:54
*** olap has quit IRC08:56
*** ccamacho has joined #tripleo08:57
*** david-lyle has quit IRC08:59
*** david-lyle has joined #tripleo09:00
jaosoriorderekh: Don't you have to export the LEAVE_RUNNING variable?09:05
derekhjaosorior: running it like this "LEAVE_RUNNING=1 ./toci_instack.sh"  makes it available to sub processes as if it was exported09:06
derekhjaosorior: gonna double check now, you have me doubting myself09:06
openstackgerritMerged openstack-infra/tripleo-ci: Don't add duplicate entries to metadata.txt  https://review.openstack.org/29636409:07
jaosoriorderekh: well... that's true, if it's only running in toci_instack.sh then it should be fine09:07
derekhjaosorior: ack, thats should the the only script reading the variable09:09
jaosoriorderekh: Then it's all good, I'll wait for it to pass the CI to workflow09:09
derekhjaosorior: cool, thanks09:09
jaosoriorderekh, shardy, have you guys seen an error importying pytx when running tripleo.sh --delorean-setup?09:12
derekhjaosorior: not me09:12
*** sambetts|afk is now known as sambetts09:14
*** ccamacho has quit IRC09:15
*** dtantsur|afk is now known as dtantsur09:18
openstackgerritMerged openstack/python-tripleoclient: Use enable-packages-install  https://review.openstack.org/30746209:18
*** akrivoka has joined #tripleo09:20
*** yamahata has quit IRC09:26
openstackgerritMerged openstack/diskimage-builder: Document upstream executable numbering convention  https://review.openstack.org/30813109:30
*** zoliXXL is now known as zoli|lunch09:33
*** zoli|lunch is now known as zoli_gone-proxy09:34
shardyjaosorior: not seen that, but tbh I've not run it for a week or so09:34
jaosoriorshardy: Uhm... alright, deploying the whole thing again, to see if it wasn't me who messed up my environment09:37
openstackgerritJuan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Avoid repetition in endpoint map  https://review.openstack.org/30818209:38
openstackgerritOpenStack Proposal Bot proposed openstack/python-tripleoclient: Updated from global requirements  https://review.openstack.org/26852809:40
*** athomas has quit IRC09:54
*** oshvartz has quit IRC09:56
*** olap has joined #tripleo09:58
*** mbound has quit IRC09:59
*** athomas has joined #tripleo10:00
*** ccamacho has joined #tripleo10:01
*** mbound has joined #tripleo10:02
*** tosky has joined #tripleo10:02
*** cmyster has joined #tripleo10:03
*** cmyster has quit IRC10:03
*** cmyster has joined #tripleo10:03
*** mbound has quit IRC10:04
*** mbound has joined #tripleo10:05
*** apetrich has quit IRC10:07
*** oshvartz has joined #tripleo10:08
*** stendulker has quit IRC10:18
*** mbound has quit IRC10:19
*** mbound has joined #tripleo10:20
*** mbound has quit IRC10:23
*** mbound has joined #tripleo10:25
*** paramite|afk is now known as paramite10:26
*** apetrich has joined #tripleo10:33
dtantsurhi folks! a quick heads-up from the ironic team: we are considering bumping the default API version used in the client10:34
dtantsurif this happens (it's not decided yet), tripleo gets broken. well, unless we merge https://review.openstack.org/#/c/272206/10:34
dtantsurplease review this patch, it will make tripleo import command behave the same way after the default API version goes beyond 1.1110:34
dtantsurthank you for your attention :)10:34
*** paramite is now known as paramite|afk10:37
*** paramite|afk is now known as paramite10:39
*** zoli_gone-proxy is now known as zoliXXL10:39
*** ccamacho is now known as ccamacho|lunch10:40
*** mcornea has quit IRC10:41
*** mcornea has joined #tripleo10:41
*** pradk has quit IRC10:45
*** mcornea_ has joined #tripleo10:45
dtantsurshardy, hi, bringing your attention to ^^^10:45
*** gfidente has quit IRC10:46
*** mcornea has quit IRC10:48
*** thrash|g0ne is now known as thrash10:49
*** jaosorior has quit IRC10:52
*** jaosorior has joined #tripleo10:53
*** pradk has joined #tripleo10:55
thrashjaosorior: for https://review.openstack.org/#/c/306333/ ... The error in the upgrades and non-ha jobs is directly related to the code.10:56
thrashjaosorior: he request you have made requires authentication. (HTTP 401) (Request-ID: req-af6814a6-d4d8-4065-9e6a-8d836ec12e82)10:57
thrashjaosorior: commented on the review...11:00
shardydtantsur: I'm fine with the change, and thanks for the heads-up but I agree this should be going into tripleo-common11:05
shardydtantsur: if we have to merge this as an interim step I guess that's OK, but mgould what are your thoughts re the refactor so it goes into tripleo-common instead?11:06
shardydtantsur: how quickly do we need to act re the version bump?11:07
* mgould mostly just wants the pain of that patch series to be over11:08
*** mikelk has quit IRC11:08
*** mikelk has joined #tripleo11:09
shardymgould: Ok, fair enough, someone else will have to do the refactor into tripleo-common then I guess11:09
mgoulddtantsur, where was the discussion about bumping the default API version? I thought we said we'd never do that11:09
mgouldshardy, no, I can probably do it11:09
dtantsurshardy, we're only starting the discussion, so not particularly urgent. yes, I also thought that, and then https://bugs.launchpad.net/python-ironicclient/+bug/157010011:10
openstackLaunchpad bug 1570100 in python-ironicclient "client defaults to old API version" [Low,Confirmed]11:10
shardymgould: I'm fine if we say it'll be a follow-up patch if that makes it easier11:10
mgouldso basically that patch, but targeting tripleo-common instead?11:10
dtantsurmgould, probably yes. with adding "initial_state" as an argument for register_nodes11:11
dtantsuralso a good chance to move some common utils there11:11
mgouldcool11:12
shardymgould: Yeah basically it needs to go in tripleo-common like11:12
shardythe register_all_nodes function11:12
dtantsurshardy, note that mistral does not use CLI, so your comment is not entirely correct11:12
mgouldcool11:12
shardy(or actually in that function)11:12
dtantsurwait11:12
mgouldI dimly remember trying it that way and it being horrible, but I can probably work something not-too-awful out11:12
dtantsurnot sure if we need that in register_all_nodes, as it's the caller who defines the API version11:12
shardydtantsur: we will want to import nodes via the UI11:13
dtantsurbut maybe it won't harm, dunno11:13
shardythe UI can never use this code11:13
dtantsurshardy, does UI use ironic CLI?11:13
dtantsurI don't think so11:13
mgouldyeah, the API version is set at the time you instantiate the client object11:13
*** pkovar has joined #tripleo11:13
*** sshnaidm has quit IRC11:14
dtantsurmgould, so yeah, lets do it in register_all_nodes. we will need a "provide" argument (bool) - whether to make sure that nodes are available11:14
*** sshnaidm has joined #tripleo11:14
dtantsurwe can't seriously check API version as we do in the current patch11:15
shardydtantsur: this is using the ironiccliennt python API is it not?11:15
mgoulddtantsur, so if the API version is too low we just fail messily?11:15
dtantsurshardy, right. we're talking about CLI now.11:15
mgouldI suppose we could keep the check in tripleoclient and document the requirement for register_all_nodes11:15
dtantsurmgould, no, we end up with what we end up with. that's kind of inevitable here11:15
dtantsurmgould, yeah - the check can stay in tripleoclient, I guess11:16
mgoulddtantsur, actually... we skip the "available" state11:16
mgouldso if you call it with a pre-1.11 API it will go straight to "available" and then do nothing11:17
dtantsurshardy, I agree that it's unclear from the bug if deva wants only a CLI change, or also change the Python API. I'd argue the latter is insane, but will see11:17
mgouldwait, I'm confused11:17
*** zoliXXL is now known as zoli|brb11:17
dtantsurmgould, you're right11:17
mgouldif you need ENROLL state, you need API >= 1.11, so we should keep the check in11:17
* shardy confused also11:17
dtantsurmgould, if you need API 1.11, call register_all_nodes with API 1.11 :)11:17
mgouldbut if you want initial_state=AVAILABLE, everything Just Works with either version11:18
shardymy point is simply that if we need this logic to interact with the ironicclient python API, we also need it in tripleo-common11:18
shardyor the UI can't use it11:18
mgouldshardy, gotcha11:18
dtantsurI'm not arguing against tripleo-common, just clarifying the scope11:18
shardydtantsur: sure, I was just confused by the UI using the CLI comment :)11:18
mgouldOK, I'll move that code into tripleo-common11:18
mgouldI noticed yesterday that my stable/liberty backport of this feature was merged several months and two rewrites ago :-)11:19
dtantsur:)11:20
mgouldso yeah, the stable/liberty equivalent of this patch is *quite different* :-)11:20
shardymgould: got a link?11:20
mgouldshardy, https://review.openstack.org/#/c/260452/11:21
mgouldoh wait, maybe that isn't it11:22
mgouldI see that's marked as a cherry-pick of a different patch, which *was* merged into master11:22
mgouldnever mind11:22
shardyYeah we shouldn't be landing any backports that didn't land on master11:23
shardyand to be clear, we can land this if there is a requirement to backport without breaking into tripleo-common11:24
shardythe tripleo-common requirement is really only for master/newton where we know our architecture is changing11:24
shardyI'd appreciate your feedback on how this can be wired in to https://review.openstack.org/#/c/300200 though11:25
mgoulddtantsur, do we need to backport this?11:25
shardythat doesn't yet have node registration, but it will need to be added via the "new" mistral interfaces11:25
*** trown|outtypewww is now known as trown11:25
mgouldmistral's an HTTP-API-scripting engine, right?11:26
mgouldso do you mean we'd need to provide an endpoint that mistral can call, or we'd need to port the logic to mistral scripts?11:26
jaosoriorthrash: hey, regarding, https://review.openstack.org/#/c/306333/2/tripleoclient/v1/overcloud_deploy.py , are you sure that's the case? we pretty much have the same flow in master and it works since you are explicitly specifying the keystone endpoint11:28
*** lucasagomes is now known as lucas-hungry11:28
jaosoriorthe kestone.initialize does the setting up of admin roles and such, but that shouldn't be a problem11:28
shardymgould: we need to implement the equivalent workflow via mistral, which is a workflow engine that interacts with openstack APIs11:29
*** electrofelix has quit IRC11:29
mgouldshardy, dtantsur: I'll do the refactor as a follow-up patch, and if there's no need to backport we can squash it into the current patch11:29
shardywhere needed we can also implement custom mistral actions, specific to tripleo, which will live inside of tripleo-common11:29
thrashjaosorior: I'm pretty sure, yeah. We saw the same outcome in downstream CI.11:29
mgouldshardy, thanks11:29
thrashjaosorior: you can see that the stack is complete, but then you get that 401 error. I'm pretty sure that's the culprit.11:30
mgouldso "port logic to mistral where possible, add endpoints where necessary"?11:30
thrashjaosorior: I've tried it in a manual session, and that's the outcome I got.11:30
jaosoriorfunky11:30
jaosoriorthrash: Now I'm confused how that works in master though11:30
shardymgould: Yes, although a new action isn't a new endpoint11:30
thrashjaosorior: me too...11:30
mgouldoh, OK11:31
*** electrofelix has joined #tripleo11:31
shardyit's just a python plugin that mistral can excecute11:31
shardySee the other patches by rbrady11:31
* mgould pulls up the mistral docs11:31
*** trozet has quit IRC11:31
shardypretty soon we're going to have to refactor *all* business logic out of tripleoclient into that system11:31
dtantsurmgould, I don't think we'll need to backport it11:31
shardyderekh: are you happy if we just land mgould's patch for now, to remove the pressure re the ironic changes11:31
shardythen we can refactor into tripleo-common as a subsequent patch11:32
*** florianf has quit IRC11:32
shardyit's passing CI so it seems like a good opportunity to just get it landed11:32
jaosoriorthrash... is something else in puppet creating the admin role?11:33
jaosoriorit might be that the admin role is created in puppet, in master, and so that's why that call works11:34
thrashjaosorior: possibly. But I don't know.11:35
*** ccamacho|lunch has quit IRC11:35
*** ccamacho has joined #tripleo11:36
jaosoriorthrash: well, if that's not working, then there is no option but to assume that the endpoints haven't been initialized, in stable/liberty, so I'll have to squash that CR with the next one that uses the values of the EndpointMap, and remove that check11:36
mgouldshardy, derekh: if you merged my patch as-is I would be very happy :-)11:36
*** electrofelix has quit IRC11:37
mgouldI'll happily submit a pair of follow-up patches to refactor it into tripleo-common11:37
*** electrofelix has joined #tripleo11:37
shardymgould: ack, I'm totally fine with that, I just wanted to discuss the plan re tripleo-common, thanks!11:37
*** electrofelix has quit IRC11:38
*** electrofelix has joined #tripleo11:38
openstackgerritmathieu bultel proposed openstack/tripleo-heat-templates: Add the default for the controller on single-nic-vlans  https://review.openstack.org/30536011:38
mgouldawesome :-) Is a patch refactoring it to python code OK, or should I try to port to Mistral templates at the same time?11:38
matbuis there any chance to get this patch reviewed https://review.openstack.org/#/c/305360/ ?11:39
*** rhallisey has joined #tripleo11:39
thrashjaosorior: should be pretty easy to tell if there is something in the puppet in master that does it.11:40
openstackgerritCarlos Camacho proposed openstack-infra/tripleo-ci: Re-write tripleo-jobs to use gerrit  https://review.openstack.org/30753211:40
*** panda has quit IRC11:41
*** panda has joined #tripleo11:41
derekhshardy: mgould I'm happy to merge it to remove pressure, the only reason I only did a +1 was because of my unfamiliarity with tripleoclient in general11:43
*** electrofelix has quit IRC11:44
*** electrofelix has joined #tripleo11:44
*** florianf has joined #tripleo11:45
mgouldmatbu, left some questions11:46
*** gfidente has joined #tripleo11:46
* shardy approves11:47
shardythanks derekh11:47
matbumgould: nice thanks11:47
*** dbecker has quit IRC11:47
shardymgould: just refactoring the python is fine as a first step11:48
mgouldshardy, will do, thanks!11:49
matbumgould: it's a good question ...11:49
matbumgould: and i'm not sure I can really answer11:50
openstackgerritJuan Antonio Osorio Robles proposed openstack/python-tripleoclient: Use values from endpoint map for service endpoints  https://review.openstack.org/30634611:51
* mgould points network-security folks at https://review.openstack.org/#/c/305360/2/network/config/single-nic-vlans/controller.yaml@10411:51
openstackgerritMerged openstack/python-tripleoclient: Use Ironic API v1.11 to support ENROLL state  https://review.openstack.org/27220611:52
mgould\o/11:52
dtantsurrebase party now :D11:54
jaosoriorthrash haha sorry, must have added those back as part of the squash11:55
openstackgerritJuan Antonio Osorio Robles proposed openstack/python-tripleoclient: Use values from endpoint map for service endpoints  https://review.openstack.org/30634611:56
thrashjaosorior: :)11:56
*** morazi has joined #tripleo11:56
ccamachoGuys, quick question. (#/301087) Are the CI tests randomly failing??. Sometimes ha fails, other times nonha, right now all except ha in my case (At the beginning they were green except for the upgrade issue last week). The solution for this is wait/recheck?11:57
*** dprince has joined #tripleo11:57
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart: Ansible 2.0.2 release broke get_url with file:// urls  https://review.openstack.org/30829011:58
jaosoriordprince: Hey dude, as per your comment from the other day, that it's quite a bummer to have to update both the endpoint_data and the enable-tls.yaml, I have this https://review.openstack.org/#/c/308182/11:58
*** MaxPC has joined #tripleo12:00
dprincejaosorior: cool, I'll check it out12:01
*** oshvartz has quit IRC12:02
jaosoriordprince: I made it to depend on gfidente's patch for adding the Mysql address to the endpoint map, so it wouldn't conflict12:02
*** dsariel has quit IRC12:03
*** zoli|brb is now known as zoli12:03
openstackgerritMerged openstack/tripleo-quickstart: Ansible 2.0.2 release broke get_url with file:// urls  https://review.openstack.org/30829012:03
*** hewbrocca-afk is now known as hewbrocca12:04
shardyccamacho: we have been suffering from a high number of false-negatives in CI for a while now12:07
openstackgerritDmitry Tantsur proposed openstack/python-tripleoclient: Allow import command to set deploy image and local boot  https://review.openstack.org/29936212:08
shardythe solution is ultimately to figure out why and fix the problems, but yes the passive solution is to wait and/or recheck12:08
*** dbecker has joined #tripleo12:08
shardyit's a good idea to check the logs and try to understand why the spurious failure happened, and report a bug if there isn't one already for the same issue12:08
jaosorior dprince: Glance is not being deployed over httpd, is it?12:11
dprincejaosorior: no, I don't think so12:12
jaosorioralright12:12
jaosorioroh, now I'm seeing this error in other places12:13
jaosoriorthere seem to be some issues initializing httpd12:13
jaosoriorthat's affecting a bunch of patches in master12:13
dprincejaosorior: yeah, I noticed that in my CI failures for the composable glance service patch too12:14
dprincejaosorior: tried hitting it locally, but I haven't yet :/12:14
jaosoriordprince: That's exactly what I was looking at. It also affected gfidente's mysql patch12:14
dprincejaosorior: ah, cool. Thanks for looking12:15
dprincejaosorior: I'll see what I can figure out as well12:15
jaosorioron the other hand, gnocchi seems to be failing too12:15
jaosoriorhttp://paste.openstack.org/show/494788/12:15
jaosoriorI see a lot of those12:16
gfidenteccamacho if same patchset in the collected attempts passed all jobs, I think you can comment it and maybe people can merge it12:16
*** bvandenh has quit IRC12:17
*** michchap_ has quit IRC12:17
*** michchap has joined #tripleo12:22
ccamachoshardy, gfidente.  I have checked the logs but wasn't able to find actual issues, i.e. ha job passed, then submit a new patch with a logged message and then ha fails.. I will try to re-check if something is failing I will try to have them all in green12:22
ccamachowill try to debug the logs12:23
shardyccamacho: sure - let us know if you need help figuring out the logs12:23
ccamachosure, thanks12:24
*** bvandenh has joined #tripleo12:29
*** michchap_ has joined #tripleo12:31
*** michchap has quit IRC12:32
sshnaidmmaybe you'll find it useful: https://cdn.rawgit.com/sshnaidm/various/tempr/index.html12:34
sshnaidmdprince, derekh ^^12:34
*** olap has quit IRC12:34
jaosoriordprince: shardy: any idea if this kind of thing would cause httpd not to start? Apr 20 13:53:09 overcloud-controller-0 python[4878]: ERROR:scss.compiler:Mixin not found: dropdown-arrow:0 (themes/default/bootstrap/components/_dropdowns.scss:61 (themes/default/bootstrap/components/_dropdowns.12:35
jaosoriorseems to be related to horizon12:35
dprincesshnaidm: I like the 'reason' column idea quite a bit12:35
dprincesshnaidm: I had a similar thing w/ smokestack a while back and it was quite helpful12:36
*** olap has joined #tripleo12:36
sshnaidmdprince, there were periodic jobs also, but broken since jenkins blocked12:36
dprincesshnaidm: I usually like to see things grouped by patches like http://tripleo.org/cistatus.html12:36
*** coolsvap has quit IRC12:37
dprincejaosorior: sounds related to horizon, it could be related...12:37
sshnaidmdprince, yeah, I scrape this page to find reasons :)12:37
shardyjaosorior: did horizon add a new dependency not yet referenced in the packaging?12:38
jaosoriorshardy: I honestly don't know :/12:39
*** snecklifter_ has joined #tripleo12:39
derekhsshnaidm: cool, looks like we need to find the reason for "Overcloud stack: FAILED."12:39
*** snecklifter has quit IRC12:40
sshnaidmderekh, sure, it never ends :)12:40
*** snecklifter_ is now known as snecklifter12:40
derekhso true12:40
dprincederekh: that is what jaosorior is talking about above. Could be a horizon related issue.12:40
*** snecklifter_ has joined #tripleo12:40
sshnaidmderekh, mostly it's for filtering out infra problems (red color)12:40
derekhdprince: ok12:41
*** oshvartz has joined #tripleo12:43
derekhsshnaidm: just took a look at one of the red problems12:44
derekhsshnaidm: it looks weird, postci looks like it never returned http://logs.openstack.org/87/301087/8/check-tripleo/gate-tripleo-ci-f22-ha/cd536f7/console.html12:44
derekhsshnaidm:  actually maybe the process didn't finish because geard went down (as it did yesterday)12:45
*** jpena is now known as jpena|lunch12:45
*** links has quit IRC12:46
sshnaidmderekh, yes, looks weird12:46
*** rlandy has joined #tripleo12:47
derekhsshnaidm: so one of our compute nodes (where ci is running), went down yesterday, it was the compute node that hosted our squid and geard instances, this should explain a lot of the infra issues for a few hours yesterday12:47
sshnaidmderekh, ok, great12:48
jaosoriorderekh, dprince, shardy: What version of horizon are we using?12:49
derekhjaosorior: it should be the pined version from last friday12:50
jaosoriorderekh: uh... not sure which that is12:50
*** lucas-hungry is now known as lucasagomes12:51
derekhjaosorior: from this repo http://trunk.rdoproject.org/centos7/current-tripleo/12:51
*** mgould is now known as mgould|lunch12:55
jaosoriorderekh: Thanks12:58
*** pradk_ has joined #tripleo12:59
derekhjaosorior: actually if your looking for a git hash, in that repo you'll see a file called versions.csv , it contains the git hash for each package12:59
jaosoriorderekh: Found it12:59
derekhjaosorior: ok12:59
jaosoriorI'm trying to figure out what's up, but haven't gotten any info from the horizon folks12:59
jaosoriorand nobody answered in #rdo :/12:59
*** tzumainn has joined #tripleo13:00
*** trown is now known as trown|brb13:00
jaosoriorbut anyway, hopefully someone will know something O_O13:00
openstackgerritMerged openstack-infra/tripleo-ci: tripleo.sh --delorean-build handle oslo.* package builds  https://review.openstack.org/30231213:03
openstackgerritJames Slagle proposed openstack/puppet-tripleo: Add destination parameter to firewall rule  https://review.openstack.org/30831713:04
*** Goneri has joined #tripleo13:06
ccamacho sshnaidm, really nice that URL!13:09
*** lblanchard has joined #tripleo13:09
jaosoriorderekh: nah... no clue what's up there :/13:14
*** weshay has joined #tripleo13:15
dprincejaosorior: we can disable horizon if it blocks us too13:15
dprincejaosorior: have you filed a bug yet?13:15
jaosoriordprince: well, I'm still trying to figure out if that's the real issue or a red herring13:16
*** mkovacik_ has joined #tripleo13:17
*** Ryjedo_ has joined #tripleo13:17
*** hrybacki has quit IRC13:17
*** cwolferh has quit IRC13:17
*** mburned has quit IRC13:18
*** mburned has joined #tripleo13:18
*** hrybacki has joined #tripleo13:18
*** trown|brb is now known as trown13:19
*** mkovacik has quit IRC13:19
*** Ryjedo has quit IRC13:19
*** Ryjedo_ is now known as Ryjedo13:19
*** cwolferh has joined #tripleo13:20
*** dustins has joined #tripleo13:21
*** mcornea_ has quit IRC13:22
*** ccamacho has quit IRC13:22
*** ccamacho has joined #tripleo13:25
*** olap has quit IRC13:27
openstackgerritMerged openstack/python-tripleoclient: Use wait_for_finish from python-ironic-inspector-client  https://review.openstack.org/29937713:28
*** olap has joined #tripleo13:29
*** jobewan has joined #tripleo13:33
*** saneax is now known as saneax_AFK13:36
*** egafford has joined #tripleo13:39
*** ccamacho has quit IRC13:43
*** ccamacho has joined #tripleo13:44
*** pradk has quit IRC13:46
openstackgerritDmitry Tantsur proposed openstack/python-tripleoclient: [WIP] Allow 'baremetal import' command to also start introspection  https://review.openstack.org/29938813:47
*** mcornea has joined #tripleo13:47
*** jpena|lunch is now known as jpena13:48
*** ebalduf has joined #tripleo13:49
*** ayoung has joined #tripleo13:50
*** jprovazn has quit IRC13:50
*** links has joined #tripleo13:52
openstackgerritJuan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Avoid repetition in endpoint map  https://review.openstack.org/30818213:52
*** jobewan has quit IRC13:53
jaosoriorayoung: dumb question, if a vhost (say horizon) fails to load, will that lead to httpd itself not loading?13:53
ayoungjaosorior, "fails to load" meaning what.\13:54
ayoungjaosorior, use apachectl configtest13:54
jaosoriorayoung: Meaning there is an issue with the build and it throws an error13:54
ayoungjaosorior, its like "Goodnight Saigon"  they will all go down together.13:54
*** trozet has joined #tripleo13:55
jaosoriorayoung: that makes sense13:55
trowndprince: in order to use https://github.com/openstack/tripleo-common/blob/master/scripts/upload-swift-artifacts to use puppet modules from the undercloud on the overcloud do I need to create the tarball in a special way wrt filepaths?13:55
jaosoriorthanks dude13:55
*** absubram has quit IRC13:56
ayoungjaosorior, I'm serious about the apachectl approach.  it makes it so much easier to deal with issues there.  What are you seeing?13:56
jaosoriorayoung: This: Apr 20 13:53:09 overcloud-controller-0 systemd-journal[491]: Suppressed 534 messages from /system.slice/httpd.service14:01
jaosoriorApr 20 13:53:09 overcloud-controller-0 python[4878]: ERROR:scss.compiler:Mixin not found: dropdown-arrow:0 (themes/default/bootstrap/components/_dropdowns.scss:61 (themes/default/bootstrap/components/_dropdowns.14:01
dprincetrown: https://review.openstack.org/#/c/299643/14:01
jaosoriorApr 20 13:53:09 overcloud-controller-0 python[4878]: ERROR:scss.compiler:Mixin not found: dropdown-arrow:0 (themes/default/bootstrap/components/_dropdowns.scss:61 (themes/default/bootstrap/components/_dropdowns.scss:61 (themes/default/bootstrap/components/_dropdowns.scss:42))14:01
trownhow germane :)14:01
trowndprince: I will give that a go. thanks!14:01
ayoungjaosorior, so that is thrown when attempting to start Apache?14:02
jaosoriorayoung: apparently. I see nothing (no file or anything) in the apache logs. And that's the closest thing I have14:02
jaosoriorayoung: Haven't been able to reproduce that in my environment, and that log is from the CI14:02
ayoungjaosorior, ah14:03
ayoungjaosorior, ask in #openstack-horizon14:03
jaosoriorayoung: I did14:03
ayoungjaosorior, Ah.14:03
jaosorioranswer was: "can't reproduce it, could be a bad build"14:03
*** mgould|lunch is now known as mgould14:04
*** jaosorior has left #tripleo14:09
*** jaosorior has quit IRC14:09
*** jaosorior has joined #tripleo14:09
trowndprince: slagle: shardy, Is jprovazn working on tripleo still? If not, any objection to me taking over packaging of tripleo-common? https://github.com/redhat-openstack/rdoinfo/blob/master/rdo.yml#L72514:11
trownI would like to actually package these helper scripts we have14:11
shardytrown: I don't believe so, so I'd say go for it14:12
trowncool14:13
dprincetrown: go for it man14:13
slagletrown: i don't think he is. he stepped down from core14:14
*** jaosorior has quit IRC14:16
*** mbound has quit IRC14:20
openstackgerritJames Slagle proposed openstack/puppet-tripleo: Add destination parameter to firewall rule  https://review.openstack.org/30831714:20
openstackgerritJames Slagle proposed openstack/puppet-tripleo: Add dport/sport parameter to firewall rule  https://review.openstack.org/30835814:20
slagleEmilienM: i am writing puppet like a mad fool today ^14:21
EmilienMI saw that14:22
EmilienMyou ok?14:22
slagleand by "writing puppet" i mean copy/pasteing patterns from other modules14:22
EmilienMgood. As long we don't copy/paste wrong code :P14:22
*** paramite has quit IRC14:28
hewbroccaslagle: code reuse14:29
hewbroccaIt's a best practice14:29
EmilienMyeah, so we dupplicate bugs14:30
*** absubram has joined #tripleo14:32
*** jprovazn has joined #tripleo14:32
*** dprince has quit IRC14:33
openstackgerritOpenStack Proposal Bot proposed openstack/python-tripleoclient: Updated from global requirements  https://review.openstack.org/26852814:33
*** yamahata has joined #tripleo14:36
openstackgerritMerged openstack-infra/tripleo-ci: Conditionally request a test env from geard  https://review.openstack.org/30650614:37
*** ccamacho has quit IRC14:37
*** electrofelix has quit IRC14:39
*** pradk_ is now known as pradk14:39
pradkcan i get some +2's on this https://review.openstack.org/#/c/304295/ .. just a backport14:39
*** ccamacho has joined #tripleo14:39
*** jaosorior has joined #tripleo14:40
jaosoriorthrash: Still around?14:40
thrashjaosorior: yes14:40
jaosoriorthrash: Hey dude, so what do you think of this one? https://review.openstack.org/#/c/306346/14:40
jaosoriorsquashed the commits and removed the check14:40
jaosoriorbut AFAIK, for stable/liberty the upgrades gate is not passing14:41
thrashjaosorior: for anything?14:41
openstackgerritJames Slagle proposed openstack/instack-undercloud: Create firewall rules with puppet  https://review.openstack.org/30837414:41
thrashat least ha and nonha went green.14:41
jaosorioryeah14:41
openstackgerritSteven Hardy proposed openstack/instack-undercloud: Work around deletion of _member_ role assignments on upgrade  https://review.openstack.org/30735214:41
jaosoriorshardy, bnemec: That was the case, right? ugprades gate is not passing for stable/liberty14:41
*** electrofelix has joined #tripleo14:42
slagleshardy: what do you think about moving all the projects off of release:independent?14:43
jaosoriorthrash: well, that's what I remember. Though, also considering that that patch shouldn't even run on upgrades (since it's post CI)14:43
jaosoriorI mean14:43
jaosoriorconst config14:43
jaosorior*post config14:43
jaosoriordammit haha14:43
shardyslagle: Yeah, I was thinking about that & decided to start with those we branch14:43
thrashjaosorior: right... because existing stack.14:43
shardyslagle: we could but I wasn't sure how we handle all the components which aren't really tied to a release14:43
bnemecjaosorior: I wasn't aware that the upgrades job was broken on liberty, but that doesn't mean it isn't.14:44
shardythey're hard-dependencies but the release: tags all seem related to a cadence aligned with the Openstack coordinated release14:44
thrashbnemec: jaosorior this particular job failed at pingtest after upgrade14:45
bnemecI thought there was an issue yesterday where we were incorrectly using the cached images for stable jobs, but that should be fixed now.14:45
slagleshardy: i was thinking release:cycle-with-intermediary14:45
jaosoriorbnemec: But that was making all gates fail for stable/liberty14:45
jaosoriorthrash: Well, I could recheck that patch, maybe I remember wrong. That's why I pinged bnemec and shardy14:45
bnemecjaosorior: Yeah, but that was the only stable problem I know of.14:45
thrashjaosorior: do another recheck.14:46
jaosoriorthrash: I did14:46
thrashjaosorior: ack14:46
shardyslagle: Yeah, I guess we could - I wasn't sure about an artificial "final" version of e.g os*config, dib, etc, which are used outside the context of tripleo14:47
thrashjaosorior: I have put a +2 on based on the current runs, but obviously it needs to pass all before merge.14:47
openstackgerritJiri Stransky proposed openstack/tripleo-common: Don't perform replacement of OS::Neutron::Port  https://review.openstack.org/30838114:47
*** athomas has quit IRC14:47
*** tosky has quit IRC14:47
shardyslagle: the independent tag appears to be there for those projects that do either continuous releases or some other cycle decoupled from the 6-monthly cycle14:48
bnemecjaosorior: How on earth did I end up as the author on that backport commit?14:48
shardyhence I ended up deciding to leave them for now - I'll be happy to seek further clarification from the release team tho14:48
hewbroccaEmilienM: bug reuse!14:48
hewbroccait's also a best practice14:48
slagleshardy: yes, but the independent tag seems to be implying some context that i wasn't aware of (based on the dib thread)14:48
slagleshardy: particularly, not "part of openstack the product"14:49
shardyslagle: Well, yeah, the recent subtext seems to be independently released things are just deps like anything else we pull in from pypi14:49
jaosoriorbnemec: squashed the commits and it ended up that way, sorry dude14:49
shardye.g they are somehow less-openstack, even if they're 100% developed and maintained by openstack teams14:50
openstackgerritJuan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Avoid repetition in endpoint map  https://review.openstack.org/30818214:50
shardyI find the argument puzzling tbh14:50
bnemecjaosorior: It's no problem, I was just surprised to see it.14:50
slagleshardy: me as well, perhaps it's worth seeking clarification14:50
thrashbnemec: blame all the things14:50
bnemecjaosorior: Why did you squash commits though?14:50
thrash:D14:50
shardyslagle: let's ask in openstack-release14:50
bnemecThat's going to make it harder to track down what got backported to liberty.14:50
openstackgerritJuan Antonio Osorio Robles proposed openstack/python-tripleoclient: Use values from endpoint map for service endpoints  https://review.openstack.org/30634614:51
jaosoriorbnemec: Cause the commit that was adding the keystone check started causing issues14:51
*** athomas has joined #tripleo14:52
*** zoli is now known as zoli|mtg14:52
jaosoriorbnemec: So thrash found out that the check for the nova user was getting an error: Unauthorized call. Or something of the sort. And it was probably because there is no admin role at that point14:52
jaosoriorso as part of the squashing of that commit, I ended up removing that part, and creating the keystoneclient instance after the keystone.initialize call14:53
*** tosky has joined #tripleo14:54
bnemecjaosorior: Okay, so what mostly happened is that you removed the init check, but because your patch was dependent on it the refactoring part was needed anyway.14:55
bnemecIt might be good to mention that in the commit message.14:55
thrashbnemec: there was a 401 Unauthorized message immediately after the STACK_COMPLETE14:57
thrashthe other option was to just catch the exception there, and assume that the exception meant it was not initialized.14:57
*** dprince has joined #tripleo14:59
dprinceslagle: upgrades job looks to have passed again https://review.openstack.org/#/c/237370/3815:02
jaosoriordprince: Yeah dude, seems that the issue was intermitent15:04
*** ErmingPei has joined #tripleo15:04
slagledprince: awesome15:05
*** florianf has quit IRC15:05
*** bvandenh has quit IRC15:05
sshnaidmderekh, does postci function works now? I suspect it still does not..15:07
sshnaidmhttp://logs.openstack.org/28/268528/36/check-tripleo/gate-tripleo-ci-f22-ha/379bd4b/console.html#_2016-04-20_11_50_07_60915:07
*** TSCHAK has joined #tripleo15:07
*** florianf has joined #tripleo15:08
beaglesgiven that it afaik it would only affect POC/virtual setups, I doubt that this is all that important, but it seems that if you configure cinder in a test/demo type configuration similar to packstack using a loopback device with lvm, there isn't anything to reconfigure the loopback device when you reboot the controller. Am I missing something?15:09
*** rajinir has joined #tripleo15:12
NobodyCamoh I've been removed for the OOO organization :(15:13
*** michchap_ has quit IRC15:15
dprinceNobodyCam: ha :). If you want back in no worries15:15
dprinceNobodyCam: I just cleaned out all the repos, etc15:16
* NobodyCam feels empty with out it15:17
*** michchap has joined #tripleo15:17
openstackgerritJiri Stransky proposed openstack/tripleo-common: Don't perform replacement of OS::Neutron::Port  https://review.openstack.org/30838115:18
EmilienMjaosorior, dprince: could we land that please ? https://review.openstack.org/#/c/286344/15:19
dprinceNobodyCam: no activity there for years so I cleaned house... thats all really15:20
*** olap has quit IRC15:21
NobodyCamdprince: I under stand,15:21
dprinceEmilienM: +215:25
*** dustins has quit IRC15:27
*** jaosorior has quit IRC15:29
openstackgerritSteven Hardy proposed openstack/instack-undercloud: Make undercloud telemetry services optional and disabled  https://review.openstack.org/30289615:31
bnemecsshnaidm: postci output got moved to a file in the logs directory.15:31
bnemecThere should be a message pointing at the file though, like there is for the other places we do that.15:32
*** links has quit IRC15:32
*** jdob has quit IRC15:33
openstackgerritEmilien Macchi proposed openstack/instack-undercloud: run nova-api with mod_proxy_uwsgi httpd module  https://review.openstack.org/30743115:33
*** ccamacho has quit IRC15:34
openstackgerritBen Nemec proposed openstack-infra/tripleo-ci: Add message pointing to postci.log  https://review.openstack.org/30841915:34
bnemecsshnaidm: ^15:34
*** ccamacho has joined #tripleo15:36
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Don't remove packages from the delorean-deps repository  https://review.openstack.org/30842115:36
sshnaidmbnemec, I see it as logs/postci.txt, no postci.log  http://logs.openstack.org/94/304094/1/check-tripleo/gate-tripleo-ci-f22-ha/fabe86b/logs/15:36
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Remove metric from previous runs  https://review.openstack.org/30842215:36
*** ccamacho has quit IRC15:36
bnemecsshnaidm: Oops, wrong extension.  Fixing.15:37
sshnaidmbnemec, why to do it, btw? Before this we could see reasons of fail in the console15:37
sshnaidmnow for heat resources you need go to different files15:37
openstackgerritMerged openstack/puppet-tripleo: IPv6 dual-stack support  https://review.openstack.org/28634415:37
EmilienMdsneddon: ^15:38
openstackgerritBen Nemec proposed openstack-infra/tripleo-ci: Add message pointing to postci.txt  https://review.openstack.org/30841915:38
EmilienMdsneddon: which means, it's now on THT to be able to provide an array of VIPs and dual stack will be possible15:38
bnemecsshnaidm: The rationale seems to be in the original commit message: https://review.openstack.org/#/c/296375/15:39
*** ccamacho has joined #tripleo15:39
*** ccamacho has quit IRC15:39
*** ccamacho has joined #tripleo15:41
*** panda has quit IRC15:41
*** panda has joined #tripleo15:42
*** mikelk has quit IRC15:52
ErmingPeiHas anybody tried to use TripleO to upgrade Openstack? How does that work? Do we need redeploy an undercloud and then destroy the existing openstack to deploy a new one or it can upgrade smoothly a new one, with keeping all the existing stuff?15:53
*** michchap_ has joined #tripleo15:57
EmilienMtrown: I found a lot of issues with newton trunk15:58
EmilienMwhat is the etherpad again?15:59
thrashErmingPei: there is definitely work to upgrade.15:59
trownEmilienM: https://etherpad.openstack.org/p/delorean_master_current_issues15:59
EmilienMtrown: thx man15:59
ErmingPeithrash: have you tried? what's the rationale to upgrade openstack?15:59
thrashErmingPei: you upgrade the undercloud, then from there, you use heat to upgrade the overcloud16:00
*** michchap has quit IRC16:01
EmilienMtrown: see the nova-schedule issue16:01
EmilienMhttp://logs.openstack.org/86/305886/12/check/gate-puppet-openstack-integration-3-scenario002-tempest-centos-7/5d57690/logs/nova/nova-scheduler.txt.gz#_2016-04-20_14_07_17_45716:01
thrashErmingPei: it's relatively new functionality, so obviously I would recommend you try the upgrade on a non-production uc/oc16:01
thrashAnd I've just realized we have zero docs on upgrading. :|16:02
ErmingPeithrash: makes sense. Does heat upgrade or re-deploy a new openstack?16:02
ErmingPeithrash: makes sense. Does heat upgrade or re-deploy a new openstack?16:02
shardyErmingPei: it is possible to upgrade the existing openstack, but we're still working out the docs etc that show how to do it16:02
shardyjistr: ^^ do you have any WIP docs you can direct ErmingPei to?16:03
thrashErmingPei: upgrade16:03
ErmingPeishardy: thanks. a doc would be great16:03
ErmingPeishardy: in the commercial version "Director". Is the upgrade functionality ready to use?16:04
*** aufi_ has quit IRC16:05
*** tremble has quit IRC16:07
jistrErmingPei: hi, unfortunately as thrash said, we don't have docs yet. What would you be upgrading from/to?16:07
ErmingPeie.g. from kilo to Liberty of mitaka16:08
ErmingPeior16:08
ErmingPeisorry16:08
*** dtantsur is now known as dtantsur|afk16:08
ErmingPeijistr: e.g. from kilo to liberty or to mitaka16:08
thrashErmingPei: i definitely would not go directly from kilo to mitaka.16:09
ErmingPeithrash: why? big difference?16:10
shardyErmingPei: upstream there is only liberty and mitaka branches16:10
jistrand upgrades are TBD on liberty -> mitaka. There are some open questions like Keystone WSGI / Ceilometer AODH migrations. https://etherpad.openstack.org/p/tripleo-migrations16:12
ErmingPeijistr: I see. thanks!16:13
*** ErmingPei is now known as Erming__16:15
*** cwolferh has quit IRC16:15
*** Erming__ is now known as EPei16:15
shardyEPei: We now have an upgrades CI job, but it's not yet fully testing the version-to-version upgrade process16:16
shardyfirst step will be fully documenting it, then we can ensure it's fully tested in TripleO CI16:17
EPeishardy: I guess it's the same situation in the commercial version, 'Director'?16:17
*** ohamada has quit IRC16:18
shardyEPei: No, the CI and documentation systems are different to TripleO upstream16:19
shardyI can't comment on the latest status of the downstream builds, because I haven't personally tested them16:20
*** olap has joined #tripleo16:20
shardyhere it's most appropriate to discuss the upstream community builds, and their status IMO16:20
jistrgotta run for an appointment, will read chat log later and will try to propose WIP upstream upgrade docs if i get the time (the docs will still need to be amended to reflect the solution for migrations before they're of any use for liberty->mitaka, but we can get them started now)16:22
shardyjistr: sounds good, thanks!16:22
*** mcornea has quit IRC16:22
*** jistr has quit IRC16:22
*** rlandy has quit IRC16:23
shardyslagle: Hey, I'm looking at adjusting the tripleo launchpad project so we have per-release series, so we can track features/bugs per-release16:24
shardyatm there's just one "tripleo" team, which doesn't map to the core group - other projects have a "drivers" group assigned to maintain the org16:24
shardyare you OK with me adding such a group, so we can accept requests from anyone to join the tripleo "team" but only those in the drivers group can adjust e.g series allocation of wishlist bugs and blueprints?16:25
shardyI think that will align better with most other projects, but wasn't sure on the history of why it's set up as it is16:26
*** bnemec has quit IRC16:26
shardydprince: also we could change the branding of https://launchpad.net/tripleo to include the owl ;)16:31
shardythere's a logo option under "change branding"16:31
*** dmacpher is now known as dmacpher-afk16:32
*** cwolferh has joined #tripleo16:32
dprinceshardy: yep, let me look16:34
dprinceshardy: okay, needs to be a 14x14. Let me create that16:35
*** rlandy has joined #tripleo16:35
*** bnemec has joined #tripleo16:35
dprinceshardy: always good to jump on the important things....16:35
shardyIt's all about the owl! :)16:36
*** dustins has joined #tripleo16:37
*** penick has joined #tripleo16:37
*** leanderthal is now known as leanderthal|afk16:37
*** adarazs has quit IRC16:39
*** lucasagomes is now known as lucas-dinner16:40
dprinceshardy: boom https://launchpad.net/tripleo16:41
shardy\o/16:41
derekhshardy: you had dprince  at "we could add the owl"16:42
*** adarazs has joined #tripleo16:42
dprincederekh: yeah well... that was a < 5 minute investment :)16:42
*** derekh has quit IRC16:44
jrist:)16:45
*** florianf has quit IRC16:46
*** Guest67082 has quit IRC16:46
hewbroccadprince: NICE all about the owl16:48
hewbroccajrist: man, that should have been on the shirt16:49
jristwhat16:49
hewbrocca"All about the owl"16:49
jrist"It's all about the owl!"16:49
jristah16:49
jristwell16:49
hewbroccayesh :D16:49
jristI asked for slogans :)16:49
jristbtw it shipped HenryG16:49
jristhewbrocca: ^16:49
hewbrocca"We're all about the owl"?16:49
hewbroccaexcellent!16:49
* shardy can't wait to get his Owl shirt :)16:50
HenryGjrist: for that slip-up you must send me an owl too ;)16:50
jristHenryG: will you be at Summit? talk to hewbrocca :)16:51
jristthere are a few extras16:51
hewbroccayep :)16:51
* HenryG will find hewbrocca at the summit16:52
openstackgerritMerged openstack-infra/tripleo-ci: Upload the overcloud images to cache  https://review.openstack.org/30649916:54
*** chem has joined #tripleo17:00
*** electrofelix has quit IRC17:05
*** mbound has joined #tripleo17:06
*** tosky has quit IRC17:07
*** fragatina has joined #tripleo17:08
hewbrocca"Owls, all the way down"17:09
hewbroccathat's probably a little too obscure17:10
trownlol17:11
trownthat is awesome though17:11
openstackgerritJason E. Rist proposed openstack/tripleo-ui: Modal fixes: header, footer, width, fluidity  https://review.openstack.org/30439017:12
*** yamahata has quit IRC17:12
*** EPei has quit IRC17:13
*** athomas has quit IRC17:13
slagleshardy: sure, a drivers team sounds good17:16
openstackgerritBrent Eagles proposed openstack-infra/tripleo-ci: Fail pingtest more quickly  https://review.openstack.org/30848617:17
shardyslagle: thanks17:17
*** penick has quit IRC17:22
*** jpena is now known as jpena|off17:22
*** penick has joined #tripleo17:22
*** zoli|mtg is now known as zoli|gone17:23
*** zoli|gone is now known as zoli_gone-proxy17:25
*** ron____ has joined #tripleo17:25
*** mbound has quit IRC17:26
*** trown is now known as trown|lunch17:27
*** rlandy has quit IRC17:28
*** penick has quit IRC17:29
*** mgould is now known as mgould|afk17:32
*** yamahata has joined #tripleo17:32
*** dustins has quit IRC17:33
* beagles boogles a cinder API blithely responding "creating" status forever even if there are no volume services available17:33
*** jdob has joined #tripleo17:37
*** rlandy has joined #tripleo17:38
*** fragatina has quit IRC17:42
*** hjensas has quit IRC17:42
*** fragatina has joined #tripleo17:42
openstackgerritAbel Lopez proposed openstack/diskimage-builder: Change to latest CentOS-6 image  https://review.openstack.org/30850417:46
*** sambetts is now known as sambetts|afk17:46
openstackgerritBen Nemec proposed openstack-infra/tripleo-ci: Add message pointing to postci.txt  https://review.openstack.org/30841917:48
*** ccamacho has quit IRC17:51
*** dsariel has joined #tripleo17:51
*** ifarkas has quit IRC17:51
*** myoung has quit IRC17:54
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Call ::sync instead of ::mysql to perform 'db sync'  https://review.openstack.org/30850517:54
*** shivrao has joined #tripleo17:57
*** gfidente has quit IRC17:57
openstackgerritMerged openstack/instack-undercloud: Add a reload of keepalived  https://review.openstack.org/30769917:58
openstackgerritMerged openstack/diskimage-builder: Don't stop dib-lint on first flake8 failure  https://review.openstack.org/30651018:00
bnemec\o/18:02
dsneddonEmilienM, I am unclear on whether a variant of this patch is still required with the modifications you made: https://review.openstack.org/#/c/289279/18:02
dsneddonEmilienM, I think it will, because that's what will provide the array, right?18:03
EmilienMdsneddon: I don't think so. The only thing we need is an array for the vip18:03
EmilienMdsneddon: you can wheter or not provide an array18:03
EmilienMbut if you do so, you enable dual stack18:03
dsneddonEmilienM, Yeah, but enabling dual-stack still requires that both IPv4 and IPv6 are enabled on the underlying interfaces, which is what my patch does. I'll rebase it. I also need to test it, preferrably with your patch as well.18:04
EmilienMdsneddon: my patch is merged in master, so you don't needs depends-on anymore18:04
dsneddonEmilienM, Oh, great news! But not 100% sure that I'll have time to test that before the summit.18:05
openstackgerritBen Nemec proposed openstack-infra/tripleo-ci: Add message pointing to postci.txt  https://review.openstack.org/30841918:05
bnemecLet's see if I managed to not screw up ^ this time18:06
openstackgerritDan Prince proposed openstack/puppet-tripleo: Add neutron profiles  https://review.openstack.org/29343618:17
*** hjensas has joined #tripleo18:18
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron ml2 and ovs agent  https://review.openstack.org/30851418:23
openstackgerritJason E. Rist proposed openstack/tripleo-ui: Sidebar-pf nav-pills fixes  https://review.openstack.org/30851618:24
*** ayoung has quit IRC18:25
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron ml2 and ovs agent  https://review.openstack.org/30851418:31
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron dhcp service  https://review.openstack.org/30338618:31
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron metadata service  https://review.openstack.org/30361818:31
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron l3 service  https://review.openstack.org/30356218:31
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable glance services  https://review.openstack.org/23737018:31
*** egafford has left #tripleo18:34
*** egafford has joined #tripleo18:34
*** penick has joined #tripleo18:35
*** dsariel has quit IRC18:39
*** trown|lunch is now known as trown18:46
openstackgerritJuan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Make compute post-deploy depend on controller's  https://review.openstack.org/29855518:53
*** rcernin has quit IRC19:05
*** ccamacho has joined #tripleo19:18
*** colonwq is now known as colonwq_afk19:18
*** pradk has quit IRC19:24
*** rcernin has joined #tripleo19:31
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart: Remove heat worker config for post-mitaka releases  https://review.openstack.org/30854519:31
trownlarsks: ^ I dont totally love that... but our special low resource environment does not work on newton. maybe you have a better idea?19:33
* larsks looks19:33
trownideally we would get the release from the image, as that is what actually determines it... but I couldnt think of a way to do that with our current images19:34
*** pradk has joined #tripleo19:35
larsksDoes getting the release from the image make sense?  I mean, don't we already know the release first, because we need that to fetch the image?19:35
larsksIt is otherwise tricky to answer the question, "which release of openstack am I running", isn't it?  There's nothing like an "openstack-release" package, for example.19:36
trown ya if we are using an image from CI, then we can get it from the URL, but that doesnt help if an arbitrary image is passed in19:37
larsksAh, I see.  But I guess my question, given an arbitrary image, how would one even determine which release of openstack is installed?19:37
trownit actually would not be awful to write an /etc/openstack-release file in the image itself19:37
larsksYeah, I suggested that last year :)19:37
larsksNot just a file, but package (like the redhat-release or fedora-release packages, etc)19:38
larsksOr right...maybe just a file, for upstream.19:38
larsksI guess that makes more sense.19:38
larsksBut yeah, it would be super useful.19:38
trownya if it was a package, that helps other consumers too, but for tripleo-quickstart I really only care about the image19:38
larsksEspecially because the version information available via --version doesn't really return anything that is really human consumable (am I running mitaka?  I dunno, nova-api --version says "13.0.1").19:39
trownhmm... we have an rdo-release package already, and there is work underway for it to install the mirrored delorean repos as well... maybe that package should just write that file19:41
larsksWhat about non-RDO environments?19:42
*** panda has quit IRC19:42
trownthat will be a while at best though... so kinda need a stop gap19:42
*** panda has joined #tripleo19:42
larsksUsing the "release" variable seems like a fine stopgap.  It just means it needs to be set correctly when using an arbitary image (and is that even a common use case?).19:42
trownI do it all the time :), but arbitrary is a strong term, any local image would need to have the release matched19:44
trownthough the consequences of not setting it currently would only affect a newton image, which there is not one that has passed ci yet19:45
trownand we cant get one to pass ci without some version-aware fix19:47
*** jprovazn has quit IRC19:47
trownalso, once there is a newton stable image published, if we switch the default to newton, the consequences of not setting the release for a mitaka or liberty would be pretty minimal... the overcloud heat would just not have heat engine workers set19:49
openstackgerritJames Slagle proposed openstack/instack-undercloud: Create firewall rules with puppet  https://review.openstack.org/30837419:49
*** pkovar has quit IRC19:52
openstackgerritJames Slagle proposed openstack/instack-undercloud: Don't use os-cloud-config/common-venv elements  https://review.openstack.org/30855019:52
*** EPei has joined #tripleo19:52
*** dbecker has quit IRC19:56
*** dbecker has joined #tripleo19:57
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart: Remove heat worker config for post-mitaka releases  https://review.openstack.org/30854519:58
openstackgerritZhiQiang Fan proposed openstack/tripleo-specs: fix wrong link of specs/juno/template.rst  https://review.openstack.org/30855119:59
openstackgerritBrad P. Crochet proposed openstack/python-tripleoclient: Remove the orchestration property from tripleoclient class  https://review.openstack.org/30855220:00
openstackgerritMerged openstack-infra/tripleo-ci: Re-write tripleo-jobs to use gerrit  https://review.openstack.org/30753220:00
*** shardy has quit IRC20:03
*** snecklifter has quit IRC20:07
EPeiHi, anybody can shed light to me about TripleO release scheme? How often?20:08
*** shardy has joined #tripleo20:12
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart: Use mirrored image location in quickstart.sh  https://review.openstack.org/30856320:13
*** rbrady has quit IRC20:15
*** ccamacho has quit IRC20:18
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart: Use mirrored image location in quickstart.sh  https://review.openstack.org/30856320:19
*** ccamacho has joined #tripleo20:23
openstackgerritMerged openstack/os-cloud-config: Updated from global requirements  https://review.openstack.org/28504920:24
*** ccamacho has quit IRC20:28
*** ccamacho has joined #tripleo20:30
*** MaxPC has quit IRC20:34
openstackgerritZhiQiang Fan proposed openstack/tripleo-specs: fix wrong link of specs/juno/template.rst  https://review.openstack.org/30855120:35
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart: Remove heat worker config for post-mitaka releases  https://review.openstack.org/30854520:38
*** lucas-dinner has quit IRC20:40
*** rajinir has quit IRC20:45
*** EPei has quit IRC20:46
*** lucasagomes has joined #tripleo20:47
*** jobewan has joined #tripleo20:49
*** trown is now known as trown|outtypewww20:53
*** shardy has quit IRC20:55
openstackgerritMerged openstack/diskimage-builder: Allow skipping the md docs check  https://review.openstack.org/30651420:56
*** jdob has quit IRC20:57
*** dprince has quit IRC20:57
openstackgerritJames Slagle proposed openstack/puppet-tripleo: Add dport/sport parameter to firewall rule  https://review.openstack.org/30835821:11
*** colonwq_afk is now known as colonwq21:12
openstackgerritJames Slagle proposed openstack/puppet-tripleo: Add dport/sport parameter to firewall rule  https://review.openstack.org/30835821:12
*** Goneri has quit IRC21:13
*** lblanchard has quit IRC21:13
*** dbecker has quit IRC21:15
*** weshay has quit IRC21:17
*** ron____ has quit IRC21:20
*** hewbrocca is now known as hewbrocca-afk21:31
*** rhallisey has quit IRC21:37
openstackgerritJason E. Rist proposed openstack/tripleo-ui: Sidebar-pf nav-pills fixes  https://review.openstack.org/30851621:38
*** thrash is now known as thrash|g0ne21:41
*** egafford has quit IRC21:45
*** EPei has joined #tripleo21:47
openstackgerritMerged openstack/diskimage-builder: Change to latest CentOS-6 image  https://review.openstack.org/30850421:52
*** EPei has quit IRC21:52
*** olap has quit IRC21:58
*** shadower has quit IRC22:04
*** ccamacho has quit IRC22:04
*** ccamacho has joined #tripleo22:07
*** lblanchard has joined #tripleo22:08
*** marcusvrn_ has quit IRC22:13
*** ccamacho has quit IRC22:13
*** ccamacho has joined #tripleo22:15
openstackgerritMerged openstack/puppet-tripleo: Add Glance profiles  https://review.openstack.org/29607622:16
*** ayoung has joined #tripleo22:33
openstackgerritJason E. Rist proposed openstack/tripleo-ui: Modal-component + changes to relevant usages  https://review.openstack.org/30865722:34
*** yamahata has quit IRC22:49
openstackgerritJason E. Rist proposed openstack/tripleo-ui: Updates to the README.md referencing old stuff  https://review.openstack.org/30866122:50
*** lblanchard has quit IRC22:52
*** EPei has joined #tripleo22:54
*** krotscheck is now known as krotscheck_dcm22:57
*** EPei has quit IRC22:58
*** jobewan has quit IRC23:04
*** dujelly has joined #tripleo23:29
*** panda has quit IRC23:41
*** panda has joined #tripleo23:42
*** penick has quit IRC23:45
*** dsneddon has quit IRC23:54
« Tuesday, 2016-04-19 Index Thursday, 2016-04-21 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!