Wednesday, 2016-05-11

*** julim has joined #tripleo00:02
*** eil397 has quit IRC00:07
*** rooooook has quit IRC00:08
*** shivrao has quit IRC00:10
*** trozet has joined #tripleo00:20
*** chlong_ has joined #tripleo00:22
*** chlong_ has joined #tripleo00:27
*** chlong_ has quit IRC00:28
*** fpan has joined #tripleo00:29
*** jobewan has joined #tripleo00:43
*** trozet has quit IRC00:45
*** mbound has quit IRC00:45
*** dmacpher-afk has quit IRC00:49
*** crinkle has joined #tripleo01:03
*** saneax is now known as saneax_AFK01:07
*** penick has quit IRC01:08
*** ramishra_ has joined #tripleo01:12
*** ramishra_ has quit IRC01:32
*** dmacpher has joined #tripleo01:33
openstackgerritMerged openstack/tripleo-heat-templates: deployment: remove Step7  https://review.openstack.org/31424201:36
*** mbound has joined #tripleo01:46
*** yamahata has quit IRC01:49
openstackgerritDan Prince proposed openstack/puppet-tripleo: Add the neutron-dnsmasq.conf to neutron profile  https://review.openstack.org/31482701:50
*** mbound has quit IRC01:51
*** jobewan has quit IRC01:52
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron ml2 and ovs agent  https://review.openstack.org/30851401:53
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron dhcp service  https://review.openstack.org/30338601:53
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron metadata service  https://review.openstack.org/30361801:53
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron l3 service  https://review.openstack.org/30356201:53
openstackgerritGerard Braad proposed openstack/tripleo-quickstart: Correct a typo in post install template  https://review.openstack.org/31483002:16
*** TSCHAK_ has quit IRC02:17
*** yamahata has joined #tripleo02:28
*** absubram has joined #tripleo02:29
*** absubram_ has joined #tripleo02:30
*** absubram has quit IRC02:34
*** absubram_ is now known as absubram02:34
*** ccamacho has quit IRC03:04
*** absubram has quit IRC03:05
*** ccamacho has joined #tripleo03:05
*** tzumainn has quit IRC03:07
*** absubram has joined #tripleo03:08
*** links has joined #tripleo03:21
*** mbound has joined #tripleo03:30
*** mbound has quit IRC03:35
*** TSCHAK_ has joined #tripleo03:54
*** masco has joined #tripleo03:55
*** TSCHAK has quit IRC03:57
*** pcaruana has joined #tripleo04:02
*** stendulker has joined #tripleo04:09
*** weshay has quit IRC04:16
*** bvandenh has joined #tripleo04:34
*** yuanying has quit IRC04:44
*** rcernin has joined #tripleo04:57
*** rlandy has quit IRC05:03
*** jaosorior has joined #tripleo05:04
openstackgerritsandhya proposed openstack/diskimage-builder: Procedure to build UEFI based images  https://review.openstack.org/30765505:04
*** oshvartz has joined #tripleo05:07
*** coolsvap has joined #tripleo05:10
openstackgerritMerged openstack/puppet-tripleo: keystone: drop usage of step 6  https://review.openstack.org/31454305:13
*** veteran has joined #tripleo05:38
*** veteran has quit IRC05:39
*** veteran has joined #tripleo05:40
*** veteran has quit IRC05:40
*** absubram has quit IRC05:40
*** absubram has joined #tripleo05:41
*** aufi has joined #tripleo05:55
*** yuanying has joined #tripleo06:01
*** yuanying has left #tripleo06:13
*** TSCHAK has joined #tripleo06:18
openstackgerritBabu Shanmugam proposed openstack/puppet-tripleo: Add OVN support for neutron profile  https://review.openstack.org/31487506:18
*** mikelk has joined #tripleo06:19
*** TSCHAK_ has quit IRC06:20
*** david-lyle has quit IRC06:23
*** david-lyle has joined #tripleo06:27
*** apetrich has joined #tripleo06:32
*** apetrich has quit IRC06:32
*** apetrich has joined #tripleo06:33
*** dsariel has joined #tripleo06:39
*** florianf has joined #tripleo06:54
*** liverpooler has joined #tripleo06:57
*** fragatina has quit IRC07:00
*** fragatina has joined #tripleo07:00
*** liverpooler has quit IRC07:01
*** liverpooler has joined #tripleo07:02
*** jprovazn has joined #tripleo07:03
*** fragatin_ has joined #tripleo07:05
*** fragatina has quit IRC07:06
*** dmacpher has quit IRC07:06
-openstackstatus- NOTICE: pip 8.1.2 broke our local python mirror, some jobs will fail with "No matching distribution found". We're investigating. Do not "recheck" until the issue is solved07:08
*** ChanServ changes topic to "pip 8.1.2 broke our local python mirror, some jobs will fail with "No matching distribution found". We're investigating. Do not "recheck" until the issue is solved"07:08
*** rcernin has quit IRC07:09
*** rcernin has joined #tripleo07:10
*** fragatin_ has quit IRC07:10
openstackgerritsandhya proposed openstack/diskimage-builder: Add support for building images capable of UEFI  https://review.openstack.org/28778407:10
*** fragatina has joined #tripleo07:12
*** shivrao has joined #tripleo07:13
*** fragatina has quit IRC07:17
*** athomas has joined #tripleo07:21
*** olap has joined #tripleo07:21
*** fragatina has joined #tripleo07:24
openstackgerritMerged openstack/instack-undercloud: TripleO common for custom Mistral actions  https://review.openstack.org/31078207:25
*** jpena|off is now known as jpena07:28
*** fragatina has quit IRC07:28
openstackgerritMerged openstack/tripleo-heat-templates: deployment: drop step6  https://review.openstack.org/31425307:29
*** fragatina has joined #tripleo07:30
*** mcornea has joined #tripleo07:31
*** ifarkas has joined #tripleo07:37
*** ohamada has joined #tripleo07:39
*** anshul has joined #tripleo07:41
*** anshul is now known as Guest4245907:41
*** chlong has quit IRC07:49
*** gbraad has quit IRC07:50
*** devvesa has joined #tripleo07:50
*** gbraad has joined #tripleo07:50
*** ebalduf has quit IRC07:50
*** devvesa has quit IRC07:50
openstackgerritDougal Matthews proposed openstack/python-tripleoclient: Remove the orchestration property from tripleoclient class  https://review.openstack.org/30855207:51
*** devvesa has joined #tripleo07:51
*** milan has quit IRC07:54
*** ifarkas has quit IRC08:04
openstackgerritJaume Devesa proposed openstack/tripleo-puppet-elements: [WIP] Modify the midonet repositories to Mitaka  https://review.openstack.org/30265708:06
*** weshay has joined #tripleo08:07
*** devvesa has quit IRC08:08
*** dmk0202 has joined #tripleo08:08
*** devvesa has joined #tripleo08:10
*** paramite has joined #tripleo08:11
ccamachofolks morning! quick question do you know if anyone is working on the composable gnocchi? xD08:17
*** ifarkas has joined #tripleo08:19
mariosccamacho: o/ morning - fwiw don't see it claimed in https://etherpad.openstack.org/p/tripleo-composable-services08:20
openstackgerritJuan Antonio Osorio Robles proposed openstack/puppet-tripleo: Create keystone endpoints in keystone profile  https://review.openstack.org/30412508:21
ccamachomarios thanks, I just wanted to double check :)08:22
mariosack08:23
*** paramite is now known as paramite|afk08:28
*** dtrainor_ has joined #tripleo08:28
*** yamahata has quit IRC08:29
*** dtrainor has quit IRC08:31
*** paramite|afk is now known as paramite08:32
*** shivrao has quit IRC08:34
*** jistr has joined #tripleo08:38
d0ugalmarios: Good spot re: the mistral requirement. I should have seen that!08:38
*** sambetts|afk is now known as sambetts08:39
*** paramite is now known as paramite|afk08:41
*** tremble has joined #tripleo08:41
*** tremble has joined #tripleo08:41
*** shardy has joined #tripleo08:43
*** gfidente has joined #tripleo08:44
mariosd0ugal: tx, just saw it  in the error log my feat was really not that impressive :)08:49
*** derekh has joined #tripleo08:51
d0ugalmarios: hah, but I looked at it about 20 times and didn't notice :)08:52
mariosheh yeah we've all done that. helps that it was first thing after spresso i guess. /me gets moar08:52
shardyMorning all08:53
openstackgerritMerged openstack-infra/tripleo-ci: Add planet blog aggregator feeds to website  https://review.openstack.org/31191008:54
openstackgerritGiulio Fidente proposed openstack-infra/tripleo-ci: Add Emilien Macchi blog to TripleO blogs feed  https://review.openstack.org/31356608:55
openstackgerritMerged openstack-infra/tripleo-ci: Add Emilien Macchi blog to TripleO blogs feed  https://review.openstack.org/31356608:57
jokke_morn08:57
*** milan has joined #tripleo09:02
*** nijaba has quit IRC09:06
*** mgould|afk is now known as mgould09:06
*** electrofelix has joined #tripleo09:06
*** nijaba has joined #tripleo09:13
*** tesseract has joined #tripleo09:20
*** fragatina has quit IRC09:24
*** fragatina has joined #tripleo09:25
*** dtantsur|afk is now known as dtantsur09:30
*** mdnadeem has joined #tripleo09:42
*** mdnadeem_ has quit IRC09:44
adarazsderekh: hi! I have a question about DLRN. I see that that what we're doing in tripleo.sh for mitaka, for example is setting distro=rpm-mitaka09:46
adarazsderekh: however, I just checked the "openstack-tripleo-heat-templates_distro" that was cloned by delorean and it doesn't have that branch, it has mitaka-rdo instead.09:47
*** tosky has joined #tripleo09:47
adarazsbut so in the end the branch is not changed from rpm-master in my run.09:47
adarazsis this a bug in tripleo.sh?09:47
*** zoli_gone-proxy is now known as zoliXXL09:48
*** rooooook has joined #tripleo09:51
derekhadarazs: hmmm, possibly, I can't look right now as I'm trying to get some prep work done for the HW upgrade, but worth looking into09:53
*** mdnadeem_ has joined #tripleo09:53
derekhadarazs: would you be able to look at the logs for a mitaka ci job to see if its possible to confirm a patch its testing was actually used09:54
adarazsderekh: I'm not sure there are verbose enough logs for it, but I take a look09:55
adarazsderekh: one more question: this is very sketchy (overwriting all branches with some commit): https://github.com/openstack-infra/tripleo-ci/blob/master/scripts/tripleo.sh#L342-L344 -- isn't there a way to tell DLRN to leave those checked out repos alone instead of something like this?09:55
*** mdnadeem has quit IRC09:56
derekhadarazs: there wasn't when we wrote that (developer mode --dev) went part of the way but wasn't suitable for some reason iirc09:57
derekhadarazs: maybe its now possible it been a long time since I look at what the RDO guys have added to delorean09:58
derekhadarazs: if its possible we shoudl use it (as I agree the branch rewrite thing is ugly)09:59
adarazsderekh: I'm using --dev move but it was still overwritten, so probably it's still not enough :)10:01
*** coolsvap_ has joined #tripleo10:01
*** mburned is now known as mburned_out10:01
adarazsderekh: oh, it's saying "Don't reset packaging git repo..." and we'd need to not reset the source repo instead.10:02
derekhadarazs: ahh, maybe it could be changed to not reset both10:02
*** athomas has quit IRC10:03
shardyit only resets the repo in ~/tripleo/delorean/data, which is deleted every build anyway, so does it matter?10:03
shardythe actual source repo is copied before the resetting happens?10:03
*** coolsvap has quit IRC10:04
adarazsshardy: yep. the actual delorean run resets the repo back to whatever is set in projects.ini for source=10:05
shardyah10:05
*** athomas has joined #tripleo10:08
adarazsI think even the origin/stable/mitaka needs to be overwritten like in tripleo.sh, as it checks it out from there directly.10:09
*** pino|work has quit IRC10:16
ccamachoderekh FYI http://logs.openstack.org/56/314056/2/check-tripleo/gate-tripleo-ci-f22-nonha/4791bb9/console.html  (There are not enough hosts available) from today10:18
*** zoliXXL is now known as zoli|lunch10:23
*** saneax_AFK is now known as saneax10:24
*** pino|work has joined #tripleo10:24
*** oshvartz has quit IRC10:29
*** mburned_out is now known as mburned10:30
derekhccamacho: this is a different problem to the one from yesterday (effecting the upgrades job), there may be some relevant logs in ironic or nova compute on the undercloud10:30
ccamachoack ill check it10:32
*** veteran has joined #tripleo10:37
*** veteran has quit IRC10:47
openstackgerritMerged openstack/puppet-tripleo: Add the neutron-dnsmasq.conf to neutron profile  https://review.openstack.org/31482710:49
*** dsariel has quit IRC10:49
*** morazi has joined #tripleo10:52
*** coolsvap_ has quit IRC10:57
*** zoli|lunch is now known as zoli10:58
*** jistr is now known as jistr|mtg11:00
*** ccamacho has quit IRC11:00
*** dprince has joined #tripleo11:06
*** stendulker has quit IRC11:06
*** ccamacho has joined #tripleo11:08
*** derekh changes topic to "================== TripleO CI cloud will be down for maintanance from 1300 UTC ===================="11:10
openstackgerritCarlos Camacho proposed openstack/tripleo-heat-templates: DO NOT MERGE TESTING stable/liberty  https://review.openstack.org/31497411:19
*** rodrigods has quit IRC11:20
*** rodrigods has joined #tripleo11:20
*** jaosorior has quit IRC11:22
*** jaosorior has joined #tripleo11:22
*** saneax is now known as saneax_AFK11:23
openstackgerritCarlos Camacho proposed openstack-infra/tripleo-ci: Fix validation for stable branch  https://review.openstack.org/31235011:26
sshnaidmderekh, hi, I'd like to clarify something - is it right that we use now in CI puppet modules from sources and not from RPM packages?11:29
*** fpan has quit IRC11:31
derekhsshnaidm: yes, for the mast branches we use puppet from source11:31
derekhsshnaidm: for the stable branches we're using the opm package but iirc some people are looking into changing it to sources11:32
sshnaidmderekh, thanks, and for liberty and mitaka - it's from delorean repos11:32
sshnaidmderekh, ok11:32
derekhsshnaidm: aslo, now that the puppet packages are split into individual packages per module, trown was going to look into switching to use those on master instead of git sources but that hasn't happened yet11:33
*** paramite|afk is now known as paramite11:34
sshnaidmderekh, and what are main reasons for this approach?11:34
sshnaidmderekh, less time to clone them?11:35
derekhsshnaidm: The most obvious reasons I can think of are that it should be quicker to build the images also it will more closely represent what we expect people to deploy in production11:36
*** dsariel has joined #tripleo11:36
*** pkovar has joined #tripleo11:37
sshnaidmok, clear now11:37
*** rhallisey has joined #tripleo11:38
*** panda has quit IRC11:42
*** panda has joined #tripleo11:42
*** ccamacho is now known as ccamacho|lunch11:43
*** zoli is now known as zoli|afk11:51
*** lucasagomes is now known as lucas-hungry11:52
*** ebarrera has joined #tripleo11:55
*** zoli|afk is now known as zoli11:56
EmilienMo/11:57
*** tosky has quit IRC11:59
-openstackstatus- NOTICE: We have a workaround for our mirrors to attempt to translate package names if a match isn't immediately obvious. A more complete fix is yet to come. It is now safe to 'recheck' any jobs that failed due to "No matching distribution found". Please join #openstack-infra if you discover more problems.12:00
*** zoli is now known as zoli|afk12:03
*** MaxPC has joined #tripleo12:03
EmilienMshardy, derekh: need your vote on https://review.openstack.org/#/c/314708/ please12:04
openstackgerritEmilien Macchi proposed openstack/puppet-tripleo: Composable role for RabbitMQ  https://review.openstack.org/31184912:08
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Deploy RabbitMQ as a composable role  https://review.openstack.org/31186312:09
EmilienMgfidente: hey, I have an alternative for https://review.openstack.org/#/c/314792/12:11
*** jistr|mtg is now known as jistr12:11
*** rbrady has joined #tripleo12:11
EmilienMgfidente: if we run db-sync in the same time as galera is prepared, but only on one node, when we're sure galera cluster is ready, it can work12:11
EmilienMgfidente: if we use clustercheck or galera-monitor12:13
*** fpan has joined #tripleo12:14
*** jayg|g0n3 is now known as jayg12:21
*** marcusvrn_ has joined #tripleo12:24
slagleEmilienM: is there any way to make puppet install all packages in a single yum call?12:26
slagleinstead of running yum a bunch of times12:26
hewbroccayou can no-op yum update :)12:27
EmilienMslagle: I don't think so and let me explain why12:27
hewbroccaalias it to /bin/true12:27
EmilienMpuppet / ansible / chef / whatever work the same :12:27
*** ccamacho|lunch is now known as ccamacho12:28
EmilienMyou declare somewhere (in a playbook, in a manifest, whatever) a list of package that you want to manage12:28
EmilienMfor each list (can be 1 or 1000 packages), the tool will manage it12:28
EmilienMAFIK there is no way to collect everything in one list and apply the policy (present, absent, latest, etc)12:28
*** dsariel has quit IRC12:29
EmilienMwhat we could investigate is to write a custom package provider12:29
*** fultonj has joined #tripleo12:29
EmilienMthat does not install packages, but actually build a file (yaml or whatever) with a global list of packages and their status12:29
EmilienMand sometimes in the catalog, manage this list once for all, using yum12:29
slagleEmilienM: yes, that's what i was wondering was already avaialble12:30
slaglegiven puppet compiles the catalog before it does anything12:30
EmilienMthe problem with that is resource ordering: sometimes (I don't have any example now) packages require some resource to be executed before12:30
slagleit knows the full list of packages12:30
slaglebut, if it's not already there, we probably dont need to spend time on it12:30
chemslagle: Emilien: I know that the saltstack yum "provider" has been enhance to support this kind of behaviour (one yum call instead of several)12:30
EmilienMand if we mix the resource, we can have dependency cycles12:30
shardyEmilienM: Hey, I think we need to investigate if that will actually work when https://review.openstack.org/#/c/312397/ lands - atm we'd build the module into the pre-update image, but really we should be using the cached image, then upgrading to the new module (which I'm not sure actually will happen with the current upgrades job implementation)12:30
chemslagle: EmilienM: I'm sure that we can collect all the required yum packages at puppet compile time from the catalog.  From there some kind of custom provider may be possible.12:32
* EmilienM in a meeting, bbl12:32
*** openstackgerrit has quit IRC12:33
*** openstackgerrit has joined #tripleo12:33
*** ayoung has quit IRC12:36
pkovardprince: hi, are you going to work on https://review.openstack.org/#/c/293091/1 ? i can fix the nits pointed out in the review so it can be merged12:37
*** jpena is now known as jpena|lunch12:42
openstackgerritMerged openstack/tripleo-heat-templates: composable neutron dhcp service  https://review.openstack.org/30338612:48
*** lucas-hungry is now known as lucasagomes12:49
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: composable neutron l3 service  https://review.openstack.org/30356212:51
*** links has quit IRC12:52
openstackgerritBrad P. Crochet proposed openstack/tripleo-heat-templates: composable heat services  https://review.openstack.org/31357712:54
*** ebalduf has joined #tripleo12:54
openstackgerritBrad P. Crochet proposed openstack/puppet-tripleo: Add Heat profiles  https://review.openstack.org/31301912:55
*** tzumainn has joined #tripleo12:58
*** Goneri has joined #tripleo12:58
*** mgould has quit IRC12:58
*** masco has quit IRC12:59
*** mgould has joined #tripleo13:00
*** milan_ has joined #tripleo13:00
openstackgerritBrad P. Crochet proposed openstack/tripleo-heat-templates: composable heat services  https://review.openstack.org/31357713:02
*** milan has quit IRC13:03
*** zoli|afk is now known as zoli13:06
openstackgerritBrad P. Crochet proposed openstack/puppet-tripleo: Add Heat profiles  https://review.openstack.org/31301913:08
EmilienMchem, slagle: (/me back from meeting) - I think we would need a custom package_provider OR using 1) deploying an image with all packages pre-installed 2) during upgrade, running yum one time 3) using noop as package_provider in catalog OR adding a step (would be step 0) that install all packages... but a bit hard to predict...13:08
*** Nakato has quit IRC13:08
*** Nakato has joined #tripleo13:08
*** rlandy has joined #tripleo13:08
openstackgerritMerged openstack-infra/tripleo-ci: Make the reviewday config file pluggable  https://review.openstack.org/30737713:09
chemEmilienM: slagle: was thinking of it, and having a yum provider based on the current existing one which check if package are installed without specific parameter and run it one time should be doable13:10
chemEmilienM: slagle: I could spend a couple of hours on it and come back here to tell you about it ... like tomorrow afternoon ?13:11
EmilienMchem: sure, feel free to investigate, if it can help us to save time13:12
EmilienMchem: we're trying to reduce deployments time13:12
chemEmilienM: ack, will toy with this tomorrow morning and see if it's doable quickly.13:13
*** lblanchard has joined #tripleo13:13
openstackgerritMerged openstack-infra/tripleo-ci: Ignore bash errors when posting metrics  https://review.openstack.org/30737413:13
*** pabelanger has quit IRC13:16
*** pabelanger has joined #tripleo13:16
openstackgerritMerged openstack/tripleo-puppet-elements: Add puppet-barbican to puppet-modules.  https://review.openstack.org/29400113:16
*** milan_ has quit IRC13:17
*** milan_ has joined #tripleo13:18
*** mgould is now known as mgould|brb13:18
*** mkovacik__ has joined #tripleo13:19
pabelangermornign13:21
pabelangererr13:21
pabelangermorning*13:21
pabelangerso, it sounds like tripleo-ci is getting an upgrade today13:21
pabelangerI also wanted to see what the status of the gate is too? I've noticed a lot of random failures related to the centos-7 patch I am testing.13:22
*** milan_ has quit IRC13:22
slaglewe haven't had a great pass rate the last couple of days13:23
slaglethe upgrades job was failing consistently until we identified the fix yesterday13:23
slagleother than that, it's been mostly spurious things13:23
slaglewhich we're hoping the hw upgrades are going to help address13:24
slaglenot sure if that's what you're seeing on your patch or not13:24
*** mkovacik__ is now known as milan13:24
pabelangersure, it looked like mostly timeouts, but never consistent pass rate. Patch in question: https://review.openstack.org/#/c/312725/13:24
pabelangerBasically, I were ready to have the discussion on when to remove fedora-22 slave and replace it with centos-713:25
pabelangeror in openstack-infra I've just finished moving all our images to DIB and would love to finish up by migrating tripleo too13:26
slaglelooks like you would have a couple runs where all jobs passed if the upgrades job hadn't been broken (since the 5th)13:26
*** mgould|brb is now known as mgould13:26
*** dmacpher has joined #tripleo13:27
openstackgerritJoe Talerico proposed openstack/tripleo-quickstart: Teardown network fix  https://review.openstack.org/31357513:28
derekhEmilienM: about your patch to switch to the upgrades job13:29
derekhEmilienM: I'm fine with it, but just want to make sure your happy that your swapping the nonha job for one that historically has been less reliable13:30
pabelangerslagle: Ya, they bounce back and forth13:30
derekhEmilienM: just wondering if you want to wait until after todays upgrade to see if the situation improves13:30
slaglepabelanger: i think we could probably merge it. might be good to wait until after the upgrade13:31
slaglejust to minimize the change going in as derekh brings our cloud back up :)13:31
pabelangerslagle: sure we can wait until the HDD are upgraded today, not an issue.13:32
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Dump IPs configuration as hieradata  https://review.openstack.org/31503513:33
gfidentedprince shardy ^^ that's the dumping part I mentioned in the call13:33
gfidentethis is pretty straight forward13:34
pabelangerI can do all the work to migrate tripleo JJB to centos-7, just need to know which day we want to land that change13:34
pabelangerI'm cleaning up some JJB this morning, so will get a WIP patch up for the migration13:34
EmilienMderekh: we want to improve our testing coverage and upgrade job is getting more and more reliable13:34
gfidentedprince shardy my problem was in looking up that data with %{hiera}13:34
EmilienMderekh: upgrade jobs is maybe less reliable than nonha, but bring more interesting feedback than nonha13:34
gfidentecause I can't get into the dict itself13:34
EmilienMderekh: the upgrade job will really help our modules to get CI/CD without introducing regressions13:35
derekhEmilienM: fair enough, just wanted to make sure you were aware that you may end up having to do more rechecks then you currently do13:35
derekhEmilienM: once I pointed that out I'm happy to +113:35
EmilienMderekh: shardy pointed some eventual issues with caching13:35
shardyderekh, EmilienM: I believe the issue atm is the upgrade job builds the image including the patch being tested, then does a deploy, then does a stack update (with the same image, and without triggering a yum update on the nodes)13:37
shardySo, we have to fix that or when we move to pre-cached overcloud images, I don't think the puppet modules will get tested at all?13:37
shardye.g we want to build an image from trunk (or use a pre-cached trunk image), then deploy, then trigger and update that pulls in the new puppet module13:38
*** chlong has joined #tripleo13:38
shardyAFAIK the current job does not do that13:38
shardyhttps://github.com/openstack-infra/tripleo-ci/blob/master/scripts/tripleo.sh#L50013:39
slaglethe only thing currently being tested by upgrades is tripleo-heat-templates13:39
derekhshardy: EmilienM using the precached image (as it is here https://review.openstack.org/#/c/312397/) , shouldn't change anything I can think of when compared to not using it13:40
slaglethat you can stack-update to new templates13:40
shardywe need to add the steps to expose the CI built delorean repo to the overcloud nodes, then drive an actual update between trunk and what is proposed13:40
shardyslagle: yup, which makes it possibly a worse choice (right now) for testing the puppet modules (than one of the normal non-upgrade jobs)13:40
derekhshardy: EmilienM the upgrades support in tripleo-ci should be fixed I think to start a deployment without what being tested then upgrade to the new patch but its not are puppet specific issue13:41
shardyhttps://review.openstack.org/#/c/314708 changes all puppet jobs to use the upgrade job, but I'm not sure it's actually providing any benefit in coverage given the current implementation of the job13:41
shardyderekh: I agree - I'm just saying we should probably do that before considering any change to project-config re the puppet module jobs13:42
derekhshardy: it probbaly would have caught the problem we've had for the last week but your correct in general it wont be testing N-1 -> N puppet module patches13:42
openstackgerritDan Prince proposed openstack/instack-undercloud: Load Mistral 'workbooks' (not 'workflows').  https://review.openstack.org/31504113:43
openstackgerritMerged openstack/tripleo-quickstart: Teardown network fix  https://review.openstack.org/31357513:44
shardyderekh: Yup, it's just going to test running puppet twice with the N module patch13:44
derekhshardy: ack13:45
*** absubram has quit IRC13:45
EmilienMshardy: ok so https://review.openstack.org/#/c/314708 is kind of useless now13:45
*** tosky has joined #tripleo13:46
shardyEmilienM: well, not useless, but I think it will provide more value if we improve the upgrades job, then make that change13:46
*** jobewan has joined #tripleo13:46
EmilienMshardy: I'm Workflow-1'ing the project-config patch now.13:46
*** jprovazn has quit IRC13:48
EmilienMslagle, shardy, derekh: ok so iiuc: current upgrade job runs deployment with THT master then THT from the patch but with same puppet modules before & after; so we would need to add support for upgrade job when running in puppet CI to run puppet modules from master first (with THT master) and then puppet patch (still THT master) -- is it correct?13:49
derekhEmilienM: yup, thats about it as i understand it also13:50
EmilienMok13:51
EmilienMso I found 1 weird thing in the scripts13:51
EmilienMhttps://github.com/openstack-infra/tripleo-ci/blob/master/scripts/deploy.sh#L118-L11913:51
derekhjust one?13:51
shardyEmilienM: Yes that's basically it13:51
shardylol :)13:51
EmilienMIMHO we should not install THT from packaging13:51
EmilienMbut rather use zuul-cloner13:51
EmilienMor use git13:51
EmilienMbecause there is a lag between what is in delorean/trunk and git/master13:52
shardyhttps://github.com/openstack-infra/tripleo-ci/blob/master/scripts/deploy.sh#L10213:52
EmilienMshardy: /me taking action to hack on the script so we can run the puppet CI with upgrade job13:52
shardyEmilienM: we build everything via delorean, including the patch under test, so it'd be a shame to not use packages to deploy the initial trunk environment13:53
EmilienMmhh ok13:53
shardyUnless rdo trunk is broken, how much lag are you referring to?13:53
EmilienMshardy: the unless :)13:53
shardyWell we depend on RDO trunk/current-tripleo for everything13:54
EmilienMok13:54
EmilienMI'll do the same for OPM then13:54
derekhEmilienM: unless their experiencing problems the lag shouldn't be excessive13:54
EmilienMderekh: we had problems in the past, but looks better now.13:54
derekhEmilienM: yup13:55
openstackgerritJay Dobies proposed openstack/tripleo-heat-templates: [WIP] Pass MysqlVirtualIP via EndpointMap  https://review.openstack.org/30575113:55
shardyEmilienM: I think we basically want to disable the delorean-ci for the duration of any image building and the initial deploy13:55
shardythen we need to expose that repo to the deployed nodes (perhaps by using DeployArtefacts to pass in a repo file), and run the minor-version package update process13:56
EmilienMshardy: does it make sense if I patch scripts/deploy.sh to add OPM package beside THT?13:56
pabelangerderekh: shardy: where can I find the build scripts for the instack.qcow2 image?13:56
shardyprobably, that should be a different option to tripleo.sh, as the current --overcloud-update is useful for just testing t-h-t updates locally13:56
EmilienMright13:57
EmilienMwe need something also for instack13:57
shardyEmilienM: I don't think we want to special case things like that - we should install *everything* from trunk, then enable delorean-ci for the upgrade13:57
derekhpabelanger: its being created by disk image builder here http://git.openstack.org/cgit/openstack-infra/tripleo-ci/tree/toci_instack.sh#n20713:57
openstackgerritCarlos Camacho proposed openstack/tripleo-heat-templates: Composable roles within services - MongoDB  https://review.openstack.org/31505113:57
EmilienMshardy: we don't upgrade instack at the moment, right?13:58
shardyEmilienM: we'll have to enable delorean-ci, update the undercloud, then do the overcloud upgrade13:58
shardyEmilienM: No we don't - that's why we've got the special-casing of the t-h-t package13:59
derekhpabelanger: the periodic job follows that path to build it and upload it to the cache nightly13:59
pabelangerderekh: great, thanks13:59
shardyEmilienM: Obviously we may end up needing some compromises due to the amount of time available, but ideally we'd fully update everything after enabling the delorean-ci repo13:59
EmilienMshardy: mhh, ok. Will look at this, /me needs to learn how it works and start hacking on it for puppet bits14:00
EmilienMshardy: well, testing THT + puppet is the least we need to do14:00
openstackgerritBob Callaway proposed openstack/tripleo-heat-templates: Enable Neutron LBaaS Integration  https://review.openstack.org/31393314:00
*** jpena|lunch is now known as jpena14:00
*** links has joined #tripleo14:01
*** ayoung has joined #tripleo14:01
*** masco has joined #tripleo14:03
openstackgerritPaul Belanger proposed openstack-infra/tripleo-ci: Build and upload instack.raw image  https://review.openstack.org/31507214:12
pabelangerderekh: ^14:12
pabelangerderekh: Also, looking at the code path, wouldn't you be uploading multiple instack images with each periodic job that is run?14:14
*** masco has quit IRC14:14
derekhpabelanger: yes, server side it only accepts the first one but look for a success status from multiple jobs for the same run before allowing the uploaded cache to be used14:15
*** jaosorior has quit IRC14:16
derekhpabelanger: I'm not sure the raw image will work, we may need to update the libvirt domain definitions from qcow, I'll upload a patch to take your code path to see if it works14:17
pabelangerderekh: ya, untested for sure.14:18
openstackgerritBrad P. Crochet proposed openstack/tripleo-heat-templates: composable heat services  https://review.openstack.org/31357714:20
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Look for a passing periodic upgrades job before moving pin  https://review.openstack.org/31507514:20
openstackgerritBrad P. Crochet proposed openstack/tripleo-heat-templates: composable heat services  https://review.openstack.org/31357714:23
openstackgerritMerged openstack/os-collect-config: Deprecated tox -downloadcache option removed  https://review.openstack.org/25680414:29
*** jprovazn has joined #tripleo14:30
*** ppradhan has quit IRC14:35
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker  https://review.openstack.org/29644014:35
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge.  https://review.openstack.org/31071314:35
*** ppradhan has joined #tripleo14:35
*** ppradhan has left #tripleo14:35
pabelangerderekh: any reason not to move the instack.qcow2 image to tarballs.o.o? Even extract the bits out of the script into a separate JJB job14:36
*** yamahata has joined #tripleo14:37
derekhpabelanger: moving them to would probably slow down the download process although we would be downloading via our squid so that might be ok14:41
*** [1]cdearborn has joined #tripleo14:41
derekhpabelanger: we'd also need logic to tripleo-ci somewhere to know when to move to a new cached image14:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Move parts of undercloud-install to puppet  https://review.openstack.org/31293114:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Use tripleo::selinux  https://review.openstack.org/31056514:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Move 02-undercloud-stack-heat-metadata to python module  https://review.openstack.org/31421914:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Move persistent journal to puppet  https://review.openstack.org/31056414:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Don't use ironic-conductor element  https://review.openstack.org/31293214:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Don't use os-cloud-config/common-venv elements  https://review.openstack.org/30855014:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Create firewall rules with puppet  https://review.openstack.org/30837414:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Don't use os-net-config element  https://review.openstack.org/31509514:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Consolidate undercloud* elements  https://review.openstack.org/31509614:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Remove tmp for service map  https://review.openstack.org/31509714:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Remove truncate nova config workaround  https://review.openstack.org/31509814:42
openstackgerritJames Slagle proposed openstack/instack-undercloud: Minimize element usage  https://review.openstack.org/31509914:42
derekhpabelanger: thats currently happening on the mirror server, it update current-tripleo when it see all the periodic ci jobs pass14:43
pabelangerderekh: AFS is another option, once we stand up an AFS mirror in tripleo-ci, it should work the same way14:43
*** jaosorior has joined #tripleo14:43
openstackgerritSteven Hardy proposed openstack/tripleo-image-elements: Don't send duplicate signals via 99-refresh-completed  https://review.openstack.org/31422614:44
derekhpabelanger: seperating out the build into its own job might be more tricky, we wouldn't then be testing it with the tripleo ci jobs so we wouldn't know if it was ok14:44
[1]cdearbornMorning foks!  I would like to specify a domain name for my overcloud nodes to use.  I tried adding 'CloudDomain: mydomain' to an environment file then deleting and redeploying the heat stack, but that had no effect.  Any ideas?14:44
shardyjaosorior: ^^ You fixed that didn't you?14:46
*** derekh changes topic to "================== TripleO CI cloud will be down for maintanance soon ===================="14:46
* shardy crosses fingers ;)14:46
jaosoriorshardy: Yeah, that should be possible now14:46
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge.  https://review.openstack.org/31071314:47
derekhshardy: just waiting on the ppl in the data center14:47
[1]cdearbornjaosorior: could you send me a pointer to the patch so I can verify I have it in the code I'm using?14:47
pabelangerderekh: sure, I'm just trying to take some burden off the tripleo team for CI stuff.14:47
*** eggmaster has joined #tripleo14:50
derekhpabelanger: yup, the more burden we can offload the better, the AFS server is probably worth exploring if we can make it work14:51
dprincepabelanger: ++ for the mirror14:53
*** links has quit IRC14:54
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge.  https://review.openstack.org/31071314:55
jaosoriorcdearborn: this hash is the commit 6d8eb3540675d6d378dd1b333553c8844898e711 that changes the structure (or the latest one)14:56
jaosorior[1]cdearborn: and this 9b43153f9822a38b4f6e29febd14d894e91425b2 introduced the parameter itself14:57
jaosoriorI gotta go though :/ was cooking14:57
pabelangerdprince: derekh: I should note, http://mirror.dfw.rax.openstack.org/centos/7/ is a thing now, so once we launch AFS in triple, we'll get a local centos-7 mirror now.14:57
pabelangernext step will be epel14:57
*** jaosorior has quit IRC14:57
pabelangerand potentially RDO, if people want it14:57
[1]cdearbornjaosorior: thx!  I'm very new to tripleo - can you tell me what repo that's in?14:58
derekhpabelanger: cool14:58
derekhpabelanger: how much space does the AFS server need?14:59
pabelangerderekh: right now, about 100GB. obviously more the better14:59
dprincepabelanger: would be nice to consider a pruned AFS, if that is even possible15:00
*** electrofelix has quit IRC15:01
pabelangerdprince: we'll, the mirror will grow as usage increases.  So, if nothing in tripleo-ci needs http://mirror.dfw.rax.openstack.org/ubuntu/ the cache won't warm up15:02
pabelangerbut, I can see triple-o using pypi and wheel15:02
*** absubram has joined #tripleo15:03
bnemec[1]cdearborn: Try https://review.openstack.org/#/c/292799/ and https://review.openstack.org/#/c/239955/15:03
pabelanger/dev/mapper/main-afscache   99G  5.4G   93G   6% /var/cache/openafs15:03
pabelangerdprince: that is the current usage on mirror.dfw.rax.o.o ^15:03
pabelanger/dev/mapper/main-afscache           197G  4.0G  193G   3% /var/cache/openafs15:04
pabelangeron mirror.regionone.osic-cloud1.openstack.org15:04
dprincepabelanger: an in rack mirror is obviously the best. But for most of our outages simply having a working mirror would be great for us too15:05
dprincepabelanger: what I'm saying is... if we simply used a mirror with better uptime than say the RDO mirror we'd be much better off15:06
*** derekh changes topic to "================== TripleO CI cloud is down for maintanance ===================="15:06
*** electrofelix has joined #tripleo15:06
[1]cdearbornbnemec: thx very much!15:06
dprincepabelanger: regardless of it being in-rack or not15:06
pabelangerdprince: agreed, that's mostly why we mirror to AFS right now. To provide better update / faster downloads15:06
pabelangerso far, it has been working well15:06
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci: Test stack-update to new OPM  https://review.openstack.org/31511715:11
EmilienMslagle: can you review https://review.openstack.org/315117 and tell me your feedback please? my goal is to run upgrade jobs in puppet CI15:11
EmilienMshardy: too ^15:12
openstackgerritMerged openstack/tripleo-ui: Add hints to run validations  https://review.openstack.org/31247615:13
*** jistr has quit IRC15:13
*** r-mibu has quit IRC15:14
shardyEmilienM: that just forces opm versions on the undercloud, we need to update them on the deployed overcloud nodes?15:14
*** r-mibu has joined #tripleo15:14
*** ebarrera has quit IRC15:14
EmilienMshardy: oh right15:15
shardyhttps://github.com/openstack-infra/tripleo-ci/blob/master/toci_instack.sh#L11315:16
shardywe need to ensure that repo is accessible to the overcloud nodes, then create a repo file which is deployed e.g via upload-swift-artifacts to the nodes at the start of the upgrade15:17
shardythen we can start of with a completely trunk deployment, and do an upgrade to what's in the CI generated delorean repo15:17
shardychanges to tripleo.sh will be required (new function) so we drive an update via the package update workflow:15:18
shardyhttp://docs.openstack.org/developer/tripleo-docs/post_deployment/package_update.html15:18
EmilienMshardy: we don't want to run yum update I think, just updating OPM15:19
shardyEmilienM: well I guess you could use upload-swift-artifacts to sync the puppet modules from undercloud->overcloud instead15:20
shardythat's not really testing a real user update/upgrade tho15:20
EmilienMright15:20
shardyit might be a simpler first step tho I guess15:21
*** milan has quit IRC15:21
shardyhttps://review.openstack.org/#/c/299643/ provides an upload-puppet-modules which may help15:21
*** dmacpher is now known as dmacpher-afk15:23
*** ebarrera has joined #tripleo15:29
*** jistr has joined #tripleo15:31
*** paramite has quit IRC15:32
*** jistr has quit IRC15:33
*** trozet has joined #tripleo15:33
*** jistr has joined #tripleo15:33
*** ifarkas has quit IRC15:35
openstackgerritJames Slagle proposed openstack/instack-undercloud: Add timestamps to puppet apply output  https://review.openstack.org/31512815:35
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable neutron server  https://review.openstack.org/31512915:37
*** mikelk has quit IRC15:37
openstackgerritMerged openstack/tripleo-puppet-elements: bump puppet-ceph to stable/hammer  https://review.openstack.org/31431115:39
*** olap has quit IRC15:41
*** panda has quit IRC15:42
*** panda has joined #tripleo15:42
*** dtrainor_ has quit IRC15:43
*** olap has joined #tripleo15:43
*** dtrainor has joined #tripleo15:44
*** ramishra has quit IRC15:44
*** ramishra has joined #tripleo15:46
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge.  https://review.openstack.org/31071315:46
*** rcernin has quit IRC15:47
EmilienMlarsks: I have a stupid question15:48
*** sshnaidm has quit IRC15:49
openstackgerritJiri Tomasek proposed openstack/tripleo-ui: Tracking nodes operations in Nodes Table  https://review.openstack.org/31423315:49
openstackgerritJiri Tomasek proposed openstack/tripleo-ui: Mistral Nodes Introspection  https://review.openstack.org/31402315:49
openstackgerritJiri Tomasek proposed openstack/tripleo-ui: Refactor Plans actions and reducer  https://review.openstack.org/31200715:49
openstackgerritJiri Tomasek proposed openstack/tripleo-ui: Nodes Registration  https://review.openstack.org/31513615:49
EmilienMlarsks: to deploy tripleo/master using quickstart, I need to download undercloud/overcloud from https://ci.centos.org/artifacts/rdo/images/master/delorean/stable/ right?15:49
*** sshnaidm has joined #tripleo15:49
larsksYou need to download an underlcoud image, yes, but the quickstart should take care of that for you.15:49
EmilienMlarsks: but how can I tell quickstart to use master and not mitaka?15:50
openstackgerritJiri Tomasek proposed openstack/tripleo-ui: Add missing key, remove redundant div in DeploymentPlan  https://review.openstack.org/31513715:50
EmilienMin RELEASE?15:50
larsksEmilienM: you can provide an explicit image_url to the master image.15:51
larsksE.g., see https://ci.centos.org/job/tripleo-quickstart-promote-master-delorean-minimal/configure15:51
EmilienMlarsks: I don't have credentials15:51
larsksAh, okay, just a sec.15:51
larskshttps://gist.github.com/larsks/f3733820bac3ff4739b9baf65691b62115:51
ccamachojokke_ If help needed ping me about creating a upstream dev env...15:52
*** rbrady has quit IRC15:52
*** rbrady has joined #tripleo15:52
EmilienMlarsks: thx!15:52
openstackgerritMerged openstack/tripleo-specs: Refactor top level puppet manifests  https://review.openstack.org/28643915:53
*** tremble has quit IRC15:53
larsksEmilienM: the RELEASE value there (the parameter after $VIRTHOST) is basically a placeholder; it just can't be "mitaka" or "liberty".15:53
EmilienMok15:54
EmilienMlarsks: I'm going to try:15:54
EmilienMash quickstart.sh --undercloud-image-url "http://artifacts.ci.centos.org/artifacts/rdo/images/master/delorean/testing/undercloud.qcow2" $VIRTHOST master15:54
*** dtantsur is now known as dtantsur|afk15:56
*** dmacpher-afk is now known as dmacpher15:57
*** devvesa has quit IRC15:58
jokke_ccamacho: cheers!15:59
pradkjokke_, http://etherpad.corp.redhat.com/profile-matching are the quick start instructions16:05
pradkjokke_, only thing you might need to do it tweak the profile match.yml to match your needs16:06
*** dmacpher is now known as dmacpher-afk16:07
*** dmk0202 has quit IRC16:07
jokke_pradk: yeah16:07
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Deploy RabbitMQ as a composable role  https://review.openstack.org/31186316:09
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates: Deploy Loadbalancer as a composable role  https://review.openstack.org/31302916:09
ccamachojokke_ Ill get you up and running faster, ill give some purifying fire to my CI and Ill show you how to do it automagically, reach you in five mins16:09
jokke_ccamacho: <3 :)16:09
*** ebarrera has quit IRC16:11
*** mcornea has quit IRC16:14
*** athomas has quit IRC16:15
ccamachojokke_ was setting up my bluejeans stuff, come https://redhat.bluejeans.com/7889947517/?src=meet_now16:15
rascahi everybody, my overcloud deploy is failing with this messages: May 11 15:54:55 overcloud-controller-0 os-collect-config[9276]: Error: Duplicate declaration: Heat_config[DEFAULT/num_engine_workers] is already declared; cannot redeclare at /etc/puppet/modules/heat/manifests/engine.pp:168 on node overcloud-controller-0.localdomain16:16
rascaMay 11 15:54:55 overcloud-controller-0 os-collect-config[9276]: Error: Duplicate declaration: Heat_config[DEFAULT/num_engine_workers] is already declared; cannot redeclare at /etc/puppet/modules/heat/manifests/engine.pp:168 on node overcloud-controller-0.localdomain16:16
rascado you have an idea of what I could check?16:16
*** ebarrera has joined #tripleo16:17
zoli see you tomorrow16:17
*** zoli is now known as zoli_gone-proxy16:18
shardyrasca: https://github.com/openstack-infra/tripleo-ci/commit/715d8a664588a65833f51b7dbcace66de0efb52316:18
EmilienMI was about pointing this too16:18
shardyrasca: there was a change to the puppet-heat module, so now you no longer have to specify num_engine_workers16:18
shardyjust specify HeatWorkers and it sets the API and engine workers16:19
rascashardy, many thanks!16:20
openstackgerritAthlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge.  https://review.openstack.org/31071316:21
*** athomas has joined #tripleo16:22
EmilienMpradk, shardy: why do we have ceilometer on undercloud? what is the use case?16:23
EmilienM(same question for aodh)16:23
shardyEmilienM: there is none, I posted a patch to remove them16:28
shardyhttps://github.com/openstack/instack-undercloud/commit/f6ce77547831a015bb7f6404c040957028022edd16:28
shardysince that landed we shouldn't enable them at all by default16:28
*** tosky has quit IRC16:36
*** tosky has joined #tripleo16:36
openstackgerritBrad P. Crochet proposed openstack/tripleo-heat-templates: composable heat services  https://review.openstack.org/31357716:37
*** myoung is now known as myoung|lunch16:39
*** thrash is now known as thrash|biab16:41
*** Guest42459 has quit IRC16:42
openstackgerritGiulio Fidente proposed openstack/puppet-tripleo: Add Cinder API/Scheduler/Volume roles  https://review.openstack.org/31093916:44
gfidentedprince ^^ give an eye to the cinder roles pls :)16:44
pradkEmilienM, we dont have ceilometer any more by default.. i thingk cloud forms were using them for something but i guess they'll enable if they want16:44
gfidenteI'll be updating the tht submission shortly so we test them16:44
*** ebarrera has quit IRC16:45
*** gfidente has quit IRC16:53
*** derekh has quit IRC16:55
*** jobewan is now known as jobewan_away16:58
*** ohamada has quit IRC17:00
*** jistr has quit IRC17:02
*** pkovar has quit IRC17:07
*** thrash|biab is now known as thrash17:08
*** jpena is now known as jpena|off17:10
*** aufi has quit IRC17:15
openstackgerritMerged openstack/instack: Correct example url in README  https://review.openstack.org/31323517:24
*** electrofelix has quit IRC17:25
openstackgerritOpenStack Proposal Bot proposed openstack/os-collect-config: Updated from global requirements  https://review.openstack.org/26444617:27
openstackgerritMerged openstack/tripleo-ui: Refactor Plans actions and reducer  https://review.openstack.org/31200717:27
*** lucasagomes is now known as lucas-dinner17:27
openstackgerritBrad P. Crochet proposed openstack/tripleo-heat-templates: composable heat services  https://review.openstack.org/31357717:30
*** mgould has quit IRC17:31
*** openstackgerrit has quit IRC17:33
*** openstackgerrit has joined #tripleo17:33
openstackgerritPradeep Kilambi proposed openstack/tripleo-heat-templates: Remove Ceilometer Alarm from the overcloud  https://review.openstack.org/28812017:34
openstackgerritPradeep Kilambi proposed openstack/tripleo-puppet-elements: Add Aodh component  https://review.openstack.org/28811917:34
*** zaneb has quit IRC17:35
*** milan has joined #tripleo17:36
*** Guest42459 has joined #tripleo17:44
*** tosky has quit IRC17:46
*** ccamacho is now known as ccamacho|away17:46
openstackgerritJames Slagle proposed openstack/instack-undercloud: Consolidate undercloud* elements  https://review.openstack.org/31509617:49
openstackgerritJames Slagle proposed openstack/instack-undercloud: Add timestamps to puppet apply output  https://review.openstack.org/31512817:49
openstackgerritJames Slagle proposed openstack/instack-undercloud: Minimize element usage  https://review.openstack.org/31509917:49
openstackgerritJames Slagle proposed openstack/instack-undercloud: Move 02-undercloud-stack-heat-metadata to python module  https://review.openstack.org/31421917:49
openstackgerritJames Slagle proposed openstack/instack-undercloud: Don't use os-net-config element  https://review.openstack.org/31509517:49
openstackgerritJames Slagle proposed openstack/instack-undercloud: Remove truncate nova config workaround  https://review.openstack.org/31518417:49
*** shivrao has joined #tripleo17:56
*** zaneb has joined #tripleo18:00
*** eil397 has joined #tripleo18:06
EmilienMpradk: right18:20
openstackgerritRyan Brady proposed openstack/tripleo-common: Initial Integration of Mistral into TripleO  https://review.openstack.org/28236618:23
*** chem has quit IRC18:28
*** krotscheck_ has joined #tripleo18:30
*** krotscheck has quit IRC18:31
dtrainornice.18:32
*** krotscheck_ is now known as krotscheck18:33
*** bfournie has joined #tripleo18:37
*** bfournie has quit IRC18:37
*** weshay has quit IRC18:41
EmilienMlarsks: how to get overcloud image from master when deploying with quickstart? I compared md5 from the one it downloaded and the one from master, and they are diff18:42
larsksIt should just download from the URL you gave it.  If it didn't work, you can download it manually, stick it on the virthost somewhere, and then set image_url to a file:// url.  And file a bug.18:42
*** zoli_gone-proxy is now known as zoliXXL18:43
larsksBut we're downloading from master automatically in ci, and it seems to be working.18:43
larsksOn the other hand, ci has much better access to the download links than I do.18:43
EmilienMso if I run: bash quickstart.sh --undercloud-image-url "http://artifacts.ci.centos.org/artifacts/rdo/images/master/delorean/testing/undercloud.qcow2" $VIRTHOST master18:43
EmilienMI should have overcloud image from master?18:44
larsksYes.18:50
*** tesseract has quit IRC18:51
openstackgerritBrad P. Crochet proposed openstack/tripleo-heat-templates: composable heat services  https://review.openstack.org/31357718:53
EmilienMok thanks18:54
*** fragatina has quit IRC19:01
*** sambetts is now known as sambetts|afk19:04
*** weshay has joined #tripleo19:10
openstackgerritJames Slagle proposed openstack/instack-undercloud: Consolidate undercloud* elements  https://review.openstack.org/31509619:17
openstackgerritJames Slagle proposed openstack/instack-undercloud: Add timestamps to puppet apply output  https://review.openstack.org/31512819:17
openstackgerritJames Slagle proposed openstack/instack-undercloud: Minimize element usage  https://review.openstack.org/31509919:17
openstackgerritJames Slagle proposed openstack/instack-undercloud: Move 02-undercloud-stack-heat-metadata to python module  https://review.openstack.org/31421919:17
openstackgerritJames Slagle proposed openstack/instack-undercloud: Remove truncate nova config workaround  https://review.openstack.org/31518419:17
openstackgerritJames Slagle proposed openstack/instack-undercloud: Don't use ironic-conductor element  https://review.openstack.org/31293219:17
openstackgerritJames Slagle proposed openstack/instack-undercloud: Don't use os-net-config element  https://review.openstack.org/31509519:17
openstackgerritTzu-Mainn Chen proposed openstack/tripleo-common: Add TripleO Mistral documentation  https://review.openstack.org/31310919:20
EmilienMslagle: do you remember why you added glance/host in ironic.conf in https://github.com/openstack/instack-undercloud/commit/0c9b39baf18095bf914e54daf510409efc4d30a7 ?19:26
*** fragatina has joined #tripleo19:26
slagleEmilienM: not really, maybe it's the old name for glance/glance_host?19:28
EmilienMmaybe19:29
EmilienMI'm doing cleanup in instack manifest19:29
EmilienMI'm doing ironic now19:29
slaglewhat type of cleanup?19:29
EmilienMso we can data driven and avoi dusing *_config providers19:29
EmilienMslagle: put all data in hieradata, and add missing params in puppet-* modules so we don't use _config providers19:29
EmilienMI also found some useless lookups19:30
slagleyou may like to see my patch series: https://review.openstack.org/#/q/topic:cull-elements+project:openstack/instack-undercloud19:30
slagleas i have a lot of changes in flight19:30
EmilienMslagle: yeah I saw it, and you did not do ironic afik19:30
slaglei know, was just mentioning it so we don't start causing each other rebase-hell19:31
EmilienMcool19:32
*** trozet has quit IRC19:32
*** jaosorior has joined #tripleo19:34
*** shardy has quit IRC19:36
*** bnemec has quit IRC19:36
*** dmk0202 has joined #tripleo19:38
*** trozet has joined #tripleo19:39
*** milan_ has joined #tripleo19:43
*** milan has quit IRC19:44
*** bnemec has joined #tripleo19:44
*** jprovazn has quit IRC19:44
*** jaosorior has quit IRC19:51
*** ccamacho|away has quit IRC19:58
*** ccamacho has joined #tripleo19:59
*** athomas has quit IRC19:59
openstackgerritRyan Brady proposed openstack/tripleo-common: Initial Integration of Mistral into TripleO  https://review.openstack.org/28236620:01
*** yamahata has quit IRC20:01
openstackgerritEmilien Macchi proposed openstack/instack-undercloud: Cleanup Puppet code that deploys Ironic  https://review.openstack.org/31523820:03
*** jayg is now known as jayg|g0n320:22
*** absubram has quit IRC20:38
*** ayoung has quit IRC20:41
*** pcaruana has quit IRC20:42
*** rcernin has joined #tripleo20:44
*** Guest42459 has quit IRC20:47
*** MaxPC has quit IRC20:47
*** yamahata has joined #tripleo20:50
bnemecGood grief.  I have an undercloud upgrade job that's been running for 1 hour 45 minutes.20:51
slagleon one of my patches? :)20:52
slaglei'm actually trying an upgrade now20:53
bnemecslagle: On all of your patches.  I have a job queue of 22 right now. :-)20:58
bnemecThey've all passed so far though.20:58
*** weshay has quit IRC20:59
*** florianf has quit IRC20:59
[1]cdearbornhey guys, tried out setting CloudDomain param, and setting it does cause the names in /etc/hosts on the overcloud nodes to be correct, but running "hostname" on the overcloud nodes still shows a domain of "localdomain".  Any idea how to override that?21:01
*** lblanchard has quit IRC21:06
*** bvandenh has quit IRC21:08
*** CaptTofu has quit IRC21:09
*** hrybacki has quit IRC21:09
*** fultonj has quit IRC21:10
bnemecTook 2 hr 4 min21:10
bnemecRidiculous.21:10
*** CaptTofu has joined #tripleo21:10
*** hrybacki has joined #tripleo21:10
bnemec[1]cdearborn: To change that you have to set the domain on the undercloud neutron too.21:10
bnemec[1]cdearborn: Note https://github.com/openstack/tripleo-heat-templates/blob/0d3dacec4b63fb168f4f039501dc6e7b9622315b/overcloud.yaml#L34621:11
[1]cdearbornbnemec: hrm - i tried setting dhcp_domain in /etc/neutron/dhcp_agent.ini and restarted the neutron services on the undercloud, but no dice21:11
EmilienMGoneri: you missed the backport to stable/mitaka ;-)21:11
GoneriEmilienM, I did the two in a row :D21:12
bnemec[1]cdearborn: I think it's in neutron.conf now:21:12
bnemec# Domain to use for building the hostnames. This option is deprecated. It has21:12
bnemec# been moved to neutron.conf as dns_domain.21:12
EmilienMGoneri: +2 on both21:13
[1]cdearbornbnemec: ok - thx - will give that a shot.  I saw that note, but figured the note regarding dhcp_domain was correct21:13
*** ayoung has joined #tripleo21:15
slaglei guess i didnt realize we were allowing  breaking the parameter interfaces when doing composable services21:19
slaglebut i see several parameters coming out of overcloud.yaml and going into the service specific nested stacks21:19
*** mburned is now known as mburned_out21:19
slagledprince: i guess that's ok? ^21:20
slagleif people are using parameters:, they will have to migrate to parameter_defaults:21:20
*** rhallisey has quit IRC21:24
openstackgerritEmilien Macchi proposed openstack/instack-undercloud: Cleanup Puppet code that deploys Ironic  https://review.openstack.org/31523821:37
EmilienMslagle: I know youre busy, but this is the kind of cleanup I'm doing : https://review.openstack.org/#/c/315238/ (and I want to do it for all services) -- please let me know feedback21:39
slagleEmilienM: yep, i like things heading in that direction, will have a closer look in a bit21:42
EmilienMcool21:43
*** julim has quit IRC21:47
*** pradk has quit IRC21:47
openstackgerritAndreas Florath proposed openstack/diskimage-builder: Fix apt-sources configuration for debian-minimal  https://review.openstack.org/31526521:47
*** Goneri has quit IRC21:48
*** [1]cdearborn has quit IRC21:50
dprinceslagle: we aren't breaking the parameters interface because python_tripleoclient now uses parameter_defaults for everything22:26
dprinceslagle: so it was generally agreed that "internal" refactoring by moving top level overcloud.yaml paramters into nested stacks was safe so long as the names stay the same22:27
*** fpan has quit IRC22:28
*** dprince has quit IRC22:29
*** Goneri has joined #tripleo22:29
*** morazi has quit IRC22:34
*** mbound has joined #tripleo22:41
*** apetrich has quit IRC22:46
*** dmk0202 has quit IRC22:47
*** rooooook has quit IRC22:55
*** zoliXXL is now known as zoli_gone-proxy22:57
*** saneax_AFK is now known as saneax22:57
*** myoung|lunch has quit IRC23:03
*** trozet has quit IRC23:08
*** sambetts|afk has quit IRC23:13
*** [1]cdearborn has joined #tripleo23:13
*** zigo has quit IRC23:14
*** pblaho has quit IRC23:14
*** pblaho has joined #tripleo23:15
*** zigo has joined #tripleo23:15
*** sambetts has joined #tripleo23:15
*** trozet has joined #tripleo23:19
*** [1]cdearborn has quit IRC23:31
*** yamahata has quit IRC23:34
*** panda has quit IRC23:42
*** panda has joined #tripleo23:42
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Update testenvs to use SSD's  https://review.openstack.org/31529023:53
*** eil397 has quit IRC23:59

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