| *** akshai has joined #tripleo | 00:14 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 00:18 |
|---|---|---|
| *** milan has joined #tripleo | 00:19 | |
| *** dmacpher has quit IRC | 00:23 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 00:28 |
| *** david-lyle has quit IRC | 00:31 | |
| *** jcoufal has joined #tripleo | 00:42 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 00:44 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Add mulitnode CI job support to tripleo-ci https://review.openstack.org/324777 | 00:51 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Add mulitnode CI job support to tripleo-ci https://review.openstack.org/324777 | 00:52 |
| *** yamahata has quit IRC | 00:55 | |
| *** jdob has quit IRC | 01:00 | |
| *** akshai has quit IRC | 01:03 | |
| *** dmacpher has joined #tripleo | 01:06 | |
| *** ebalduf has quit IRC | 01:08 | |
| *** jdob has joined #tripleo | 01:09 | |
| *** jdob has quit IRC | 01:16 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 01:26 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 01:36 |
| *** jdob has joined #tripleo | 01:39 | |
| *** egafford has quit IRC | 01:51 | |
| *** akshai has joined #tripleo | 01:55 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 02:01 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 02:07 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 02:14 |
| *** akshai has quit IRC | 02:30 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 02:37 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: [WIP] workaround 'Requested operation is not valid: cannot set autostart for transient network' https://review.openstack.org/341870 | 02:46 |
| *** lblanchard has quit IRC | 03:07 | |
| *** coolsvap_ is now known as coolsvap | 03:10 | |
| *** rajinir has quit IRC | 03:15 | |
| *** akuznetsov has joined #tripleo | 03:16 | |
| *** yamahata has joined #tripleo | 03:48 | |
| *** shivrao has quit IRC | 03:52 | |
| *** julim has quit IRC | 03:56 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Revert "Revert "Pre-install pip/virtualenv packages"" https://review.openstack.org/341934 | 03:56 |
| *** julim has joined #tripleo | 03:57 | |
| *** julim has quit IRC | 03:57 | |
| *** dmacpher has quit IRC | 03:59 | |
| *** ebalduf has joined #tripleo | 04:03 | |
| *** dmacpher has joined #tripleo | 04:07 | |
| *** akuznetsov has quit IRC | 04:10 | |
| *** eggmaster has quit IRC | 04:16 | |
| *** ramishra has quit IRC | 04:19 | |
| *** ramishra has joined #tripleo | 04:21 | |
| *** anshul has joined #tripleo | 04:24 | |
| *** links has joined #tripleo | 04:27 | |
| *** coolsvap_ has joined #tripleo | 04:34 | |
| *** shivrao has joined #tripleo | 04:36 | |
| *** jcoufal has quit IRC | 04:37 | |
| *** coolsvap has quit IRC | 04:37 | |
| *** numan_ has joined #tripleo | 04:39 | |
| *** eggmaster has joined #tripleo | 05:01 | |
| *** HenryG has quit IRC | 05:12 | |
| *** coolsvap_ is now known as coolsvap | 05:12 | |
| *** HenryG has joined #tripleo | 05:15 | |
| *** jcoufal has joined #tripleo | 05:23 | |
| openstackgerrit | Merged openstack/diskimage-builder: Make Fedora 24 the default https://review.openstack.org/339681 | 05:26 |
| *** pgadiya has joined #tripleo | 05:27 | |
| *** radek_ has joined #tripleo | 05:38 | |
| *** bswartz has quit IRC | 05:41 | |
| *** radek__ has joined #tripleo | 05:43 | |
| *** radek__ has quit IRC | 05:43 | |
| *** radek__ has joined #tripleo | 05:44 | |
| *** ccamacho has quit IRC | 05:46 | |
| *** Goneri has joined #tripleo | 05:54 | |
| *** rcernin has joined #tripleo | 06:01 | |
| *** ramishra_ has joined #tripleo | 06:06 | |
| *** ramishra has quit IRC | 06:07 | |
| *** dmacpher has quit IRC | 06:07 | |
| *** fragatina has joined #tripleo | 06:12 | |
| *** fragatina has quit IRC | 06:12 | |
| *** fragatina has joined #tripleo | 06:13 | |
| *** oshvartz has joined #tripleo | 06:13 | |
| *** fragatina has quit IRC | 06:18 | |
| *** jcoufal has quit IRC | 06:26 | |
| *** Goneri has quit IRC | 06:28 | |
| *** fragatina has joined #tripleo | 06:33 | |
| *** fragatina has quit IRC | 06:34 | |
| *** fragatina has joined #tripleo | 06:35 | |
| *** Goneri has joined #tripleo | 06:38 | |
| *** liverpooler has joined #tripleo | 06:38 | |
| *** pcaruana has joined #tripleo | 06:40 | |
| openstackgerrit | sandhya proposed openstack/diskimage-builder: Add support for building images capable of UEFI https://review.openstack.org/287784 | 06:42 |
| *** anshul has quit IRC | 06:46 | |
| *** limao has joined #tripleo | 06:48 | |
| *** saneax_AFK is now known as saneax | 06:52 | |
| *** saneax is now known as saneax_AFK | 06:53 | |
| *** saneax_AFK is now known as saneax | 06:53 | |
| saneax | GM folks | 06:53 |
| *** jubapa has joined #tripleo | 06:57 | |
| *** trown|outtypewww has quit IRC | 06:57 | |
| *** anshul has joined #tripleo | 06:58 | |
| *** anshul has quit IRC | 06:58 | |
| *** trown has joined #tripleo | 06:59 | |
| *** anshul has joined #tripleo | 06:59 | |
| *** rcernin has quit IRC | 07:05 | |
| *** athomas has joined #tripleo | 07:06 | |
| *** dsariel has joined #tripleo | 07:08 | |
| *** jaosorior has joined #tripleo | 07:08 | |
| *** tesseract- has joined #tripleo | 07:10 | |
| *** mcornea has joined #tripleo | 07:10 | |
| *** ccamacho has joined #tripleo | 07:11 | |
| *** shivrao has quit IRC | 07:14 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable keystone to use the SSL middleware https://review.openstack.org/341354 | 07:16 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable glance to use the SSL middleware https://review.openstack.org/341367 | 07:16 |
| *** akuznetsov has joined #tripleo | 07:16 | |
| *** ebalduf has quit IRC | 07:19 | |
| *** shardy has joined #tripleo | 07:21 | |
| *** rcernin has joined #tripleo | 07:21 | |
| *** sshnaidm|afk is now known as sshnaidm | 07:24 | |
| *** jubapa has quit IRC | 07:25 | |
| *** jubapa_ has joined #tripleo | 07:25 | |
| *** aufi has joined #tripleo | 07:25 | |
| *** zoli_gone-proxy is now known as zoliXXL | 07:27 | |
| *** ohamada has joined #tripleo | 07:28 | |
| *** ebarrera has joined #tripleo | 07:28 | |
| coolsvap | shardy, hi | 07:30 |
| *** lazy_prince has quit IRC | 07:30 | |
| jaosorior | shardy: Hey dude, tried installing puppet-tripleo from a package, but it seems to me that it gets ignored by the undercloud install :/ | 07:30 |
| coolsvap | shardy, while doing the bug triaging in tripleo we do not have target milestones for stable branches | 07:31 |
| *** lazy_prince has joined #tripleo | 07:32 | |
| *** fragatina has quit IRC | 07:36 | |
| *** fragatina has joined #tripleo | 07:36 | |
| *** anshul has quit IRC | 07:38 | |
| shardy | coolsvap: No we don't currently have stable milestones, I can create some, but to date the stable releases have been somewhat ad-hoc | 07:43 |
| coolsvap | shardy, alright np | 07:44 |
| shardy | jaosorior: stick it in a local yum repo and set the priority up so it's higher than everything else? | 07:44 |
| coolsvap | I did not find it so wanted to confirm | 07:44 |
| shardy | jaosorior: I thought it'd be a higher version thus wouldn't be upgraded, but I've mostly been testing puppet-tripleo on the overcloud lately | 07:45 |
| shardy | coolsvap: If you'd find it helpful we could just create some, just let me know | 07:45 |
| coolsvap | shardy, i think it will be helpful for tracking | 07:46 |
| *** liverpooler has quit IRC | 07:51 | |
| *** abehl has joined #tripleo | 07:51 | |
| *** liverpooler has joined #tripleo | 07:51 | |
| *** fragatina has quit IRC | 07:52 | |
| jaosorior | shardy: What do we do in CI for puppet-tripleo? | 07:54 |
| shardy | jaosorior: we build the package and put it in a repo with higher priority than all the rest | 07:54 |
| jaosorior | I see | 07:55 |
| jaosorior | uh... how do I do that then? O_o | 07:55 |
| *** numan_ has quit IRC | 07:56 | |
| shardy | cd /etc/yum.repos.d/; grep priority ./* | 07:57 |
| shardy | you will see the delorean* repos are 10, 20, 30 | 07:57 |
| shardy | just create a delorean-local.repo with priority=1 | 07:57 |
| shardy | then do yum check-update to prove it's pulling puppet-tripleo from it | 07:57 |
| *** jrist has joined #tripleo | 07:59 | |
| *** d0ugal has joined #tripleo | 08:00 | |
| *** rcernin has quit IRC | 08:00 | |
| *** zoliXXL is now known as zoli|mtg | 08:01 | |
| *** jpich has joined #tripleo | 08:02 | |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: WIP: DONT MERGE TESTING https://review.openstack.org/316436 | 08:04 |
| *** akuznetsov has quit IRC | 08:04 | |
| jaosorior | shardy: Any specific name I should consider for naming the local repo? tried the check-update and it failed cause of the lack of name | 08:04 |
| *** saneax is now known as saneax_AFK | 08:05 | |
| jaosorior | shardy: Well... gave some name,. but not sure what baseurl to give | 08:07 |
| jaosorior | cause it seems to be failing on that now | 08:08 |
| shardy | file:///path/to/local/repo | 08:08 |
| shardy | or http://path/to/repo | 08:08 |
| shardy | https://wiki.centos.org/HowTos/CreateLocalRepos | 08:09 |
| *** jistr is now known as jistr|off | 08:11 | |
| *** florianf has joined #tripleo | 08:12 | |
| *** tremble has joined #tripleo | 08:12 | |
| *** derekh has joined #tripleo | 08:12 | |
| *** abehl has quit IRC | 08:12 | |
| *** rcernin has joined #tripleo | 08:14 | |
| *** dtrainor has joined #tripleo | 08:14 | |
| openstackgerrit | Marios Andreou proposed openstack/puppet-tripleo: Ensure the Heat Domain parameters before starting heat-engine pcmk service https://review.openstack.org/342032 | 08:15 |
| *** athomas has quit IRC | 08:16 | |
| *** hewbrocca-afk is now known as hewbrocca | 08:17 | |
| *** athomas has joined #tripleo | 08:19 | |
| *** ifarkas has joined #tripleo | 08:19 | |
| hewbrocca | Morning folks | 08:19 |
| *** akuznetsov has joined #tripleo | 08:19 | |
| *** dbecker has joined #tripleo | 08:22 | |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: Remove the unused service_host arg from node registration https://review.openstack.org/326036 | 08:23 |
| *** abehl has joined #tripleo | 08:25 | |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: Remove the underscore from the base action client methods https://review.openstack.org/335920 | 08:26 |
| *** shardy has quit IRC | 08:29 | |
| *** lucas|afk is now known as lucasagomes | 08:29 | |
| *** leanderthal|afk is now known as leanderthal | 08:30 | |
| *** jubapa_ has quit IRC | 08:39 | |
| *** abehl has quit IRC | 08:40 | |
| *** abehl has joined #tripleo | 08:41 | |
| *** anshul_ has joined #tripleo | 08:52 | |
| *** abehl has quit IRC | 08:54 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud: Use certmonger for automatic cert generation https://review.openstack.org/332774 | 08:55 |
| d0ugal | Is there a way to filter in gerrit for gate-tripleo-ci-centos-7-ovb-ha filters? or is that impossible because there is no voting? | 08:55 |
| *** milan has quit IRC | 08:58 | |
| *** jtomasek has joined #tripleo | 09:00 | |
| *** zoli|mtg is now known as zoli | 09:01 | |
| *** zoli is now known as zoliXXL | 09:01 | |
| *** saneax_AFK is now known as saneax | 09:05 | |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: Adds Plan List Action https://review.openstack.org/300221 | 09:05 |
| *** shardy has joined #tripleo | 09:08 | |
| jaosorior | shardy, marios: Would you guys be opposed to the undercloud openstack services always being behind HAProxy? I'm working on TLS everywhere, and currently haproxy is only used in the undercloud if we enable TLS for the public endpoints in the undercloud. However, I think that's a quite comon usecase and it wouldn't really hurt to just have it there always to make the manifests simpler | 09:08 |
| jaosorior | so, for TLs everywhere, then I wouldn't need to change much, as we would already be accessing the openstack services through HAProxy, since currently that's only the case for the public endpoints | 09:09 |
| marios | jaosorior: i'd say the worst you can do is ut a review out to point at | 09:09 |
| marios | jaosorior: i mean, i don't immediately have any thoughts on why doint that would be a bad idea, but perhaps others will | 09:10 |
| jaosorior | alright | 09:10 |
| jaosorior | will do | 09:10 |
| shardy | jaosorior: Hmm, that's a question for the ML I think - I guess it would be OK and may make things easier should we allow multiple undercloud nodes in future | 09:10 |
| shardy | bit of a weird pattern when there's only one node tho | 09:10 |
| shardy | (or just push a review as marios said) | 09:11 |
| jaosorior | I'll go for the review then | 09:11 |
| jaosorior | thanks guys | 09:11 |
| *** sambetts|afk is now known as sambetts | 09:13 | |
| hewbrocca | we're gonna want multiple undercloud nodes at some point relatively soon | 09:13 |
| *** akuznetsov has quit IRC | 09:13 | |
| *** mgould|afk is now known as mgould | 09:14 | |
| shardy | The question is do we deploy an "overcloud" as the multi-node undercloud with ironic enabled tho I think | 09:14 |
| shardy | vs the current undercloud implementation that jaosorior proposes to modify | 09:14 |
| *** akuznetsov has joined #tripleo | 09:14 | |
| shardy | e.g the old seed->undercloud->overcloud model | 09:14 |
| hewbrocca | mmm yeah the first thing would make more sense | 09:15 |
| hewbrocca | since we already know how to do that... | 09:15 |
| *** fragatina has joined #tripleo | 09:15 | |
| shardy | I guess the alternative is to orchestrate multiple undercloud nodes with some other tool, but that undercloud/overcloud symmetry and reuse is something TripeO has always aimed for | 09:15 |
| hewbrocca | Yeah I have no idea why we would do that | 09:16 |
| hewbrocca | Bring on the seed cloud! | 09:16 |
| hewbrocca | :D | 09:16 |
| shardy | clouds all the way down :) | 09:16 |
| *** akuznets_ has joined #tripleo | 09:17 | |
| *** akuznetsov has quit IRC | 09:17 | |
| jaosorior | is it gonna end up being septupleO? | 09:17 |
| hewbrocca | at least | 09:17 |
| jaosorior | O_O | 09:18 |
| hewbrocca | :D | 09:18 |
| shardy | jaosorior: hehe, yup :) | 09:18 |
| *** tosky has joined #tripleo | 09:18 | |
| *** limao has quit IRC | 09:18 | |
| shardy | we might do something like run the seed as an appliance in a container on one of the undercloud nodes | 09:19 |
| shardy | to remove the need for additional nodes | 09:19 |
| hewbrocca | Yeah | 09:19 |
| hewbrocca | funnily enough it might actually make deployment easier | 09:19 |
| hewbrocca | "here, just spin up this container" | 09:19 |
| jaosorior | so, is the vision to have the undercloud be deployed in the same way as the overcloud? | 09:19 |
| jaosorior | so the undercloud will now be the seed cloud node? | 09:20 |
| shardy | jaosorior: Yes, that's one way we could do it | 09:20 |
| shardy | TripleO used to work like that before we switched to instack-undercloud | 09:20 |
| jaosorior | why was it changed? | 09:20 |
| shardy | jaosorior: because we wanted to configure the undercloud with the same puppet modules used for the overcloud, and remove the need for an additional seed node | 09:21 |
| shardy | previously the seed/undercloud were configured with some scary shell scripts and os-apply-config | 09:21 |
| jaosorior | I see | 09:22 |
| shardy | The current implementation is cleaner and easier to use, but we lost the ability to do multi-node underclouds in the process | 09:22 |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Fixes for dlrn gate repo injection while using devmode https://review.openstack.org/334691 | 09:22 |
| * shardy remembers he needs to retire the tripleo-incubator repo which still contains most of those scripts | 09:23 | |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Add third party gating to quickstart https://review.openstack.org/332223 | 09:23 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-incubator: WIP retire tripleo-incubator https://review.openstack.org/342065 | 09:27 |
| shardy | derekh: Hey, if you get a moment, perhaps you can comment on ^^ - I'm aware there are some things there which will need to move to tripleo-ci, but you'll probably be able to list them whereas I'll have to iterate on CI failures to figure them out :) | 09:28 |
| *** saneax is now known as saneax_AFK | 09:28 | |
| derekh | shardy: will do | 09:29 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-incubator: WIP retire tripleo-incubator https://review.openstack.org/342065 | 09:29 |
| jaosorior | Hey, can anyone give this a check https://review.openstack.org/#/c/341390/ ? It's the base certmonger puppet manifest that just installs certmonger and sets some defaults | 09:31 |
| jaosorior | just needs a +A | 09:31 |
| *** masco is now known as masco|afk | 09:33 | |
| shardy | jaosorior: shouldn't that be in a profile like all the composable services profiles? | 09:33 |
| *** sambetts has quit IRC | 09:36 | |
| derekh | shardy: I remember taking a look in the passed, iirc it was only scripts/wait_for that we used in CI but I'll give it a closer look | 09:37 |
| *** mbound has joined #tripleo | 09:38 | |
| *** saneax_AFK is now known as saneax | 09:38 | |
| derekh | shardy: btw, this got past the nonha problem https://review.openstack.org/#/c/340717/ | 09:38 |
| derekh | shardy: introspection failed | 09:39 |
| *** milan has joined #tripleo | 09:40 | |
| jaosorior | shardy: well... it could I guess. Though there is no such thing as HA certmonger, and certmonger isn't really listening on a port, nor it's a REST service... should I move it then? | 09:40 |
| jaosorior | it's a daemon that handles the local certificates, so I saw it as something such as the firewall or even selinux. It's a local service that should be per-node | 09:41 |
| *** mbound has quit IRC | 09:43 | |
| *** radek2 has joined #tripleo | 09:46 | |
| shardy | jaosorior: stuff like kernel configuration, keepalived and memcached are going in manifests/profile/base so I think perhaps it should go there? | 09:49 |
| shardy | derekh: ack, thanks, good news the ceph pin may now be working :) | 09:50 |
| *** jaosorior has quit IRC | 09:50 | |
| *** jaosorior has joined #tripleo | 09:50 | |
| shardy | jaosorior: not a blocker, maybe lets chat with EmilienM about it when he comes online | 09:51 |
| jaosorior | will do | 09:51 |
| jaosorior | shardy: But... why are those going to the profiles? Actually those to me make no sense there | 09:51 |
| shardy | jaosorior: everything is moving into the profiles AIUI | 09:52 |
| *** akrivoka has joined #tripleo | 09:55 | |
| sshnaidm | derekh, FYI, https://bugs.launchpad.net/tripleo/+bug/1602976 | 09:55 |
| openstack | Launchpad bug 1602976 in tripleo "CI: Istrospection timeout in nonha jobs" [Critical,Triaged] - Assigned to Sagi (Sergey) Shnaidman (sshnaidm) | 09:55 |
| sshnaidm | derekh, I try to figure out now how to configure 3816 port in /etc/ironic-inspector/dnsmasq.conf with hiera, like here: https://review.openstack.org/#/c/340339/5/scripts/deploy.sh | 09:57 |
| *** mbound has joined #tripleo | 09:57 | |
| sshnaidm | derekh, if you know the key it would be great, trying now simple sed solution | 09:58 |
| *** bvandenh has joined #tripleo | 09:59 | |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Enable Manila integration - as a composable controller service https://review.openstack.org/188137 | 10:01 |
| openstackgerrit | Marios Andreou proposed openstack/python-tripleoclient: Adds ManilaPassword to the tripleoclient https://review.openstack.org/339437 | 10:11 |
| *** dtantsur|afk is now known as dtantsur | 10:12 | |
| marios | shardy: should we just land this https://review.openstack.org/#/c/339109/ | 10:13 |
| marios | jaosorior: ^^ pls | 10:13 |
| shardy | marios: sure | 10:14 |
| marios | shardy: ack will do in a bit if no-one beats me to it, i'll consider it sufficient time for objections :) | 10:15 |
| openstackgerrit | Saravanan KR proposed openstack/os-net-config: WIP: Add support for OVS DPDK Bridge and Port https://review.openstack.org/338694 | 10:15 |
| openstackgerrit | Merged openstack/tripleo-docs: Mark Manila integration as WIP https://review.openstack.org/339109 | 10:18 |
| *** akuznets_ has quit IRC | 10:18 | |
| marios | thanks jaosorior | 10:19 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Remove empty test file https://review.openstack.org/332671 | 10:20 |
| d0ugal | Anyone want some trivial reviews? https://review.openstack.org/#/q/topic:trivial+is:open+owner:%22Dougal+Matthews+%253Cdougal%2540redhat.com%253E%22 | 10:24 |
| d0ugal | jaosorior: Thanks! | 10:27 |
| openstackgerrit | Marios Andreou proposed openstack/python-tripleoclient: Adds ManilaPassword to the tripleoclient https://review.openstack.org/339437 | 10:28 |
| jaosorior | d0ugal no worries | 10:28 |
| saneax | folk need some help debugging failures in gate-tripleo-ci-centos-7-ovb-nonha | 10:32 |
| saneax | http://logs.openstack.org/96/341596/1/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha/70fa2ba/logs/postci.txt.gz#_2016-07-13_18_44_02_000 | 10:32 |
| shardy | saneax: that job has been broken by puppet-ceph, we're trying to work around it by https://review.openstack.org/#/c/340717/ | 10:33 |
| shardy | we're not ready for jewel yet so need to pin to stable/hammer | 10:33 |
| d0ugal | shardy: Thanks! | 10:34 |
| openstackgerrit | Merged openstack/tripleo-common: Remove execution_id from the workflows https://review.openstack.org/332003 | 10:34 |
| saneax | shardy, thanks | 10:34 |
| *** coolsvap_ has joined #tripleo | 10:34 | |
| derekh | sshnaidm: ok, here is the patch slagle originally user to make it configurable, there might be some clues in there | 10:35 |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Add third party gating to quickstart https://review.openstack.org/332223 | 10:37 |
| openstackgerrit | Merged openstack/python-tripleoclient: Remove unused exceptions and redundant pass statements https://review.openstack.org/332672 | 10:37 |
| openstackgerrit | Merged openstack/python-tripleoclient: Remove empty test file https://review.openstack.org/332671 | 10:37 |
| *** coolsvap has quit IRC | 10:38 | |
| openstackgerrit | Merged openstack/tripleo-common: Move Workflow comments into the workflow description property https://review.openstack.org/341590 | 10:39 |
| *** coolsvap_ is now known as coolsvap | 10:41 | |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: Remove "type: direct" from workflows as it is the default https://review.openstack.org/341617 | 10:48 |
| *** akuznetsov has joined #tripleo | 10:50 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud: Remove hardcoded endpoints in hiera https://review.openstack.org/342103 | 10:55 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud: Remove hardcoded endpoints in hiera https://review.openstack.org/342103 | 10:55 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Temporarily remove puppet-ceph from ci slaves https://review.openstack.org/340717 | 10:59 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Temporarily disable introspection https://review.openstack.org/342106 | 10:59 |
| derekh | shardy: it failed introspection again, sshnaidm is working on a fix, I think we should just merge ^^ in the mean time | 11:00 |
| sshnaidm | derekh, or revert prev patch.. | 11:00 |
| shardy | derekh: ack, will do | 11:00 |
| derekh | sshnaidm: yup, we could do that either, any idea how long a fix would take? | 11:01 |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Temporarily remove puppet-ceph from ci slaves https://review.openstack.org/340717 | 11:03 |
| shardy | derekh: I fast-approved them both to hopfully unbreak things | 11:04 |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Temporarily disable introspection https://review.openstack.org/342106 | 11:04 |
| sshnaidm | derekh, the good solution will require a change in puppet-ironic, the port is just hardcoded there, and it will take a time | 11:04 |
| sshnaidm | derekh, the hack might be faster, but it's not ready yet | 11:05 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud: Remove hardcoded endpoints in hiera https://review.openstack.org/342103 | 11:05 |
| shardy | The puppet folks are pretty responsive, if we get a patch up adding a parameter I'm sure we can get it merged pretty soon | 11:05 |
| shardy | For now lets just disable introspection | 11:05 |
| derekh | shardy: ok | 11:05 |
| *** pkovar has joined #tripleo | 11:06 | |
| derekh | sshnaidm: ok, lets see if we can get puppet-ironic patched and get it back on ASAP | 11:06 |
| sshnaidm | derekh, ok, perhaps I'll need some help with this | 11:07 |
| sshnaidm | derekh, why don't we run 2 jobs on each patch? | 11:07 |
| sshnaidm | derekh, resources lack? | 11:08 |
| derekh | sshnaidm: since move to rh2 we don't have capacity, once we are back on rh1 again we can run all the jobs | 11:08 |
| sshnaidm | derekh, and how does it know which job to run? | 11:09 |
| derekh | sshnaidm: if you need to check the nonha job you can do it in the tripleo-ci repositoty with check experimental | 11:09 |
| derekh | sshnaidm: we're running the nonha job on the puppet modules and DIB , all the tripleo projects have the ha job | 11:09 |
| sshnaidm | derekh, ok | 11:10 |
| derekh | sshnaidm: its configured in here, you can check each repository to see what jobs are run for it http://git.openstack.org/cgit/openstack-infra/project-config/tree/zuul/layout.yaml | 11:10 |
| derekh | sshnaidm: sorry, you probably knew that | 11:11 |
| jaosorior | shardy, is there any manipulation of the hieradata that comes from the config_settings in the composable services? | 11:13 |
| jaosorior | I think the map_merge that is done in the base services is messing up some of it | 11:13 |
| openstackgerrit | Martin André proposed openstack/tripleo-validations: Port the network environment validation https://review.openstack.org/341586 | 11:13 |
| jaosorior | in EmilienM's patch https://review.openstack.org/#/c/341616/2 it seems that if we do "%{::fqdn}" it ends up without the quotes | 11:13 |
| jaosorior | for some reason | 11:13 |
| shardy | jaosorior: I'd guess it's the yaml parsing stripping the quotes | 11:14 |
| shardy | probably needs double quoting? | 11:14 |
| jaosorior | shardy: I guess using something like '"%{::fqdn}"' | 11:15 |
| jaosorior | would work | 11:15 |
| shardy | yeah, IIRC we already do that in some other places | 11:16 |
| jaosorior | I'll try that then | 11:16 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable keystone to use the SSL middleware https://review.openstack.org/341354 | 11:21 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable glance to use the SSL middleware https://review.openstack.org/341367 | 11:21 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Migrate Puppet Hieradata to composable services https://review.openstack.org/341616 | 11:21 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert Swift ringbuilder to composable services format https://review.openstack.org/338551 | 11:21 |
| *** lucasagomes is now known as lucas-hungry | 11:23 | |
| sshnaidm | does anybody know who populates file /httpboot/inspector.ipxe ? | 11:24 |
| *** masco|afk is now known as masco | 11:25 | |
| jaosorior | sshnaidm: I think it comes from the element | 11:25 |
| jaosorior | sshnaidm: https://github.com/openstack/instack-undercloud/tree/master/elements/ipxe | 11:26 |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: Add Plan Deletion https://review.openstack.org/299937 | 11:27 |
| shardy | sshnaidm: FYI there are plans to move that to use puppet instead: | 11:28 |
| shardy | https://review.openstack.org/#/c/320014/ | 11:28 |
| shardy | I'm not sure on the status of that | 11:28 |
| dtantsur | sshnaidm, you're setting ironic port, not inspector port.. what exactly are you trying to achieve? | 11:28 |
| dtantsur | sshnaidm, oh damn, we have a hardcoded port: https://github.com/openstack/puppet-ironic/blob/master/templates/inspector_dnsmasq_http.erb#L13 | 11:30 |
| sshnaidm | dtantsur, it's inspector port: https://paste.fedoraproject.org/391043/14684958/ | 11:30 |
| sshnaidm | dtantsur, yes, and not only there | 11:30 |
| dtantsur | sshnaidm, wanna propose a patch to puppet-ironic? | 11:30 |
| sshnaidm | dtantsur, also in /httpboot/inspector.ipxe | 11:30 |
| dtantsur | yeah https://github.com/openstack/puppet-ironic/blob/master/templates/inspector_ipxe.erb | 11:31 |
| openstackgerrit | Dmitry Tantsur proposed openstack/instack-undercloud: Enable LLDP collection during introspection if inspection_extras is on https://review.openstack.org/342117 | 11:31 |
| *** zoliXXL is now known as zoli|lunch | 11:31 | |
| *** yamahata has quit IRC | 11:32 | |
| dtantsur | shardy, these file are managed by puppet already, we just need an option in puppet-ironic | 11:32 |
| sshnaidm | dtantsur, I'm not so familiar with openstack puppet modules, so if you want take this, it'd be great | 11:32 |
| dtantsur | okie, lemme do it | 11:32 |
| sshnaidm | dtantsur, thanks | 11:33 |
| dtantsur | do we have the nonha job running on puppet modules? | 11:33 |
| dtantsur | otherwise I'll have to guess if I did it right or not... | 11:33 |
| shardy | dtantsur: Yes we do but you'll have to test the fix via a patch reverting https://review.openstack.org/#/c/342106/ with a Depends-On the puppet change | 11:34 |
| dtantsur | got it | 11:35 |
| openstackgerrit | Martin André proposed openstack/tripleo-validations: Port the network environment validation https://review.openstack.org/341586 | 11:41 |
| sshnaidm | dtantsur, can we use this parameter for everything? https://review.openstack.org/#/c/340339/5/scripts/deploy.sh | 11:42 |
| dtantsur | sshnaidm, I'm not sure how it works in puppet.. also in theory this could be different ports | 11:43 |
| *** jcoufal has joined #tripleo | 11:44 | |
| sshnaidm | dtantsur, why so? | 11:44 |
| dtantsur | if you have different http servers? | 11:46 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud: Remove hardcoded endpoints in hiera https://review.openstack.org/342103 | 11:47 |
| dtantsur | anyway, there is the weirdness of puppet-ironic: it contains two loosely-coupled services | 11:47 |
| EmilienM | hello | 11:52 |
| *** tzumainn has joined #tripleo | 11:52 | |
| *** tzumainn has quit IRC | 11:52 | |
| *** tzumainn has joined #tripleo | 11:52 | |
| jaosorior | EmilienM: Hey dude, how's it going? | 11:52 |
| jaosorior | EmilienM: Hey, regarding the certmonger puppet patches. https://review.openstack.org/#/c/341390/ Are they OK as they are or should I be moving it to the profiles directory? | 11:54 |
| EmilienM | jaosorior: looking | 11:54 |
| EmilienM | jaosorior: it's an helper, it's good to me as it is now | 11:55 |
| jaosorior | shardy, marios, could you take a look again at https://review.openstack.org/#/c/341390/ and https://review.openstack.org/#/c/341405/2 ? | 11:56 |
| *** zoli|lunch is now known as zoli | 11:57 | |
| *** zoli is now known as zoliXXL | 11:57 | |
| marios | jaosorior: ack noted | 11:58 |
| EmilienM | jaosorior: thx for your patch on hiera thing | 11:59 |
| jaosorior | EmilienM: No worries. | 11:59 |
| *** rodrigods has quit IRC | 11:59 | |
| *** rodrigods has joined #tripleo | 11:59 | |
| *** zoliXXL is now known as zoli|afk | 12:05 | |
| openstackgerrit | Martin André proposed openstack/tripleo-validations: Port the network environment validation https://review.openstack.org/341586 | 12:06 |
| EmilienM | shardy: hey, any chance to review https://review.openstack.org/#/c/340548/ and https://review.openstack.org/#/c/340547/ please? | 12:12 |
| *** adarazs is now known as adarazs_brb | 12:13 | |
| *** pradk has joined #tripleo | 12:14 | |
| *** adarazs_brb is now known as adarazs | 12:16 | |
| *** tzumainn has quit IRC | 12:19 | |
| dtantsur | sshnaidm, https://review.openstack.org/#/c/342126/ | 12:20 |
| sshnaidm | dtantsur, thanks! | 12:22 |
| *** lucas-hungry is now known as lucasagomes | 12:24 | |
| *** anshul_ has quit IRC | 12:26 | |
| *** adarazs is now known as adarazs_brb | 12:26 | |
| dtantsur | sshnaidm, I see ironic still using the default port: https://github.com/openstack/instack-undercloud/blob/master/elements/puppet-stack-config/puppet-stack-config.yaml.template#L422 | 12:27 |
| dtantsur | was it even changed? | 12:27 |
| sshnaidm | dtantsur, it's changed in hiera in CI: https://review.openstack.org/#/c/340339/5/scripts/deploy.sh | 12:28 |
| dtantsur | ah, so we need to amend this patch | 12:28 |
| *** dprince has joined #tripleo | 12:28 | |
| * dtantsur does | 12:28 | |
| openstackgerrit | Martin André proposed openstack/tripleo-validations: Port the network environment validation https://review.openstack.org/341586 | 12:29 |
| *** bswartz has joined #tripleo | 12:30 | |
| *** fzdarsky has joined #tripleo | 12:31 | |
| openstackgerrit | Dmitry Tantsur proposed openstack-infra/tripleo-ci: Properly configure introspection HTTP port and reenable it in gate https://review.openstack.org/342146 | 12:31 |
| dtantsur | sshnaidm, shardy ^^^ | 12:32 |
| shardy | dtantsur: thanks for looking into it :) | 12:32 |
| openstackgerrit | Dmitry Tantsur proposed openstack/instack-undercloud: Always configure inspector to use the same HTTP port as Ironic https://review.openstack.org/342147 | 12:34 |
| dtantsur | sshnaidm, shardy, this is also a fix actually ^^^ | 12:34 |
| sshnaidm | dtantsur, thanks, let's wait for CI | 12:35 |
| sshnaidm | maybe it's better to run all jobs on tripleo-ci repo, just checking experimental on each patch there | 12:35 |
| dtantsur | EmilienM, hi! I'm not sure I get what you mean by "environment file" on https://review.openstack.org/#/c/341433/ do you maybe have an example handy? | 12:37 |
| *** athomas has quit IRC | 12:37 | |
| rbrady | still looking for reviews on the deployment workflows patches: http://paste.openstack.org/show/532791/ | 12:37 |
| rbrady | dprince, d0ugal, shardy, jdob ^^ | 12:38 |
| jaosorior | EmilienM: So now the issue is here https://review.openstack.org/#/c/341616/3/puppet/services/nova-base.yaml in line 57 | 12:38 |
| *** jayg|g0n3 is now known as jayg | 12:38 | |
| jaosorior | the quotes get removed and the resulting value is %{hiera('mysql_virtual_ip')} | 12:38 |
| EmilienM | dtantsur: https://github.com/openstack/tripleo-heat-templates/blob/master/environments/neutron-plumgrid.yaml | 12:39 |
| EmilienM | this is an environment file | 12:39 |
| jaosorior | however, I cannot surround that by single quotes like I did for the fqdns because of the single quotes that are already there | 12:39 |
| EmilienM | jaosorior: oh damn | 12:39 |
| jaosorior | so any other suggestions? | 12:39 |
| openstackgerrit | Merged openstack/puppet-tripleo: Implement Pacemaker service profile https://review.openstack.org/340547 | 12:39 |
| jdob | rbrady: right, sorry, I looked at the stuff we were talking about yesterday but didnt get a chance for a full review. i'll look today | 12:39 |
| dtantsur | EmilienM, but I already have a resource registry, what should I put there? :/ | 12:40 |
| EmilienM | dtantsur: to allow people to use this environment when they want to deploy ironic | 12:40 |
| *** links has quit IRC | 12:40 | |
| EmilienM | the registry is different | 12:40 |
| dtantsur | I don't get it, sorry | 12:40 |
| shardy | rbrady: ack, I'm planning to get to them today | 12:41 |
| EmilienM | jaosorior: "%{hiera('mysql_bind_host')}" <--- in heat-engine.yaml already :/ | 12:41 |
| shardy | d0ugal: is there a WIP tripleoclient patch anywhere I can try? | 12:41 |
| shardy | I don't see one in the review queue but perhaps I'm missing it | 12:41 |
| EmilienM | dtantsur: right now, in your patch, there is no way someone deploys Ironic on compute nodes | 12:42 |
| dtantsur | correct | 12:42 |
| EmilienM | what you need to do is to create an env file that loads the service | 12:42 |
| *** athomas has joined #tripleo | 12:42 | |
| EmilienM | like here https://github.com/openstack/tripleo-heat-templates/blob/master/environments/neutron-plumgrid.yaml#L11 | 12:42 |
| dtantsur | the last time I've checked, folks said I can't append to list of services on controller | 12:43 |
| jaosorior | so for some reason it's behaving differently for nova | 12:43 |
| EmilienM | so when people will want to deploy ironic on compute, they will load this env file | 12:43 |
| dtantsur | EmilienM, no, people won't | 12:43 |
| EmilienM | why? | 12:43 |
| dtantsur | EmilienM, ironic is deployed on a controller, completely | 12:43 |
| dtantsur | there are not compute nodes in ironic | 12:43 |
| d0ugal | shardy: Not yet, I've not had time this week yet and I don't actually know if all the deploy patches are working. | 12:43 |
| EmilienM | right, whatever node | 12:43 |
| d0ugal | rbrady: Have you done a full deploy? | 12:43 |
| EmilienM | we need to stop thinking controller/compute | 12:43 |
| dtantsur | EmilienM, the last I've tried, people said I'll have to replicate the whole list of services on controller | 12:44 |
| d0ugal | shardy: I'll try and get something rough together tomorrow (I go home tonight) | 12:44 |
| EmilienM | dtantsur: do we have a spec about it? | 12:44 |
| jaosorior | EmilienM: Not sure why it's behaving differently, but yeah, stuff coming from nova-base ends up without the quotes. | 12:44 |
| jaosorior | for instance, this | 12:44 |
| jaosorior | nova::db::mysql::host: %{hiera('mysql_virtual_ip')} | 12:44 |
| jaosorior | which ends up causing the error | 12:44 |
| dtantsur | EmilienM, yep https://review.openstack.org/320995 | 12:44 |
| EmilienM | or a place to discuss? I'm sure dprince & shardy have thoughts on this, we already did something similar for NovaComputeLibvirt | 12:44 |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: Add Mistral profiles https://review.openstack.org/323431 | 12:45 |
| dtantsur | EmilienM, I probably just don't understand how https://github.com/openstack/tripleo-heat-templates/blob/master/environments/neutron-plumgrid.yaml applies to my case.. | 12:45 |
| shardy | d0ugal: ack, thanks, I may try smoke-testing the mistral patches directly via mistral CLI then | 12:46 |
| *** skramaja has quit IRC | 12:46 | |
| *** zoli|afk is now known as zoli | 12:46 | |
| *** zoli is now known as zoliXXL | 12:47 | |
| EmilienM | jaosorior: looking | 12:47 |
| d0ugal | shardy: Thanks, that would be really useful. If you do and have success could you send me a log of what you did? (might save me a bit of time) | 12:47 |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: DO NOT MERGE: testing ironic https://review.openstack.org/329872 | 12:47 |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: Add composable service for configuring Nova compute with Ironic https://review.openstack.org/341433 | 12:47 |
| dtantsur | EmilienM, fix 2 other issues ^^^ | 12:48 |
| EmilienM | dtantsur: I would like to talk with dprince when he has time about this ironic thing | 12:48 |
| EmilienM | jaosorior: so it looks like when it's not in an array it does not work? | 12:50 |
| dprince | EmilienM: whats up? | 12:52 |
| *** masco has quit IRC | 12:52 | |
| *** weshay has joined #tripleo | 12:53 | |
| EmilienM | dprince: wdyt about https://review.openstack.org/#/c/341433/ and how to enable ironic driver? | 12:53 |
| *** d0ugal has quit IRC | 12:53 | |
| *** rlandy has joined #tripleo | 12:53 | |
| shardy | EmilienM: Really we need https://review.openstack.org/#/c/330414/ in heat before we can have an environment that enables a service | 12:55 |
| shardy | we don't have a way to append a service to the ComputeServices list atm, other than copying the entire list | 12:56 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Composable Pacemaker service https://review.openstack.org/340548 | 12:56 |
| EmilienM | shardy: ok | 12:56 |
| dprince | EmilienM: I think it is okay | 12:56 |
| dprince | EmilienM: +2 | 12:56 |
| dtantsur | reminder: s/ComputeServices/ControllerServices/ (just for clarity) | 12:56 |
| EmilienM | good then :) | 12:56 |
| shardy | EmilienM: personally I think it's OK to land it anyway, and we can document copying the services list until the heat stuff lands | 12:56 |
| EmilienM | dtantsur: you addressed the 2 other comments, thanks. Lgtm now | 12:57 |
| shardy | jdob: ^^ no pressure or anything :) | 12:57 |
| dprince | EmilienM: it may or may not layer on top of the libvirt driver. But that is okay because I really see this as being something that is more usuable once we have composable roles (hopefully soon) | 12:57 |
| openstackgerrit | Adriano Petrich proposed openstack/tripleo-quickstart: WIP gate upgrade https://review.openstack.org/342161 | 12:57 |
| dtantsur | thanks all! | 12:58 |
| jdob | shardy: believe me, I'd love to get back to doing real work, but OSP 9 shit keeps getting in the way | 12:58 |
| *** adarazs_brb is now known as adarazs | 12:58 | |
| shardy | jdob: no worries - would you be OK with me taking a first pass at it if I get time? | 12:58 |
| jdob | and that's likely to only get worse over the next 2-3 weeks, so if it's time sensitive, it might make sense to reassign :( | 12:58 |
| EmilienM | dprince: ack | 12:59 |
| *** cdearborn has joined #tripleo | 12:59 | |
| jdob | ya, sucks because i'd have liked to actually do something, but I understand if it's needed | 12:59 |
| shardy | jdob: Ok, no worries, I'll see if I can get it going later this week | 12:59 |
| EmilienM | dprince: so we can land his patch here: https://review.openstack.org/#/c/341420/ | 12:59 |
| jdob | shardy: literally in the middle of this conversation, I got pinged about more osp9 bugs to look at :) | 12:59 |
| shardy | jdob: Don't worry, there's plenty more on my Heat TODO list I can pass your way when you've got time :) | 12:59 |
| jdob | sounds good | 12:59 |
| jrist | haha | 13:00 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Move MySQL Galera within composable services https://review.openstack.org/340448 | 13:00 |
| EmilienM | dprince: do you have thoughts about the hiera thing we discussed with jaosorior ? see at 08:44:43 | 13:01 |
| dprince | EmilienM: my thought (initially) is that we shouldn't ever need to do inline hiera lookups in heat templates for VIPs because now we have the endpoint map | 13:02 |
| dprince | EmilienM: so rather than | 13:02 |
| dprince | EmilienM: nova::db::mysql::host: %{hiera('mysql_virtual_ip')} | 13:02 |
| dprince | EmilienM: why not just look it up in Heat via the EndpointMap? | 13:03 |
| EmilienM | I just want to use hiera | 13:03 |
| dprince | EmilienM: then you can explicitly quote it as you wish too | 13:03 |
| EmilienM | and again I can't | 13:03 |
| *** pgadiya has quit IRC | 13:03 | |
| dprince | EmilienM: got a patch? link? | 13:03 |
| EmilienM | dprince: https://review.openstack.org/#/c/341616/ | 13:03 |
| dprince | EmilienM: we are using Hiera | 13:03 |
| dprince | EmilienM: this is largely a stylistic question I think | 13:04 |
| EmilienM | feel free to push over if you have an idea | 13:04 |
| jaosorior | EmilienM: I had a similar issue in another commit. Another option is to use double quotes https://review.openstack.org/#/c/342103/4/elements/puppet-stack-config/puppet-stack-config.yaml.template | 13:04 |
| jaosorior | as I did in line 252 | 13:04 |
| *** lblanchard has joined #tripleo | 13:05 | |
| EmilienM | jaosorior: we already use double quotes I think | 13:05 |
| jaosorior | EmilienM: I mean, double quotes instead of the single ones | 13:05 |
| jaosorior | right now it's single quotes surrounding single ones | 13:05 |
| EmilienM | oh inside! | 13:06 |
| dprince | EmilienM: https://review.openstack.org/#/c/341616/3/puppet/services/nova-base.yaml | 13:06 |
| dprince | EmilienM: commented. There is also an ordering concern... which again using the Heat function avoids | 13:06 |
| *** julim has joined #tripleo | 13:06 | |
| jaosorior | yep | 13:06 |
| dprince | EmilienM: I really would prefer using EndpointMap for your patch | 13:06 |
| EmilienM | {get_param: [EndpointMap, MysqlVirtual, host]} | 13:07 |
| EmilienM | ? | 13:07 |
| *** jubapa has joined #tripleo | 13:07 | |
| dprince | we've gone to the trouble of wiring EndpointMap in, and it is the only required service param | 13:07 |
| dprince | EmilienM: yes, something like that | 13:07 |
| jaosorior | dprince: Thought the idea was to ultimately get rid of EndpointMap | 13:07 |
| dprince | jaosorior: is someone pushing a patch for this? | 13:07 |
| shardy | dprince: Hey, I was wondering if you'd given any thought to how we can convert cluster-wide stuff like swift-devices-and-proxy-config.yaml to the new composable model? | 13:08 |
| dprince | jaosorior: regardless, until we do we have a comperable mechanism that avoids ordering issues I'd prefer to keep using it | 13:08 |
| shardy | I've been looking at it today and can see a few ways it might work, but wondered if you'd already been down this path | 13:08 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Migrate Puppet Hieradata to composable services https://review.openstack.org/341616 | 13:08 |
| EmilienM | dprince: good ? https://review.openstack.org/#/c/341616/3..4/puppet/services/nova-base.yaml | 13:08 |
| shardy | jaosorior: the idea wasn't to get rid of it, but compose it from the service templates | 13:09 |
| jaosorior | dprince: To me that's not an issue, but if we're gonna keep using it, then I rather have it's limitations fixed. So, can you give an opinion about this patch? https://review.openstack.org/#/c/308182/ if you think it's a good idea I might as well re-take it | 13:09 |
| openstackgerrit | Dmitry Tantsur proposed openstack/instack-undercloud: Always configure inspector to use the same HTTP port as Ironic https://review.openstack.org/342147 | 13:09 |
| shardy | vs having a giant map which always contains all the services | 13:09 |
| shardy | regardless of what is actually deployed | 13:09 |
| jaosorior | uhm, I still don't see how to get it composed, as we're now getting into a chicken-egg problem here | 13:10 |
| jaosorior | composable services need the endpoint map, so how can it be defined inside them? | 13:10 |
| dprince | EmilienM: it is okay I think. But just make sure your larger patch does the same thing and in all the places. | 13:10 |
| *** bfournie has joined #tripleo | 13:10 | |
| shardy | jaosorior: it needs the map of endpoint IPs for public/internal etc, but the formatting of each endpoint string belongs with the service definition | 13:11 |
| EmilienM | the last thing I see would be bootstrap_nodeid | 13:11 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable keystone to use the SSL middleware https://review.openstack.org/341354 | 13:12 |
| dprince | shardy: I don't have any prototypes for it yet but we'll need/want a composable interface for cluster wide stuff I think | 13:12 |
| shardy | dprince: ack, yeah I think we need it for custom-roles to work, I'll have try a few things | 13:12 |
| *** akuznetsov has quit IRC | 13:12 | |
| shardy | thanks | 13:12 |
| dprince | jaosorior: I agree on iterating to get rid of the maintenance of the static EndpointMap mind you. But EndpointMap itself can't go away I think | 13:13 |
| dprince | jaosorior: do we agree on that? | 13:13 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Migrate Puppet Hieradata to composable services https://review.openstack.org/341616 | 13:13 |
| jaosorior | dprince: I honestly don't care if it stays or leaves. I do urgently need something to work with to modify the endpoitns though | 13:13 |
| dprince | jaosorior: composing the endpoing map in such a way that it better aligns with the composable services is a great idea | 13:13 |
| jaosorior | preferably before newton. Cause I need it for the internal TLS work. Else that just won't happen | 13:14 |
| dprince | jaosorior: I think we should be careful about getting rid of the endpoint map at this point in the release. I'd prefer it stay. For your purposes couldn't we just allow the end user to pass in parameters to control the endpoint map generation? | 13:16 |
| EmilienM | dprince, jaosorior: ok I think I got the hiera thing ok now. Thanks again | 13:16 |
| jaosorior | dprince: We can keep doing it as we do for the public tls | 13:16 |
| jaosorior | to just pass in the map for the endpoints | 13:17 |
| jaosorior | but it becomes very problematic and messy | 13:17 |
| jaosorior | that's why I proposed that patch that uses the map_merge | 13:17 |
| jaosorior | cause then you only overwrite what you need | 13:17 |
| jaosorior | which is this CR https://review.openstack.org/#/c/308182/ | 13:17 |
| jaosorior | EmilienM: where is the bootstrap_nodeid coming from? | 13:18 |
| jaosorior | you're using get_param but I don't see it defined as a paramter | 13:18 |
| jaosorior | *parameter | 13:18 |
| EmilienM | jaosorior: overcloud.yaml | 13:18 |
| dprince | jaosorior: I will look at your patch. If it cleans up the maintenance aspects but still maintains the interfaces I'd go for it | 13:18 |
| jaosorior | dprince: that is indeed the point | 13:18 |
| shardy | jaosorior: it's written via puppet/all-nodes-config.yaml by selecting the first node in each group of nodes | 13:19 |
| jaosorior | shardy: so will it work like this? https://review.openstack.org/#/c/341616/4..5/puppet/services/database/redis-base.yaml | 13:19 |
| jaosorior | thought you needed to pass it as a parameter in that template for that get_param to work | 13:20 |
| shardy | jaosorior: No, that won't work, you have to reference it via hiera | 13:20 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable keystone to use the SSL middleware https://review.openstack.org/341354 | 13:20 |
| jaosorior | EmilienM ^^ | 13:20 |
| jaosorior | EmilienM: I think the bootstrap_nodeid needs to go back to being fetched from hiera | 13:21 |
| EmilienM | but dprince said to use endpointmaps | 13:21 |
| *** akuznetsov has joined #tripleo | 13:22 | |
| EmilienM | jaosorior: just for those 2? | 13:22 |
| jaosorior | EmilienM: see shardy's comment. But yeah, bootstrap_nodeid is not an endpoint. and that get_param won't work there; you would need it in the template as a parameter and then pass it in overcloud.yaml | 13:22 |
| jaosorior | yeah, just the bootstrap_nodeid's | 13:23 |
| *** noslzzp has joined #tripleo | 13:23 | |
| EmilienM | ok | 13:23 |
| *** egafford has joined #tripleo | 13:25 | |
| jaosorior | shardy, by the way, talked to EmilienM about this https://review.openstack.org/#/c/341390/ and being more of a utility than a service he said it's alright that way, can we merge that? :D | 13:25 |
| jaosorior | dprince ^^ | 13:26 |
| shardy | jaosorior: Ok, thats fine with me | 13:26 |
| shardy | It's a bit confusing, we may want to add docs to clarify what goes in a profile and what goes elsewhere | 13:26 |
| jaosorior | this one is on the same track https://review.openstack.org/#/c/341405/ | 13:26 |
| dprince | jaosorior: cool +A | 13:28 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Migrate Puppet Hieradata to composable services https://review.openstack.org/341616 | 13:29 |
| jaosorior | dprince: Thanks dude :D\ | 13:29 |
| derekh | EmilienM: FYI, until we get the ceph hammer/jewel thing sorted out tripleo-ci isn't actually testing puppet-ceph, https://review.openstack.org/#/c/340717/6/toci_gate_test.sh | 13:29 |
| derekh | EmilienM: the nonha test should be back working again now (well we'll know soon | 13:30 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable keystone to use the SSL middleware https://review.openstack.org/341354 | 13:30 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable glance to use the SSL middleware https://review.openstack.org/341367 | 13:31 |
| jaosorior | dprince: Can you check this one out too? https://review.openstack.org/#/c/341405/ I've been testing it for the undercloud' | 13:31 |
| jaosorior | undercloud's certificate autogeneration | 13:32 |
| jaosorior | which is this https://review.openstack.org/#/c/332774/ | 13:32 |
| *** dsariel has quit IRC | 13:34 | |
| *** coolsvap_ has joined #tripleo | 13:35 | |
| *** rhallisey has joined #tripleo | 13:36 | |
| *** zoliXXL is now known as zoli|afk | 13:36 | |
| *** dsariel has joined #tripleo | 13:37 | |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Add third party gating to quickstart https://review.openstack.org/332223 | 13:38 |
| openstackgerrit | Merged openstack/puppet-tripleo: Add base certmonger class https://review.openstack.org/341390 | 13:38 |
| *** coolsvap has quit IRC | 13:39 | |
| slagle | reminder: tripleo deep dive on tripleo-heat-templates in 20 minutes | 13:40 |
| saneax | slagle, ack | 13:40 |
| slagle | https://etherpad.openstack.org/p/tripleo-deep-dive-topics | 13:40 |
| shardy | We're going to be talking about tripleo-heat-templates and heat software-config | 13:41 |
| shardy | if anyone has specific things related to t-h-t you'd like to see covered, feel free to add them to the etherpad | 13:41 |
| shardy | and I'll try to include them | 13:42 |
| *** mbound has quit IRC | 13:43 | |
| *** oshvartz has quit IRC | 13:43 | |
| *** dsariel has quit IRC | 13:45 | |
| jaosorior | slagle, shardy, where do I join there? | 13:47 |
| shardy | jaosorior: link is in the etherpad | 13:48 |
| *** coolsvap_ has quit IRC | 13:49 | |
| EmilienM | jaosorior: i'm having some hard time with https://review.openstack.org/#/c/325983/ and https://review.openstack.org/#/c/328347/ | 13:49 |
| jaosorior | EmilienML: I'll check em out | 13:50 |
| jaosorior | EmilienM: Found something here https://review.openstack.org/#/c/328347/13 | 13:53 |
| trown | slagle: shardy I think meeting ID in the etherpad is not right https://bluejeans.com/914505526 | 13:53 |
| trown | I get "Invalid Meeting Link" | 13:53 |
| saneax | I guess that id is expired, | 13:53 |
| *** r-mibu has quit IRC | 13:54 | |
| *** r-mibu has joined #tripleo | 13:54 | |
| *** ohamada has quit IRC | 13:55 | |
| *** jubapa has quit IRC | 13:56 | |
| EmilienM | jaosorior: damn | 13:57 |
| EmilienM | jaosorior: thx for the second pair of eyes | 13:57 |
| ccamacho | Hey, Invalid Meeting Link for bluejeans.. | 13:58 |
| ccamacho | owww reading now | 13:58 |
| shardy | sec, I'll create a new link | 13:58 |
| shardy | unless slagle is already on it | 13:58 |
| slagle | i'm looking now | 13:59 |
| shardy | kk | 13:59 |
| *** ebalduf has joined #tripleo | 13:59 | |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: profile/base/nova: declare nova class properly https://review.openstack.org/328347 | 13:59 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Remove usage of ::nova class in THT https://review.openstack.org/325983 | 13:59 |
| jmiu | any working link for meeting? | 13:59 |
| jaosorior | jmiu: It's being figured out | 13:59 |
| slagle | bluejenas is being quite slow for me | 14:00 |
| jmiu | i'm so excited for t | 14:00 |
| *** links has joined #tripleo | 14:01 | |
| * beagles observes the peeps waiting in the queue at the door are getting restless ;) | 14:01 | |
| slagle | https://bluejeans.com/176756457/ | 14:01 |
| jaosorior | if I want to run again the experimental nonha, how do I do it? | 14:04 |
| jaosorior | issuing check experimental again didn't seem to work' | 14:04 |
| jaosorior | jmiu: bluejeans is working | 14:07 |
| *** akshai has joined #tripleo | 14:09 | |
| *** coolsvap_ has joined #tripleo | 14:10 | |
| *** zoli|afk is now known as zoli | 14:11 | |
| *** zoli is now known as zoliXXL | 14:11 | |
| *** jubapa has joined #tripleo | 14:12 | |
| *** coolsvap_ has quit IRC | 14:17 | |
| *** ohamada has joined #tripleo | 14:22 | |
| beagles | composable services ftw!!!!! | 14:23 |
| *** greghaynes has quit IRC | 14:23 | |
| *** greghaynes has joined #tripleo | 14:25 | |
| jaosorior | slagle, EmilienM, dprince, shardy: Hey guys, can you take a look at this? https://review.openstack.org/#/c/342103/ | 14:25 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: Do not fail the playbook when libvirt network auto start fails https://review.openstack.org/341870 | 14:30 |
| *** jaosorior has quit IRC | 14:32 | |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: PoC with openvswitch - Composable service validation within profiles https://review.openstack.org/342202 | 14:32 |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Add third party gating to quickstart https://review.openstack.org/332223 | 14:35 |
| openstackgerrit | Adriano Petrich proposed openstack/tripleo-quickstart: WIP gate upgrade https://review.openstack.org/342161 | 14:38 |
| *** pcaruana has quit IRC | 14:39 | |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: PoC Validation - Make it fail (test) https://review.openstack.org/342205 | 14:40 |
| dtantsur | shardy, how do I run the nonha job on https://review.openstack.org/#/c/342146/ ? | 14:43 |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: Add Mistral profiles https://review.openstack.org/323431 | 14:43 |
| *** mbound has joined #tripleo | 14:44 | |
| openstackgerrit | Ryan Brady proposed openstack/tripleo-quickstart: Fix devmode command in README https://review.openstack.org/342209 | 14:46 |
| *** d0ugal has joined #tripleo | 14:47 | |
| EmilienM | dtantsur: check experimental i think | 14:47 |
| dtantsur | thnx | 14:48 |
| *** mbound has quit IRC | 14:48 | |
| *** fzdarsky has quit IRC | 14:49 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Composable Mistral services https://review.openstack.org/323436 | 14:50 |
| *** milan has quit IRC | 14:55 | |
| *** dsariel has joined #tripleo | 14:55 | |
| *** rasca has quit IRC | 15:00 | |
| trown | really good talk shardy++ thanks! | 15:01 |
| marios | shardy: ++ very well done | 15:02 |
| slagle | +1! | 15:02 |
| *** aufi has quit IRC | 15:02 | |
| *** noslzzp has quit IRC | 15:02 | |
| slagle | i rated it a Great talk in bluejeans! | 15:02 |
| trown | lol | 15:02 |
| *** shardy_ has joined #tripleo | 15:02 | |
| jmiu | shardy++ thanks so much | 15:02 |
| saneax | shardy, can you please share your slide as well | 15:02 |
| bfournie | shardy: thank you, that was very useful | 15:02 |
| shardy_ | saneax: sure, I'll add a link to the etherpad | 15:02 |
| saneax | thanks shardy | 15:03 |
| *** fzdarsky has joined #tripleo | 15:03 | |
| *** noslzzp has joined #tripleo | 15:03 | |
| saneax | I also would be waiting for the youtube upload, as I want to go through once more :) | 15:04 |
| saneax | GN folks, thanks again | 15:05 |
| bnemec | derekh: shardy_ EmilienM: Want to point out https://review.openstack.org/#/c/294787 because I think it relates to what we do with rh1 when it comes back. | 15:08 |
| *** dmsimard is now known as dmsimard|afk | 15:08 | |
| bnemec | See my latest comment on it for why. | 15:08 |
| shardy_ | Slide link added to the etherpad | 15:08 |
| *** coolsvap has joined #tripleo | 15:08 | |
| *** shardy has quit IRC | 15:10 | |
| derekh | bnemec: thanks, zoliXXL started on an OVB deployment for rh1 this morning anyways | 15:13 |
| *** mgarciam has joined #tripleo | 15:13 | |
| bnemec | derekh: Okay, that's exactly what I needed to know. Thanks. | 15:14 |
| derekh | bnemec: when this merges how many more networks/ports will be created per env ? https://review.openstack.org/#/c/341161/ | 15:14 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: PoC Validation - Make it fail (test) https://review.openstack.org/342205 | 15:14 |
| bnemec | derekh: I think 5 extra networks, and 25-ish ports. | 15:15 |
| bnemec | So, quite a few. | 15:15 |
| derekh | bnemec: I'm thinking we could just bump nodepools quota way up, see who much they end up peaking at, then push a putch to update their quota accordingly | 15:15 |
| bnemec | But less than we had in the old testenvs. | 15:15 |
| *** mgarciam has quit IRC | 15:16 | |
| *** saneax is now known as saneax_AFK | 15:16 | |
| bnemec | derekh: Yeah, or just leave it high. I doubt we're going to be limited by number of vnets or ports in CI. | 15:16 |
| *** d0ugal has quit IRC | 15:16 | |
| *** links has quit IRC | 15:17 | |
| *** d0ugal has joined #tripleo | 15:18 | |
| derekh | bnemec: ok, I'll bump up the quota and merge both of those patches in the morning (when things are a little less busy) and keep an eye on it | 15:18 |
| bnemec | derekh: Sounds good. It will be nice to get those in so we can start adding net-iso back to CI. | 15:19 |
| derekh | bnemec: or we could do it over the weekend but I'm not around to watch it | 15:19 |
| openstackgerrit | Steven Hardy proposed openstack/puppet-tripleo: Add swift ringbuilder profile https://review.openstack.org/337803 | 15:19 |
| derekh | bnemec: yup | 15:19 |
| bnemec | derekh: Yeah, I'm out of town this weekend too, so that's probably not ideal. | 15:19 |
| bnemec | Although I guess worst case would be it breaks horribly and we have to revert. | 15:20 |
| derekh | bnemec: ok, will do it first thing in the morning, watch and revert if needed | 15:20 |
| bnemec | derekh: ++ | 15:21 |
| dtantsur | folks, how does one debug a failed deployment in gate? :/ | 15:22 |
| dtantsur | where are the logs, where is anything that actually happened? | 15:23 |
| dtantsur | the only thing I see is that nova-conductor didn't start | 15:23 |
| panda | EmilienM: is there a way to diff between patchsets in gertty ? | 15:24 |
| EmilienM | not afik | 15:25 |
| *** ramishra_ has quit IRC | 15:25 | |
| *** ebarrera has quit IRC | 15:25 | |
| *** fragatina has quit IRC | 15:26 | |
| dtantsur | panda, press "p" when in diff view | 15:26 |
| *** ramishra has joined #tripleo | 15:26 | |
| *** radek_ has quit IRC | 15:26 | |
| panda | dtantsur++ | 15:26 |
| EmilienM | oh nice | 15:26 |
| EmilienM | right, F1 gives all options | 15:27 |
| panda | I thought I couldn't hit F1 in diff mode, don't know why | 15:27 |
| panda | mind traps | 15:27 |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart: Add third party gating to quickstart https://review.openstack.org/332223 | 15:28 |
| *** radek__ has quit IRC | 15:28 | |
| bnemec | dtantsur: Check postci.txt in the logs directory. Or the conductor logs from the overcloud node. | 15:28 |
| dtantsur | "Create timed out" aha | 15:28 |
| dtantsur | thanks bnemec | 15:28 |
| *** leanderthal is now known as leanderthal|afk | 15:29 | |
| dtantsur | bnemec, still, is there a way of getting the puppet run logs except for digging into /var/log/messages on every node? | 15:29 |
| bnemec | dtantsur: If it caused a SoftwareDeployment to fail then we should show that in postci. Although for a timeout I'm not positive how it would work. | 15:31 |
| bnemec | dtantsur: Can you link me to the logs? | 15:32 |
| dtantsur | bnemec, http://logs.openstack.org/72/329872/8/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/92a180d/ | 15:32 |
| dtantsur | maybe I'm just missing something | 15:32 |
| *** akuznetsov has quit IRC | 15:33 | |
| bnemec | dtantsur: No, I think the timeout confused our error logging. The nested resources are still shown as in_progress instead of failed, so we didn't report them. | 15:34 |
| dtantsur | I see | 15:35 |
| dtantsur | I just wonder if timeout was caused by my patch | 15:36 |
| bnemec | dtantsur: It actually looks to me like nova-compute may have hung. I see "Jul 14 14:21:16 localhost systemd: Starting OpenStack Nova Compute Server..." with no corresponding "Started" message. | 15:36 |
| *** jaosorior has joined #tripleo | 15:36 | |
| *** jaosorior has quit IRC | 15:36 | |
| bnemec | dtantsur: Oh, but that was caused by no conductor. | 15:37 |
| dtantsur | yep | 15:38 |
| *** jaosorior has joined #tripleo | 15:38 | |
| bnemec | I don't see any attempt to start conductor in /var/log/messages. | 15:38 |
| openstackgerrit | Dougal Matthews proposed openstack/tripleo-common: Minor update to the upgrade-non-controller.sh https://review.openstack.org/324546 | 15:40 |
| jaosorior | bnemec: Hey dude, can you review https://review.openstack.org/#/c/341405/ and this https://review.openstack.org/#/c/342103/ when you get time? Would really appreciate it | 15:41 |
| bnemec | jaosorior: Did you see dprince's comment on the latter? | 15:43 |
| jaosorior | bnemec: I just answered it | 15:43 |
| *** yamahata has joined #tripleo | 15:43 | |
| bnemec | ok | 15:43 |
| *** numan_ has joined #tripleo | 15:46 | |
| jaosorior | EmilienM: So those bootstrap_nodeid ones are problematic | 15:46 |
| jaosorior | I think the only option left is to do use the double quotes like I had mentioned :/ | 15:46 |
| jaosorior | so it would end up as '"%{hiera(\"bootstrap_nodeid\")}"' | 15:47 |
| EmilienM | except if dprince has a better idea? | 15:47 |
| sshnaidm | derekh, how much time the devenv from etherpad will be accessible yet? | 15:49 |
| jaosorior | dprince: HALP | 15:49 |
| *** sshnaidm is now known as sshnaidm|afk | 15:50 | |
| derekh | sshnaidm: It will be available until we switch to rh1 again, at least 2 weeks id say. | 15:50 |
| derekh | sshnaidm|afk: then I think we should set up the same thing on rh1 | 15:50 |
| jaosorior | bnemec, dprince, regarding this patch https://review.openstack.org/#/c/342103/ I can add another hiera key there. But I don't think it would be appropriate ot have something like keystone_identity_uri_v2. That really doesn't say much | 15:52 |
| jaosorior | what about keystone_admin_auth_uri_v2 ? | 15:52 |
| bnemec | I don't have a strong opinion one way or another. It's only used in one place, and it shouldn't be used in any new configuration so I'm not sure a new hiera value is even necessary. | 15:54 |
| bnemec | I was just trying to clarify what was being suggested. | 15:54 |
| jaosorior | well, that's true. | 15:54 |
| jaosorior | and hopefully we can move out of using that v2.0 eventually | 15:54 |
| jaosorior | I tried submitting a patch for that | 15:54 |
| *** florianf has quit IRC | 15:54 | |
| jaosorior | but haven't gotten it working | 15:54 |
| jaosorior | bnemec: Would oy be OK with merging that then? | 15:55 |
| jaosorior | by the way, how can I run the nonha gate on a patch nowadays? | 15:55 |
| *** tremble has quit IRC | 15:55 | |
| jaosorior | check experimental or is there something else? | 15:55 |
| *** coolsvap_ has joined #tripleo | 15:55 | |
| jaosorior | cause check experimental is the multinode job | 15:55 |
| bnemec | jaosorior: I think I'd be okay with it. What's the deal with the ironic endpoint though? Why do we have to append the version in hiera? | 15:56 |
| bnemec | jaosorior: There are two experimental queues now. | 15:56 |
| bnemec | jaosorior: The multinode job is the regular infra queue, but there should also be a tripleo-ci one for the nonha job. | 15:56 |
| jaosorior | bnemec: I actually have no idea | 15:56 |
| jaosorior | bnemec: So how do I run it? | 15:56 |
| bnemec | jaosorior: check experimental should do both, as I understand it. | 15:57 |
| *** mbound has joined #tripleo | 15:57 | |
| *** coolsvap has quit IRC | 15:57 | |
| jaosorior | bnemec: but it seems that if I run check experimental once, I can't re-run it :/ | 15:57 |
| jaosorior | bnemec: by the way, I tried switching ironic to v3 in this patch https://review.openstack.org/#/c/340652/2/elements/puppet-stack-config/puppet-stack-config.yaml.template but that didn't really work out | 15:58 |
| *** ramishra has quit IRC | 15:58 | |
| bnemec | jaosorior: I see we aren't appending a version in the code, so I guess that's fine: https://github.com/openstack/instack-undercloud/blob/master/instack_undercloud/undercloud.py#L663 | 15:58 |
| jaosorior | dtantsur: How acquainted are you with configuring the keystone bits for ironic? | 15:58 |
| dtantsur | not so much, but I have some idea | 15:58 |
| openstackgerrit | Flavio Percoco proposed openstack/tripleo-heat-templates: WIP: Containerized Services for Composable Roles https://review.openstack.org/330659 | 15:58 |
| jaosorior | bnemec: Ideally we should be doing everything versionless though | 15:59 |
| *** ramishra has joined #tripleo | 15:59 | |
| bnemec | jaosorior: Not for the services themselves, I don't think. That's just keystone. | 15:59 |
| bnemec | AFAIK, anyway. | 15:59 |
| jaosorior | bnemec: It should be for the services themselves too | 15:59 |
| dtantsur | I don't think ironic supports V3 for real | 15:59 |
| *** fragatina has joined #tripleo | 15:59 | |
| jaosorior | bnemec: that should be handled by the keystone libraries. (talked to ayoung about it) | 16:00 |
| jaosorior | dtantsur: Why not? | 16:00 |
| jaosorior | dtantsur: Are there hardcoded instances of v2.0 somewhere? | 16:00 |
| dtantsur | jaosorior, well, it's not implemented yet :) patches are up for both | 16:00 |
| dtantsur | well, it's not as simple as s/2.0/3/ | 16:00 |
| jaosorior | dtantsur: I did that work for barbican. Depends on how it was implemented initially | 16:00 |
| dtantsur | essentially we need to switch to keystoneauth lib | 16:00 |
| jaosorior | ah, thought you were using that already | 16:01 |
| jaosorior | well that makes sense then | 16:01 |
| dtantsur | jaosorior, the nova side is https://review.openstack.org/300154 | 16:01 |
| jaosorior | though using keystonemiddleware should have handled that too | 16:01 |
| dtantsur | the ironic side is https://review.openstack.org/236982 | 16:02 |
| *** mbound has quit IRC | 16:02 | |
| EmilienM | I see a lot of no valid host found | 16:02 |
| EmilienM | ex: http://logs.openstack.org/83/325983/23/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/ae73eef/logs/postci.txt.gz#_2016-07-14_15_16_32_000 | 16:02 |
| EmilienM | are we running out of testenvs? | 16:02 |
| jaosorior | EmilienM: We are quite limited on the testenvs actually :/ | 16:03 |
| EmilienM | yeah but why our CI would fail on that and not wait a testenv is free | 16:03 |
| jaosorior | dtantsur: Now it makes sense. Thanks | 16:05 |
| jaosorior | bnemec: Yep, so it seems we do need that endpoint there for ironic anyway. | 16:05 |
| bnemec | No valid host is not a testenv problem. | 16:06 |
| bnemec | AFAICT it's the nova-ironic scheduler race. I've seen a bunch of the nodeassociated errors in the logs on those failed jobs. | 16:07 |
| EmilienM | ah | 16:07 |
| openstackgerrit | Rodrigo Duarte proposed openstack/tripleo-quickstart: Fix minor typo https://review.openstack.org/342260 | 16:07 |
| EmilienM | bnemec: is it fixed by your patch in nova? | 16:08 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: WIP Mistral Deploy https://review.openstack.org/342261 | 16:09 |
| bnemec | EmilienM: No, there's no patch for it that I'm aware of. I think fixing it properly requires a significant rework of the nova scheduler. | 16:10 |
| EmilienM | ok | 16:10 |
| bnemec | We do have workarounds, and I'm not entirely sure why they aren't helping. | 16:10 |
| bnemec | Although it may have to do with the fact that we only deploy one extra node now. I think we had more in the old testenvs. | 16:11 |
| panda | bnemec: do you have a description of this race problem ? | 16:11 |
| *** coolsvap_ has quit IRC | 16:14 | |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: WIP Mistral Deploy https://review.openstack.org/342261 | 16:15 |
| bnemec | panda: https://bugs.launchpad.net/nova/+bug/1341420 and https://bugzilla.redhat.com/show_bug.cgi?id=1257950 | 16:15 |
| openstack | Launchpad bug 1341420 in tripleo "gap between scheduler selection and claim causes spurious failures when the instance is the last one to fit" [Undecided,New] | 16:15 |
| openstack | bugzilla.redhat.com bug 1257950 in openstack-nova "NodeAssociated errors due to apparent race condition between nova-scheduler and ironic" [Medium,Closed: errata] - Assigned to lmartins | 16:15 |
| panda | bnemec: thanks. | 16:16 |
| weshay | adarazs, so the thought is.. if we can run 3rd party CI internally on RHEL we can avoid this problem. .https://review.openstack.org/#/c/340503/1 | 16:16 |
| *** milan has joined #tripleo | 16:16 | |
| weshay | adarazs, I'm going to open a card in rdo-infra to see if they can get us space the logs | 16:16 |
| weshay | myoung, fyi.. | 16:16 |
| adarazs | weshay: okay | 16:17 |
| myoung | weshay, ack | 16:17 |
| *** zoliXXL is now known as zoli|gone | 16:18 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use delorean-current for puppet modules https://review.openstack.org/341429 | 16:18 |
| *** zoli|gone is now known as zoli_gone-proxy | 16:18 | |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: WIP Mistral Deploy https://review.openstack.org/342261 | 16:18 |
| derekh | shardy_: untested but I thinks that most of what would be removed https://review.openstack.org/341429 | 16:19 |
| shardy_ | derekh: awesome, thanks! | 16:20 |
| *** rcernin has quit IRC | 16:21 | |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: WIP Mistral Deploy https://review.openstack.org/342261 | 16:24 |
| *** trown is now known as trown|lunch | 16:24 | |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: WIP Mistral Deploy https://review.openstack.org/342261 | 16:27 |
| *** akshai has quit IRC | 16:27 | |
| *** akshai has joined #tripleo | 16:30 | |
| openstackgerrit | Ana Krivokapic proposed openstack/tripleo-heat-templates: Add constraint for ControllerCount https://review.openstack.org/342272 | 16:30 |
| *** dtantsur is now known as dtantsur|afk | 16:32 | |
| *** dsariel has quit IRC | 16:33 | |
| *** lucasagomes is now known as lucas|afk | 16:35 | |
| d0ugal | shardy_: https://review.openstack.org/#/c/342261/ | 16:38 |
| d0ugal | shardy_: Very very WIP-y. It doesn't work yet. Not sure why either. | 16:38 |
| d0ugal | ... but you did ask you see it early :) | 16:38 |
| d0ugal | rbrady: FYI too ^ | 16:39 |
| d0ugal | The two patches it depends on conflict with each other amusingly :) | 16:39 |
| *** numan_ has quit IRC | 16:40 | |
| *** numans has joined #tripleo | 16:40 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable keystone to use the SSL middleware https://review.openstack.org/341354 | 16:41 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Enable glance to use the SSL middleware https://review.openstack.org/341367 | 16:41 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Migrate Puppet Hieradata to composable services https://review.openstack.org/341616 | 16:41 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: WIP Mistral Deploy https://review.openstack.org/342261 | 16:43 |
| *** r2bit has quit IRC | 16:43 | |
| jaosorior | bnemec: Is there a way to run the same nonha job that runs for puppet? I just tried doing check experimental and I noticed it only runs the nonha multinode. Which seems to be different | 16:44 |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Fixup os-collect-config unit file so it won't fail on update https://review.openstack.org/342278 | 16:44 |
| bnemec | jaosorior: It looks like experimental-tripleo might be hung on the tempest job. See http://status.openstack.org/zuul/ | 16:45 |
| jaosorior | bnemec: But the experimental-tripleo seems to only run for tripleo-ci | 16:47 |
| jaosorior | not for any other changes | 16:47 |
| bnemec | jaosorior: Ah, that's possible. I don't know much about how it's configured in infra. | 16:47 |
| jaosorior | aaah | 16:47 |
| jaosorior | alright | 16:47 |
| jaosorior | I guess that's the case | 16:47 |
| jaosorior | I couldn't figure out how to run it :/ | 16:48 |
| *** d0ugal has quit IRC | 16:49 | |
| *** r2bit has joined #tripleo | 16:50 | |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: PoC with openvswitch - Composable service validation within profiles https://review.openstack.org/342202 | 16:51 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: Do not fail the playbook when libvirt network auto start fails https://review.openstack.org/341870 | 16:51 |
| *** derekh has quit IRC | 16:51 | |
| *** julim has quit IRC | 16:53 | |
| EmilienM | jaosorior: I was about to do the same \o/ | 16:53 |
| jaosorior | EmilienM: The patch looks good though. I forgot to score it | 16:56 |
| EmilienM | still having bad time with https://review.openstack.org/#/c/325983/ | 16:56 |
| EmilienM | err | 16:56 |
| EmilienM | timeout | 16:56 |
| jaosorior | EmilienM: Last time that one failed was because of the no valid host | 16:57 |
| jaosorior | so it wasn't really an issue with the patch | 16:57 |
| jaosorior | I think I issued a recheck there | 16:57 |
| jaosorior | yep, I did | 16:57 |
| jaosorior | and then you did | 16:57 |
| jaosorior | :P | 16:57 |
| EmilienM | ahah | 16:57 |
| *** jpich has quit IRC | 16:57 | |
| *** ifarkas has quit IRC | 16:57 | |
| *** weshay is now known as weshay_food | 16:58 | |
| ccamacho | Hey jaosorior, EmilienM, shardy, dprince I finally made this pass CI https://review.openstack.org/#/c/323436/ It is the Mistral composable submission, when having some time, can you give it a try?! | 16:59 |
| openstackgerrit | Adriano Petrich proposed openstack/tripleo-quickstart: WIP gate upgrade https://review.openstack.org/342161 | 17:00 |
| openstackgerrit | Michele Baldessari proposed openstack/puppet-tripleo: WIP Remove cinder-volume pacemaker constraints https://review.openstack.org/341493 | 17:00 |
| EmilienM | ccamacho: I will, thanks for your work | 17:00 |
| *** akrivoka has quit IRC | 17:00 | |
| jaosorior | ccamacho: reviewing | 17:01 |
| ccamacho | EmilienM I'll start now with the Zaqar one, when passing CI ill let you know. | 17:01 |
| ccamacho | EmilienM jaosorior, thanks!!! :) | 17:01 |
| jaosorior | ccamacho | 17:01 |
| jaosorior | where is mistral in haproxy? | 17:02 |
| *** dtrainor has quit IRC | 17:02 | |
| jaosorior | I don't see it defined there | 17:02 |
| jaosorior | and it's also missing from the enable-tls.yaml environment file | 17:03 |
| jaosorior | almost there | 17:03 |
| jaosorior | everything else looks pretty good though | 17:03 |
| *** weshay_food is now known as weshay | 17:04 | |
| *** jtomasek has quit IRC | 17:04 | |
| *** dbecker has quit IRC | 17:04 | |
| *** jrist has quit IRC | 17:04 | |
| ccamacho | jaosorior, not sure if that is needed. Ill work on that then :) As the pacemaker profiles were removed | 17:05 |
| jaosorior | ccamacho: maybe I'm misunderstanding something then. But to access the mistral API, it needs to be proxied by HAProxy, therefore it needs to have an entry there | 17:07 |
| jaosorior | I don't find that part though | 17:07 |
| jaosorior | did I miss it> | 17:07 |
| jaosorior | uhm... no, I really think it's not there | 17:07 |
| *** akshai has quit IRC | 17:08 | |
| jaosorior | ccamacho: But basically you need to add it here https://github.com/openstack/puppet-tripleo/blob/master/manifests/haproxy.pp like the other services | 17:08 |
| jaosorior | and subsequently turn on the mistral flag (that will need to be written too) on the profiles in t-h-t | 17:09 |
| ccamacho | jaosorior, thanks for that hint! :) | 17:09 |
| jaosorior | ccamacho: No worries dude. I know it can get really tedious with all the moving parts | 17:10 |
| jaosorior | ccamacho: If you have any questions regarding that haproxy.pp file, let me know. I've been hitting my head against the wall because of it for long enough that I can now help :D | 17:11 |
| ccamacho | jaosorior sure thing :) | 17:12 |
| *** athomas has quit IRC | 17:14 | |
| *** ohamada has quit IRC | 17:19 | |
| *** dmsimard|afk is now known as dmsimard | 17:20 | |
| *** akshai has joined #tripleo | 17:20 | |
| *** noslzzp has quit IRC | 17:20 | |
| *** noslzzp has joined #tripleo | 17:20 | |
| *** jaosorior has quit IRC | 17:22 | |
| *** jaosorior has joined #tripleo | 17:24 | |
| jaosorior | EmilienM: By the way | 17:25 |
| jaosorior | what will we do with your nova patches? | 17:25 |
| jaosorior | seems that the puppet-tripleo one will fail | 17:25 |
| jaosorior | and then it will be fixed by the t-h-t one | 17:25 |
| jaosorior | is that the plan? To merge them at the same time? | 17:25 |
| *** shardy_ has quit IRC | 17:25 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Composable Mistral services https://review.openstack.org/323436 | 17:26 |
| *** rcernin has joined #tripleo | 17:32 | |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: Add Mistral profiles https://review.openstack.org/323431 | 17:33 |
| jaosorior | ccamacho: Nice! | 17:34 |
| ccamacho | jaosorior but that's only because your feedback! Thanks dude! | 17:34 |
| jaosorior | uhm... | 17:35 |
| jaosorior | gotta check if we need to add any firewall rules | 17:35 |
| ccamacho | I already added them to /puppet/hieradata/controller.yaml right? into tripleo::firewall::firewall_rules: | 17:36 |
| jaosorior | true that | 17:36 |
| jaosorior | that's all I can think of then | 17:37 |
| jaosorior | +2 | 17:37 |
| ccamacho | lets wait for CI :) | 17:37 |
| jaosorior | yep | 17:37 |
| *** jaosorior_ has joined #tripleo | 17:43 | |
| *** tesseract- has quit IRC | 17:44 | |
| *** jaosorior has quit IRC | 17:46 | |
| *** tosky has quit IRC | 17:47 | |
| *** trown|lunch is now known as trown | 17:47 | |
| *** jaosorior_ is now known as jaosorior | 17:47 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: Do not fail the playbook when libvirt network auto start fails https://review.openstack.org/341870 | 17:49 |
| jaosorior | EmilienM: alright... so now it failed with something pretty weird.. | 17:49 |
| *** mbound has joined #tripleo | 17:49 | |
| jaosorior | resources.ControllerNodesPostDeployment.resources.ControllerOvercloudServicesDeployment_Step5.resources[0]: Deployment to server failed: deploy_status_code: Deployment exited with non-zero status | 17:50 |
| jaosorior | no clue what that actually means | 17:50 |
| jaosorior | ah | 17:50 |
| jaosorior | nevermind | 17:50 |
| jaosorior | found the trace now | 17:50 |
| *** shivrao has joined #tripleo | 17:51 | |
| pradk | any ideas on debugging this ci failure? doesnt tell me much in the logs sadly http://logs.openstack.org/56/333556/17/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/e11c5e0/console.html | 17:55 |
| jaosorior | pradk: Resource CREATE failed: The Referenced Attribute (ControllerServiceChain role_data) is incorrect. | 17:56 |
| pradk | hmm where do you see that? | 17:57 |
| jaosorior | pradk http://logs.openstack.org/56/333556/17/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/e11c5e0/logs/postci.txt.gz | 17:57 |
| pradk | hmm thought i fixed that | 17:58 |
| *** julim has joined #tripleo | 17:59 | |
| jaosorior | well guys, I gotta go | 18:01 |
| jaosorior | have a good day everyone! | 18:01 |
| *** jaosorior has quit IRC | 18:02 | |
| *** mcornea has quit IRC | 18:05 | |
| *** pkovar has quit IRC | 18:10 | |
| *** radek2 has quit IRC | 18:10 | |
| *** rajinir has joined #tripleo | 18:10 | |
| openstackgerrit | Michele Baldessari proposed openstack/puppet-tripleo: WIP Remove cinder-volume pacemaker constraints https://review.openstack.org/341493 | 18:17 |
| *** mgould is now known as mgould|afk | 18:24 | |
| openstackgerrit | Merged openstack/instack-undercloud: Remove hardcoded endpoints in hiera https://review.openstack.org/342103 | 18:29 |
| *** dsariel has joined #tripleo | 18:34 | |
| *** mbound_ has joined #tripleo | 18:42 | |
| *** mbound has quit IRC | 18:43 | |
| *** weshay is now known as weshay_afk | 18:45 | |
| *** mbound_ has quit IRC | 18:50 | |
| jmiu | any tips on figuring out what went wrong when deploying with tripleo-quickstart? i mean, aside from the -v you pass into ansible | 18:50 |
| *** dprince has quit IRC | 18:54 | |
| openstackgerrit | Michele Baldessari proposed openstack/puppet-tripleo: WIP Remove cinder-volume pacemaker constraints https://review.openstack.org/341493 | 19:03 |
| *** fzdarsky has quit IRC | 19:05 | |
| trown | jmiu: what TASK did it fail on? | 19:05 |
| jmiu | trown: it failed on the task where it tries to grab the undercloud vm ip | 19:08 |
| jmiu | trown: wait what. i just reran it out of hopelessness and it's made it past that step already | 19:09 |
| trown | jmiu: could you paste a log somewhere? fpaste.org or chunk.io or the like | 19:09 |
| trown | hmm | 19:09 |
| jmiu | trown: it seems to be an unstable failure/success (to me). maybe you can help me understand why sometimes it fails and sometimes it succeeds | 19:10 |
| jmiu | trown: working off the same git co and the same virthost | 19:10 |
| trown | jmiu: ya hard to say without logs, but if it failed to get the undercloud IP, my best guess is that the undercloud VM did not come up for some reason | 19:11 |
| jmiu | trown: you want the logs of the quickstart run? | 19:11 |
| trown | jmiu: ya that would be the first place I would look, maybe something failed earlier that should have been fatal but quickstart continued running | 19:12 |
| *** shivrao has quit IRC | 19:13 | |
| jmiu | trown: unsuccessful run only, right? | 19:16 |
| trown | jmiu: sure | 19:18 |
| jmiu | trown: warning: super verbose output ahead | 19:20 |
| jmiu | trown: http://chunk.io/f/fae51ebb6ba34c7080c4e6f921ab8b24 | 19:20 |
| *** mbound has joined #tripleo | 19:22 | |
| *** mbound has quit IRC | 19:26 | |
| *** shivrao has joined #tripleo | 19:26 | |
| trown | jmiu: hmm, it seems like the undercloud just did not come up in the default amount of allowed time... all of the previous tasks look legitimately successful, and show the undercloud image being usable since there are tasks for injecting ssh keys and an instackenv.json that worked | 19:30 |
| openstackgerrit | Michele Baldessari proposed openstack/puppet-tripleo: Make ::tripleo::profile::base classes work with multiple nodes https://review.openstack.org/338387 | 19:30 |
| trown | jmiu: there is a variable to increase the number of retries if you hit it again: https://git.openstack.org/cgit/openstack/tripleo-quickstart/tree/roles/libvirt/setup/undercloud/tasks/main.yml#n182 | 19:30 |
| trown | jmiu: so you could add "--extra-vars undercloud_ip_retries=40" to the quickstart.sh command | 19:31 |
| *** noslzzp has quit IRC | 19:31 | |
| jmiu | trown: nice! thank you | 19:31 |
| *** noslzzp has joined #tripleo | 19:32 | |
| *** fragatina has quit IRC | 19:32 | |
| *** radek2 has joined #tripleo | 19:36 | |
| jmiu | trown: how did you identify this? in the task right before, the "start undercloud vm" task -- it seemingly says that state did change to "running" | 19:36 |
| trown | jmiu: ya, it is a bit of a guess... the previous task changes immediately though, as soon as libvirt starts to boot the VM. The reason we have the loop to grab the IP is to wait for the undercloud network to actually come up. | 19:38 |
| trown | jmiu: in your case that part did not happen... they why is still a bit of a mystery, but given it succeeded on the next try, my guess is it hit the timeout the first time | 19:39 |
| *** fragatina has joined #tripleo | 19:40 | |
| jmiu | trown: i see...so the information reported in the logs, even with -v on, doesn't reflect everything that is happening. i guess it also depends on how you define what "start undercloud" means haha | 19:41 |
| trown | jmiu: ya -v is just ansible debug... it is really only useful for connection issues since it shows the verbose SSH logging | 19:42 |
| jmiu | trown: that's not that helpful, especially if the problems aren't with the connexion but rather with the vms coming up, etc | 19:43 |
| *** Goneri has quit IRC | 19:43 | |
| trown | jmiu: ya I am not sure how to best introspect libvirt in that case, but I agree it would be nice to provide some more information for that failure scenario | 19:44 |
| *** weshay_afk is now known as weshay | 19:46 | |
| jmiu | trown: are there any resources on how to debug tripleo-quickstart failures in general? sure, it is a bunch of ansible playbooks that do things with libvirt (among other things), so you *could* make the argument that i am really asking how better to debug libvirt in general | 19:47 |
| trown | jmiu: well, it is not supposed to fail :) | 19:47 |
| jmiu | trown: but it seems to me that oooq is meant to simplify the whole tripleo setup thing, | 19:47 |
| *** julim has quit IRC | 19:47 | |
| trown | jmiu: and when it does it is usually after the virt-setup stuff... everything after that creates logs on the undercloud that can usually point to what went wrong | 19:48 |
| jmiu | trown: that's what i mean -- things fail from time to time. i feel that i haven't been doing really effective debugging when this fails, just because i don't know where to start | 19:48 |
| *** julim has joined #tripleo | 19:49 | |
| jmiu | trown: those logs are gold. in the past i've had overcloud deploys fail | 19:49 |
| trown | jmiu: ya for anything after the undercloud is booted and reachable via ssh I think we have pretty good logging at this point | 19:51 |
| jmiu | trown: yeah, at that point it seems rather a question of "where do i find these logs" | 19:51 |
| trown | jmiu: to answer your explicit question though, there are not really any "troubleshooting quickstart failures" docs, mostly because we want to make those failures obvious | 19:52 |
| trown | jmiu: everything is on the undercloud in the /home/stack dir | 19:52 |
| jmiu | trown: the failed_deployment_uuid.log step is brilliant. i don't know who added that but thank you | 19:53 |
| trown | jmiu: nice, that is something we should surface in tripleo itself better, but I put it in for CI purposes | 19:54 |
| trown | jmiu: we could even improve that log quite a bit... I use https://paste.fedoraproject.org/391206/52610814/ to download the logs from CI and just return the colorized puppet stderr | 19:55 |
| jmiu | trown: in fact, some of shardy's steps do that | 19:56 |
| *** fragatina has quit IRC | 20:04 | |
| openstackgerrit | Andreas Florath proposed openstack/diskimage-builder: Refactor: block-device handling (local loop) https://review.openstack.org/319591 | 20:05 |
| trown | jmiu: ya that was distilled from shardy's blog post on debugging heat | 20:05 |
| *** fragatina has joined #tripleo | 20:06 | |
| *** akshai has quit IRC | 20:06 | |
| jmiu | trown: thank you so much, by the way. i really appreciate it :) | 20:10 |
| *** pradk has quit IRC | 20:11 | |
| *** akshai has joined #tripleo | 20:11 | |
| trown | jmiu: no problem, glad it is useful | 20:12 |
| matbu | trown: where can i get the list of the packages which are installed on the undercloud.qcow2 image? | 20:12 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: Do not fail the playbook when libvirt network auto start fails https://review.openstack.org/341870 | 20:13 |
| *** pradk has joined #tripleo | 20:13 | |
| trown | matbu: you could look at the promote job artifacts that creates it to get a real list of what is there, but the list that is used is in the image building role: https://github.com/redhat-openstack/ansible-role-tripleo-image-build/blob/master/vars/default_package_list.yml | 20:15 |
| trown | matbu: note I have work in progress to get rid of that list though... still working out how to convert a DIB built overcloud image into an undercloud image to do that though | 20:15 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: PoC with openvswitch - Composable service validation within profiles https://review.openstack.org/342202 | 20:17 |
| *** yamahata has quit IRC | 20:17 | |
| matbu | trown: ack thanks | 20:17 |
| trown | matbu: https://ci.centos.org/artifacts/rdo/jenkins-tripleo-quickstart-promote-master-delorean-minimal-358/undercloud/var/log/rpm.list.gz is what I mean by checking promote job artifacts | 20:17 |
| *** radek2 has quit IRC | 20:20 | |
| openstackgerrit | Michele Baldessari proposed openstack/puppet-tripleo: Make ::tripleo::profile::base classes work with multiple nodes https://review.openstack.org/338387 | 20:21 |
| *** jschlueter is now known as jschlueter|afk | 20:26 | |
| *** lblanchard has quit IRC | 20:33 | |
| matbu | trown: i don't understand how a package is present on my deployment and not on the rpm.list on the job artifact | 20:39 |
| trown | matbu: maybe it got pulled in during undercloud install? | 20:40 |
| trown | matbu: it is a different version, or a net new package? | 20:41 |
| matbu | trown: maybe but the timestamp in the yum.log match with the timestamp of the yum.log in the job artifact | 20:41 |
| matbu | trown: a new package | 20:41 |
| trown | matbu: odd... and timestamp of the new package is at the same time as the rest of the undercloud packages? | 20:42 |
| matbu | trown: nop | 20:43 |
| trown | matbu: it is maybe from undercloud install time then? | 20:43 |
| matbu | trown: the weird package is at July 14 06:05 and undercloud is at 20:00 | 20:43 |
| matbu | trown: it's really like this package comes with a undercloud image | 20:44 |
| matbu | trown: and on my old UC image I don't have this pacakge | 20:45 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: PoC with openvswitch - Composable service validation within profiles https://review.openstack.org/342202 | 20:46 |
| trown | matbu: what package is this? | 20:46 |
| trown | matbu: maybe the centos base image added it | 20:46 |
| matbu | trown: python2-barbicanclient | 20:46 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: PoC Validation - Make it fail (test) https://review.openstack.org/342205 | 20:46 |
| trown | wtf... that wouldnt be in the centos base image | 20:46 |
| trown | matbu: this is the RDO mitaka image? | 20:47 |
| matbu | trown: nop liberty | 20:47 |
| trown | matbu: it is in the rpm list on the promote job https://ci.centos.org/artifacts/rdo/jenkins-tripleo-quickstart-promote-liberty-delorean-minimal-461/undercloud/var/log/rpm.list.gz | 20:49 |
| *** akshai has quit IRC | 20:49 | |
| trown | matbu: and it was installed as a dependency of something in the package list in the image build role https://ci.centos.org/artifacts/rdo/jenkins-tripleo-quickstart-promote-liberty-delorean-minimal-461/undercloud/var/log/yum.log.gz | 20:50 |
| matbu | trown: wow i tired :) | 20:50 |
| matbu | idk why i missed it | 20:50 |
| trown | could be the python2 thing... I hate that :P | 20:50 |
| trown | I did not see it at first glance | 20:50 |
| matbu | trown: is there a way to skipped it ? | 20:51 |
| matbu | trown: i think it comes as dependencies of sahara | 20:51 |
| trown | matbu: you would have to build a custom image... or just unistall sahara, that should probably not be installed on the undercloud anyways | 20:52 |
| trown | matbu: in fact adding sahara to https://github.com/redhat-openstack/ansible-role-tripleo-image-build/blob/master/defaults/main.yml#L37 would probably be a good thing | 20:52 |
| matbu | trown: k i don't know much about sahara but | 20:54 |
| trown | matbu: because even if we get rid of the other package list we will still have those two small lists for converting overcloud to undercloud | 20:54 |
| trown | matbu: it Big Data as a service, no reason to have it on the undercloud | 20:54 |
| matbu | trown: this package breaks upgrade for few days | 20:55 |
| *** akshai has joined #tripleo | 20:55 | |
| trown | matbu: k, liberty promote has been very consistent, so if we fix it in the image building role, we can get a new image pretty quick | 20:56 |
| *** akshai has quit IRC | 20:56 | |
| *** akshai has joined #tripleo | 21:03 | |
| matbu | trown: k (sorry my network connectivity is really weak atm) | 21:07 |
| matbu | trown: do you want me to make a review on the role ? | 21:08 |
| trown | matbu: sure, or I can do it tomorrow, but if you submit it I can merge it :) | 21:08 |
| matbu | trown: ack | 21:09 |
| trozet | hi everyone. We are seeing some deployments have no keystone services or endpoints at the end of a deployment, but deployment returns a success. I follow the rabbit hole in tripleoclient and end up here: https://github.com/openstack/os-cloud-config/blob/master/os_cloud_config/keystone.py#L333 | 21:11 |
| *** trown is now known as trown|outtypewww | 21:11 | |
| trozet | it doesn't look there is any exception raised if services/endpoints cannot be created? | 21:11 |
| *** egafford has quit IRC | 21:18 | |
| ccamacho | trozet, something like http://paste.openstack.org/show/532920/ ???? | 21:21 |
| ccamacho | Im getting those in my CI since a few days ago "randomly" | 21:22 |
| *** ccamacho is now known as ccamacho|out | 21:22 | |
| trozet | ccamacho|out: i'm not sure if that error was present. Will need to check | 21:23 |
| trozet | ccamacho|out: this is on stable/mitaka | 21:23 |
| *** rcernin has quit IRC | 21:24 | |
| *** rhallisey has quit IRC | 21:24 | |
| *** jayg is now known as jayg|g0n3 | 21:24 | |
| *** [1]cdearborn has joined #tripleo | 21:28 | |
| *** hparekh has quit IRC | 21:30 | |
| *** akshai has quit IRC | 21:31 | |
| *** cdearborn has quit IRC | 21:31 | |
| *** weshay is now known as weshay_afk | 21:34 | |
| *** fzdarsky has joined #tripleo | 21:39 | |
| *** yamahata has joined #tripleo | 21:43 | |
| *** hparekh has joined #tripleo | 21:43 | |
| *** tzumainn has joined #tripleo | 21:55 | |
| *** openstackgerrit has quit IRC | 22:03 | |
| *** openstackgerrit has joined #tripleo | 22:03 | |
| *** tzumainn has quit IRC | 22:18 | |
| openstackgerrit | Merged openstack/puppet-tripleo: Add Mistral profiles https://review.openstack.org/323431 | 22:23 |
| *** rlandy has quit IRC | 22:24 | |
| * bnemec notices that CI is rather lightly loaded | 22:32 | |
| * bnemec rechecks all the things | 22:32 | |
| *** [1]cdearborn has quit IRC | 22:47 | |
| *** noslzzp has quit IRC | 22:57 | |
| *** noslzzp has joined #tripleo | 22:57 | |
| *** noslzzp has quit IRC | 23:01 | |
| *** fzdarsky has quit IRC | 23:02 | |
| *** jcoufal has quit IRC | 23:06 | |
| *** jcoufal has joined #tripleo | 23:06 | |
| *** fragatina has quit IRC | 23:07 | |
| *** jerrygb has joined #tripleo | 23:08 | |
| *** jcoufal has quit IRC | 23:08 | |
| *** fzdarsky has joined #tripleo | 23:15 | |
| *** saneax_AFK is now known as saneax | 23:16 | |
| *** fzdarsky has quit IRC | 23:24 | |
| *** jerrygb has quit IRC | 23:24 | |
| *** rook- has quit IRC | 23:28 | |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Add mulitnode CI job support to tripleo-ci https://review.openstack.org/324777 | 23:31 |
| *** pradk_ has quit IRC | 23:32 | |
| *** bfournie has quit IRC | 23:36 | |
| *** fpan has quit IRC | 23:41 | |
| *** pradk has quit IRC | 23:47 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!