| *** rwsu has quit IRC | 00:08 | |
| openstackgerrit | Merged openstack/diskimage-builder: secure_path in sudoers: deal with possible quotes https://review.openstack.org/350582 | 00:21 |
|---|---|---|
| *** abehl has quit IRC | 00:24 | |
| *** weshay has quit IRC | 00:36 | |
| *** limao has joined #tripleo | 00:39 | |
| *** Guest7941 is now known as saneax_AFK | 00:39 | |
| slagle | EmilienM: i yamld again | 00:39 |
| *** trozet has quit IRC | 00:54 | |
| *** rhallisey has quit IRC | 01:00 | |
| *** trozet has joined #tripleo | 01:06 | |
| *** penick has quit IRC | 01:07 | |
| *** ebalduf has quit IRC | 01:19 | |
| openstackgerrit | Harry Rybacki proposed openstack/tripleo-quickstart: Expand templated bash to allow for doc generation https://review.openstack.org/347592 | 01:24 |
| *** chlong|mtg has joined #tripleo | 01:56 | |
| *** jdob has quit IRC | 02:03 | |
| *** bfournie has joined #tripleo | 02:04 | |
| *** maeca1 has joined #tripleo | 02:12 | |
| *** jcoufal has joined #tripleo | 02:17 | |
| *** tzumainn has quit IRC | 02:21 | |
| dtrainor | trying to debug a keystone issue here. /etc/keystone/keystone.conf is now supposed to reference /etc/keystone/keystone-paste.ini for pasteconfig, correct? trying to figure out what role /usr/share/keystone now plays, if any, in the runtime configuration of keystone under apache's wsgi | 02:24 |
| dtrainor | line 200 of major_upgrade_controller_pacemarker_1.sh provides the only real clue that I've seen so far | 02:25 |
| dtrainor | keystone_spec.rb has an override_params value of paste_config to still be /usr/share/keystone/keystone-paste.ini | 02:27 |
| *** morazi has quit IRC | 02:29 | |
| dtrainor | I'm trying to configure keystone for cors and it's as if none of my changes are taking effect, that's why i ask | 02:34 |
| *** penick has joined #tripleo | 02:35 | |
| *** penick_ has joined #tripleo | 02:40 | |
| *** penick has quit IRC | 02:41 | |
| *** penick_ is now known as penick | 02:41 | |
| *** jeckersb is now known as jeckersb_gone | 02:52 | |
| *** jcoufal has quit IRC | 02:55 | |
| *** yamahata has quit IRC | 02:58 | |
| *** krotscheck_ has joined #tripleo | 03:12 | |
| *** krotscheck has quit IRC | 03:13 | |
| *** krotscheck_ is now known as krotscheck | 03:14 | |
| *** julim has joined #tripleo | 03:26 | |
| *** limao has quit IRC | 03:26 | |
| *** trozet has quit IRC | 03:32 | |
| *** links has joined #tripleo | 03:37 | |
| *** limao has joined #tripleo | 03:38 | |
| *** dsariel has quit IRC | 03:41 | |
| *** dsariel has joined #tripleo | 03:43 | |
| *** sshnaidm|afk has quit IRC | 03:47 | |
| *** beagles has quit IRC | 03:47 | |
| *** wfoster has quit IRC | 03:47 | |
| *** jpeeler has quit IRC | 03:47 | |
| *** jtomasek has quit IRC | 03:48 | |
| *** bana_k has joined #tripleo | 03:49 | |
| *** wfoster has joined #tripleo | 03:49 | |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/tripleo-quickstart: Add undercloud reboot https://review.openstack.org/350489 | 03:51 |
| *** jeckersb_gone has quit IRC | 03:51 | |
| *** coolsvap has joined #tripleo | 03:51 | |
| *** b3nt_pin has joined #tripleo | 03:52 | |
| *** b3nt_pin is now known as Guest54563 | 03:52 | |
| *** jpeeler has joined #tripleo | 03:53 | |
| *** jpeeler has quit IRC | 03:53 | |
| *** jpeeler has joined #tripleo | 03:54 | |
| *** jeckersb_gone has joined #tripleo | 03:55 | |
| *** fragatina has quit IRC | 03:57 | |
| *** sshnaidm|afk has joined #tripleo | 03:58 | |
| *** fragatina has joined #tripleo | 03:58 | |
| *** fragatina has quit IRC | 03:58 | |
| *** jtomasek has joined #tripleo | 03:59 | |
| *** dsariel has quit IRC | 04:08 | |
| *** rlandy has quit IRC | 04:13 | |
| *** dsariel has joined #tripleo | 04:22 | |
| *** julim has quit IRC | 04:27 | |
| *** maeca1 has quit IRC | 04:34 | |
| *** radeks has joined #tripleo | 04:46 | |
| *** dsariel has quit IRC | 04:49 | |
| *** numan_ has joined #tripleo | 04:56 | |
| *** masco has joined #tripleo | 04:56 | |
| *** bana_k has quit IRC | 05:03 | |
| *** skramaja has joined #tripleo | 05:19 | |
| *** jaosorior has joined #tripleo | 05:23 | |
| *** oshvartz has quit IRC | 05:23 | |
| *** fultonj has quit IRC | 05:25 | |
| *** rcernin has joined #tripleo | 05:29 | |
| *** dsariel has joined #tripleo | 05:30 | |
| jaosorior | marios: How's it going? | 05:33 |
| jaosorior | marios: Got anything in need for review? | 05:33 |
| *** liverpooler has joined #tripleo | 05:42 | |
| *** david-lyle_ has joined #tripleo | 05:43 | |
| *** liverpooler is now known as yrabl|WFH | 05:43 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Remove Compute role parameters from overcloud.yaml https://review.openstack.org/347926 | 05:43 |
| *** yrabl|WFH is now known as liverpooler | 05:43 | |
| *** liverpooler has quit IRC | 05:44 | |
| *** liverpooler has joined #tripleo | 05:44 | |
| *** beekneemech has joined #tripleo | 05:45 | |
| *** ramishra_ has joined #tripleo | 05:45 | |
| *** r2bit__ has joined #tripleo | 05:46 | |
| *** r-mibu_ has joined #tripleo | 05:46 | |
| *** StevenK_ has joined #tripleo | 05:46 | |
| *** ramishra has quit IRC | 05:48 | |
| *** david-lyle has quit IRC | 05:48 | |
| *** r2bit has quit IRC | 05:48 | |
| *** r-mibu has quit IRC | 05:48 | |
| *** yolanda has quit IRC | 05:48 | |
| *** karthiks has quit IRC | 05:48 | |
| *** bnemec has quit IRC | 05:48 | |
| *** StevenK has quit IRC | 05:48 | |
| *** yolanda has joined #tripleo | 05:49 | |
| marios | sup jaosorior | 05:49 |
| *** liverpoo1er has joined #tripleo | 05:49 | |
| *** karthiks has joined #tripleo | 05:50 | |
| *** liverpooler has quit IRC | 05:50 | |
| jaosorior | marios: Chilling, having a review round and getting coffee. How's it going over there? | 05:50 |
| *** liverpooler has joined #tripleo | 05:51 | |
| jaosorior | marios: Anyway, so if you need some reviews, I can check stuff out. Currently I'm just going through the list in no particular order. | 05:57 |
| *** cmyster has quit IRC | 05:59 | |
| marios | jaosorior: thanks man will do, theres a bunch of stuff i still need to revisit/rebase after long context switch - want to try and poke at the manila netap at some point | 06:00 |
| jaosorior | marios: Talking about manila | 06:02 |
| *** jprovazn has joined #tripleo | 06:02 | |
| jaosorior | marios: manila::keystone::auth assumes two versions of manila will always be deployed | 06:02 |
| jaosorior | so with the switch to puppet handling the keystone endpoints, it creates an extra endpoint pointing to localhost | 06:02 |
| jaosorior | so I ended up adding an endpoint for it in t-h-t to address that https://review.openstack.org/#/c/352353/ | 06:03 |
| jaosorior | if you can take a look, it's not a big CR | 06:03 |
| jaosorior | I'm not sure if I needed to capitalize those v1 though | 06:04 |
| marios | jaosorior: ack will look | 06:04 |
| jaosorior | marios: Nevermind, I needed to capitalize them | 06:04 |
| jaosorior | just checked | 06:04 |
| jaosorior | lol, sorry dude | 06:04 |
| marios | jaosorior: np man added to list anyway | 06:04 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Add hieradata for Manila v1 and v2 endpoints https://review.openstack.org/352353 | 06:07 |
| jaosorior | marios: If you have time, there is a really similar patch addressing cinder v3 https://review.openstack.org/#/c/353364/3 | 06:07 |
| jaosorior | that one I did capitalize | 06:07 |
| *** bana_k has joined #tripleo | 06:20 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Convert EndpointMap to not require per-service VIP parameters https://review.openstack.org/342900 | 06:23 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Convert ServiceNetMap to a nested template https://review.openstack.org/353032 | 06:23 |
| *** florianf has joined #tripleo | 06:24 | |
| *** oshvartz has joined #tripleo | 06:30 | |
| *** dbecker has joined #tripleo | 06:41 | |
| *** pgadiya has joined #tripleo | 06:44 | |
| *** pcaruana has joined #tripleo | 06:46 | |
| *** dbecker has quit IRC | 06:46 | |
| *** tesseract- has joined #tripleo | 06:53 | |
| *** saneax_AFK is now known as saneax | 06:55 | |
| *** cmyster has joined #tripleo | 06:55 | |
| *** saneax is now known as Guest20032 | 06:55 | |
| *** bana_k has quit IRC | 06:57 | |
| *** liverpoo1er has quit IRC | 06:58 | |
| *** limao has quit IRC | 07:00 | |
| *** Goneri has joined #tripleo | 07:01 | |
| *** limao has joined #tripleo | 07:01 | |
| *** ioggstream has joined #tripleo | 07:03 | |
| *** panda|zZ is now known as panda | 07:04 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud: HAProxy should start before ironic and inspector https://review.openstack.org/353865 | 07:04 |
| *** mcornea has joined #tripleo | 07:06 | |
| *** limao has quit IRC | 07:06 | |
| *** shardy has joined #tripleo | 07:10 | |
| *** milan has joined #tripleo | 07:11 | |
| jaosorior | shardy: Hey dude, just so you know, I rebased a couple of your patches | 07:11 |
| shardy | jaosorior: ack, thanks | 07:11 |
| jaosorior | this https://review.openstack.org/#/c/342900/ (also fixed the missing replacement for the nobrackets SQL) | 07:12 |
| jaosorior | and this https://review.openstack.org/#/c/353032/3 | 07:12 |
| jaosorior | * no brackets mysql endpoint | 07:12 |
| shardy | jaosorior: thanks for that, I meant to do it yesterday but ran out of time | 07:12 |
| jaosorior | shardy: Also, I think I found the issue that I had wrongly thought was a heat issue. | 07:12 |
| shardy | aha, what was it? | 07:12 |
| jaosorior | in the nonha case | 07:13 |
| jaosorior | haproxy is started sometimes after some openstack services | 07:13 |
| jaosorior | so, for instance, ironic inspector tries to ping ironic | 07:13 |
| jaosorior | and it can't | 07:13 |
| jaosorior | so one of the two (haven't figured out which) ends up in an error state because of that | 07:14 |
| *** jprovazn has quit IRC | 07:14 | |
| jaosorior | that's why those test runs always failed when trying to provision the nodes | 07:14 |
| jaosorior | and somehow that ended up being reflected further in the overcloud deploy | 07:14 |
| shardy | jaosorior: interesting, so we can probably fix that with some puppet orchestration in the undercloud manifest? | 07:14 |
| jaosorior | shardy: this is my attempt to do so https://review.openstack.org/353865 | 07:14 |
| *** jpena|off is now known as jpena | 07:15 | |
| jaosorior | but yeah, it's quite an intermittent issue in the nonha case | 07:16 |
| shardy | jaosorior: ack, thanks for tracking this down, lets see what CI thinks but the patch lgtm | 07:16 |
| jaosorior | cause of puppet's random ordering | 07:16 |
| shardy | it'd also be good to discuss with dtantsur|afk if this is an ironic/inspector bug (e.g should it retry instead of fail) | 07:16 |
| shardy | perhaps you can raise a bug with details so that can be discussed? | 07:16 |
| jaosorior | shardy: checking the logs, it seems it retries around 7 times | 07:17 |
| shardy | ah, perhaps that's configurable | 07:17 |
| shardy | but +1 on the puppet orchestration regardless | 07:17 |
| jaosorior | shardy: Do we know of any cause of those: NodeLocked: Node a7dc9450-7a4c-4746-b000-827e50a45985 is locked by host localhost.localdomain, please retry after the the current operation is completed. | 07:20 |
| *** akuznetsov has joined #tripleo | 07:22 | |
| *** ifarkas_afk is now known as ifarkas | 07:23 | |
| *** ebarrera has joined #tripleo | 07:28 | |
| *** mbound has joined #tripleo | 07:31 | |
| jaosorior | shardy: Hey dude, could you check this out? https://review.openstack.org/#/c/353655/ it's a commit bumping up the overcloud nodes' memory a bit | 07:32 |
| *** ohamada has joined #tripleo | 07:33 | |
| marios | jaosorior: there goes my 32 gb minimum virt box setup | 07:33 |
| marios | jaosorior: and that is my "good" box, the other one has 24 gb but it does nothing in its life cos 24gb | 07:34 |
| *** aufi has joined #tripleo | 07:35 | |
| jaosorior | marios: FWIW it's for the ovb jobs | 07:35 |
| marios | jaosorior: yah was just looking at that. was going to comment about why 5.5 not 6 but i saw it wasn't for instack-undercloud | 07:36 |
| jaosorior | marios: I have a 32gb box and it works pretty good for a basic virtual HA setup | 07:41 |
| marios | jaosorior: yeah me too i was thinking it was suddenly not enough on master | 07:42 |
| marios | jaosorior: though, 32gb is just enough for 3control 1 compute 5gb nodes with a slightly bigger undercloud node | 07:43 |
| *** jpich has joined #tripleo | 07:43 | |
| jaosorior | yeah | 07:43 |
| jaosorior | it's not optimal :( | 07:43 |
| jaosorior | wish I could get a 64GB machine | 07:43 |
| marios | jaosorior: yeah exactly i have heard rumours hewbrocca-afk was handing some out behind the bike shed yesterday but it could all be lies. srsly though there have been rumblings about a new round of h/w for devs so ask whoever approves those things for you | 07:45 |
| *** ramishra_ has quit IRC | 07:45 | |
| marios | jaosorior: at least i've heard of beaker boxes getting bumped to 64 gb anyway so hoping mine will at some point too carlos was looking into that | 07:46 |
| jaosorior | nice! | 07:46 |
| *** chlong|mtg has quit IRC | 07:46 | |
| jaosorior | marios: I actually have a physical machine bellow my desk :P | 07:46 |
| *** ramishra has joined #tripleo | 07:46 | |
| jaosorior | might need to upgrade it myself | 07:46 |
| jaosorior | used to have a node in beaker. But we don't have as much resources (different org) | 07:47 |
| marios | jaosorior: yeah my main box is actually a machine here with 32gb ram but would be great to have another beaker box too. | 07:47 |
| marios | jaosorior: like setup for >1 version cos upgrades testing/bz etc | 07:47 |
| jaosorior | yeah, every time I need to test a different version I need to whipe out my dev setup | 07:48 |
| marios | jaosorior: indeed, been meaning to open this box up cos not sure what is happening with the ram slots... i suspect 4x8 | 07:48 |
| *** ramishra has quit IRC | 07:50 | |
| *** ramishra has joined #tripleo | 07:50 | |
| *** jaosorior is now known as jaosorior_brb | 07:53 | |
| *** jprovazn has joined #tripleo | 07:56 | |
| *** limao has joined #tripleo | 07:57 | |
| *** akuznetsov has quit IRC | 07:58 | |
| *** jtomasek_ has joined #tripleo | 08:02 | |
| *** gfidente has joined #tripleo | 08:04 | |
| *** florianf has quit IRC | 08:05 | |
| *** TSCHAK has quit IRC | 08:07 | |
| *** florianf has joined #tripleo | 08:10 | |
| *** hewbrocca-afk is now known as hewbrocca | 08:11 | |
| *** amoralej|off is now known as amoralej | 08:12 | |
| hewbrocca | Morning folks | 08:14 |
| hewbrocca | Yes, ram is raining from the heavens | 08:14 |
| shadower | \o/ | 08:15 |
| marios | onto the golden flocks of roaming boxen | 08:16 |
| *** lucas-dinner is now known as lucasagomes | 08:16 | |
| panda | if it is this kind of ram https://upload.wikimedia.org/wikipedia/commons/8/80/Bighorn_ram_animal_ovis_canadensis.jpg we may be in trouble | 08:16 |
| *** dbecker has joined #tripleo | 08:17 | |
| hewbrocca | LOL | 08:19 |
| *** dtantsur|afk is now known as dtantsur | 08:29 | |
| *** paramite has joined #tripleo | 08:30 | |
| *** dtantsur is now known as dtantsur|bbl | 08:32 | |
| openstackgerrit | Martin André proposed openstack/tripleo-common: Allow running validation against different plans https://review.openstack.org/318194 | 08:32 |
| openstackgerrit | Martin André proposed openstack/tripleo-common: Disable retry files for ansible validations https://review.openstack.org/329039 | 08:32 |
| openstackgerrit | Martin André proposed openstack/tripleo-common: Workflow to setup validations https://review.openstack.org/313632 | 08:32 |
| openstackgerrit | Martin André proposed openstack/tripleo-common: Workflows to load validations https://review.openstack.org/353898 | 08:32 |
| openstackgerrit | Martin André proposed openstack/tripleo-common: Workflows to run validations https://review.openstack.org/353899 | 08:32 |
| *** electrofelix has joined #tripleo | 08:41 | |
| *** coolsvap is now known as _coolsvap_ | 08:41 | |
| *** mbound has quit IRC | 08:43 | |
| *** jaosorior_brb is now known as jaosorior | 08:44 | |
| jaosorior | Simple commit removing a parameter that's no longer used and passing CI https://review.openstack.org/#/c/352907/ | 08:45 |
| jaosorior | By the way, it appears that liberty is broken :/ | 08:46 |
| *** iogg has joined #tripleo | 08:48 | |
| ccamacho | jaosorior quick question im getting several OOM errors in CI for the Mistral service, the OOM fix re sahara that you pushed "SaharaWorkers = 0" worked? | 08:48 |
| *** limao_ has joined #tripleo | 08:48 | |
| *** ioggstream has quit IRC | 08:48 | |
| jaosorior | ccamacho: no | 08:48 |
| ccamacho | ;( | 08:48 |
| jaosorior | ccamacho: Seems that sahara already had a default of 0 | 08:48 |
| shardy | For most services workers=0 means use the number of CPUs | 08:48 |
| shardy | I would say try workers=1 | 08:48 |
| shardy | we already do that for heat and some other services in CI | 08:48 |
| jaosorior | shardy: sahara's documentation claims that 0 is one thread for all sahara | 08:49 |
| shardy | yay for openstack consistency | 08:49 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci: Set Sahara workers to 1 https://review.openstack.org/352729 | 08:50 |
| *** aqkhan__ has quit IRC | 08:50 | |
| ccamacho | Im adding the workers param to puppet-mistral and then to the submission and also to CI to see if this helps | 08:50 |
| jaosorior | sahrdy, ccamacho: restored the commit, there it is ^^ | 08:50 |
| jaosorior | ccamacho: does mistral take workers as a parameter? I thought it didn't | 08:50 |
| *** aqkhan has joined #tripleo | 08:50 | |
| *** limao has quit IRC | 08:51 | |
| jaosorior | * shardy | 08:51 |
| *** aqkhan_ has joined #tripleo | 08:51 | |
| ccamacho | jaosorior not in puppet-mistral but is a parameter supported by mistral API, so first im adding it to puppet-mistral, then the THT and finally to CI | 08:51 |
| ccamacho | jaosorior https://github.com/openstack/mistral/blob/dd195650792129b2f41c87c5160f6ff904ca31ca/mistral/api/service.py#L29 | 08:51 |
| jaosorior | alright | 08:52 |
| jaosorior | ccamacho: Good idea! | 08:52 |
| ccamacho | hopefully this helps a little bit there | 08:52 |
| shardy | ccamacho: you could test it with the generic config interface without changing puppet-mistral | 08:53 |
| shardy | https://github.com/openstack/puppet-mistral/blob/master/manifests/config.pp | 08:53 |
| shardy | by passing that in ControllerExtraConfig | 08:53 |
| shardy | https://github.com/openstack-infra/tripleo-ci/blob/master/test-environments/worker-config-mitaka-and-below.yaml#L5 | 08:53 |
| shardy | like that but s/heat/mistral | 08:54 |
| jaosorior | gfidente: hey dude, could you check this out? https://review.openstack.org/#/c/353364/ | 08:54 |
| ccamacho | shardy =-O cool that at least will show me if it works | 08:54 |
| gfidente | jaosorior, morning | 08:54 |
| shardy | ccamacho: Yeah, it's another approach if you don't want to wait on the puppet-mistral patch | 08:54 |
| jaosorior | gfidente: morning, dude. How's it going? | 08:54 |
| ccamacho | shardy I'll test it then :P thanks! | 08:55 |
| *** aqkhan has quit IRC | 08:55 | |
| shardy | jaosorior: re https://review.openstack.org/#/c/353582/ I can't add a Depends-On or it will be circular | 08:55 |
| shardy | as I already commented | 08:55 |
| shardy | see https://review.openstack.org/#/c/353637/ | 08:55 |
| shardy | I'd suggest we just land them together? | 08:56 |
| jaosorior | I see | 08:56 |
| jaosorior | shardy | 08:56 |
| jaosorior | sure, lets do that | 08:56 |
| jaosorior | shardy: done | 08:57 |
| shardy | jaosorior: thanks! | 08:57 |
| Jokke_ | shardy: I think the worker value behavior really depends if that part of the project was implemented before or after Nova changed the behavior | 09:00 |
| jaosorior | gfidente: thanks for checking it out dude | 09:00 |
| *** bkopilov has joined #tripleo | 09:00 | |
| *** bkopilov has joined #tripleo | 09:01 | |
| *** iogg has quit IRC | 09:01 | |
| shardy | Jokke_: Yeah, but to be fair oslo.service (and before that processutils in oslo-incubator) have been around for years | 09:03 |
| shardy | there's been plenty of time for per-project interfaces to align with the patterns there | 09:03 |
| shardy | I'm sure we'll get more consistent eventually | 09:04 |
| jaosorior | then again, it's openstack | 09:04 |
| Jokke_ | shardy: that, but again it's not that often people are willing to break established ops facing patterns just because someone merged different behavior in oslo | 09:04 |
| shardy | Jokke_: I guess that's my point, options are deprecated all the time, and a really good reason to take that pain is to establish consistent conventions | 09:05 |
| Jokke_ | like we deploy glance on single worker 64 threads just because 0 ment never "cpu count" for glance | 09:05 |
| shardy | some projects aren't willing to do it, which I accept but find frustrating | 09:05 |
| *** abehl has joined #tripleo | 09:05 | |
| shardy | in the long term, it's harmful to everyone involved with deploying openstack | 09:05 |
| shardy | Jokke_: sure, so we have to stick with the status-quo which is all operators need to know the historical quirks of every single service | 09:06 |
| shardy | there are ways it could be done in a backwards compatible way, e.g a new common config section for all projects or something | 09:07 |
| Jokke_ | shardy: I think the biggest problem there is that none of those are actually results of crossproject discussions. Someone implements A into nova dozen others copy it from there, someone else changes it in nova and next guy goes and implements it in oslo purely based on Nova behavior | 09:07 |
| shardy | Jokke_: Yeah, I know, sad isn't it | 09:07 |
| shardy | :( | 09:08 |
| jaosorior | adding an endpoint for manila ready for merging https://review.openstack.org/#/c/352353/4 | 09:08 |
| Jokke_ | shardy: yup and even more frustrating | 09:08 |
| jaosorior | shardy, Jokke_ wasn't that what the cross-project meetings were supposed to solve? | 09:08 |
| *** akrivoka has joined #tripleo | 09:08 | |
| jaosorior | well, issues like that | 09:08 |
| jaosorior | I think it'll get better | 09:09 |
| shardy | gfidente: Hey, if you have time please can you revisit https://review.openstack.org/#/c/342900/ (and the subseqent patch in the series :) | 09:10 |
| Jokke_ | jaosorior: problem with cross project meetings seems to be that they attract that exactly same group of people ;) | 09:11 |
| shardy | gah | 09:11 |
| jaosorior | shardy: I actually think we're gonna need to re-fresh that commit | 09:11 |
| jaosorior | with the merging of the cinder endpoint that was missing | 09:11 |
| jaosorior | the endpoint_map will change | 09:11 |
| shardy | jaosorior: yeah one of us is going to get hit with a rebase | 09:11 |
| Jokke_ | jaosorior: and AFAIK that meeting got discontinued due to the fact that it was all bikeshedding and nothing got done :P | 09:11 |
| shardy | Ok I guess lets land the manila one then I'll rebase it again | 09:12 |
| jaosorior | shardy: I can do the rebase, it's not a big deal | 09:12 |
| shardy | anyone have anything else endpoint related we need to land before I rebase this? | 09:12 |
| jaosorior | shardy: not that I know. But the commits providing a new service will need to be rebased too (mistral and barbican, that I know of) but those are easy | 09:13 |
| jaosorior | shardy, lets land manila https://review.openstack.org/#/c/352353/ and I'll push the rebase for your patches | 09:13 |
| gfidente | shardy, ack | 09:14 |
| social | EmilienM: I already separated it to different class - it's https://review.openstack.org/#/c/350526 and https://review.openstack.org/#/c/334881 | 09:14 |
| social | EmilienM: only issue is that I thought it's mostly ironic inspector dependancy as that is taking care of putting pxe configs in place | 09:14 |
| openstackgerrit | Carlos Camacho proposed openstack-infra/tripleo-ci: TEST DO NOT MERGE - Adding api_workers = 1 for composable mistral https://review.openstack.org/353922 | 09:15 |
| shardy | jaosorior: ack, done, thanks | 09:15 |
| shardy | that EndpointMap patch was always going to be tough to land but I'd like to try to get it done soon if possible | 09:16 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Do not merge, checking composable mistral. https://review.openstack.org/352947 | 09:17 |
| jaosorior | shardy: I know the feeling. I'm really eager to get those patches for puppet managing the services/endpoints merged. Already have a special celebration beer and everything bought for when that merges | 09:18 |
| jaosorior | :P | 09:19 |
| hewbrocca | Not to repeat myself, but | 09:19 |
| hewbrocca | os-cloud-config.die.die.die | 09:19 |
| hewbrocca | might as well | 09:19 |
| shardy | lol | 09:19 |
| * shardy needs to get some special custom-roles beer in ;) | 09:19 | |
| marios | i totally missed this special beer thing for manila :/ | 09:20 |
| gfidente | shardy, thanks for host_nobrackets :) | 09:20 |
| * d0ugal needs to get some Mistral deploy beers | 09:20 | |
| d0ugal | actually | 09:20 |
| d0ugal | s/beers/whisky/ | 09:21 |
| shardy | d0ugal: Hey, good idea :) What's the status of the tripleoclient deploy patch? | 09:22 |
| shardy | ready for it to be pulled, tested, reviewed? | 09:22 |
| * shardy hasn't tried if for a few days | 09:22 | |
| *** bootsha has joined #tripleo | 09:23 | |
| openstackgerrit | Paul Belanger proposed openstack/python-tripleoclient: Initial support for bindep https://review.openstack.org/353929 | 09:26 |
| *** bootsha has quit IRC | 09:27 | |
| jaosorior | shardy, d0ugal by the way, what if the endpoint_map.yaml would be generated from a mistral workflow? That way we wouldn't need to maintain that file like we do now | 09:27 |
| jaosorior | and adding endpoints to tripleo would be way easier | 09:27 |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add cinder-backup profiles https://review.openstack.org/311218 | 09:27 |
| jaosorior | * slightly easier | 09:27 |
| shardy | jaosorior: Yeah we could do but it would make things a bit more opaque for template authors | 09:28 |
| *** psanchez has joined #tripleo | 09:29 | |
| Jokke_ | anyone here familiar enough with puppet that could have a look on my manila cephfs manifest and point me to right direction with it? | 09:30 |
| Jokke_ | https://review.openstack.org/#/c/353928/1 | 09:30 |
| d0ugal | shardy: I am currently trying to figure out the best way to store the list of Heat environments. Just adding it to the Mistral env breaks the UI I think. | 09:31 |
| d0ugal | shardy: and I some rough path munging code that I need to create an obscure heat env to test with | 09:32 |
| hewbrocca | social: ^^^ is familiar with puppet Jokke_ | 09:32 |
| shardy | d0ugal: Hmm, I assumed it would be added to the mistral envioronment | 09:32 |
| d0ugal | shardy: Yeah, but I don't think that is enough for the UI | 09:32 |
| shardy | it will be very confusing if the UI stores the list there and the CLI stores it somewhere else | 09:32 |
| d0ugal | shardy: Indeed, they will both store it there - but I don't think it is that simple. jtomasek said something but I forget what it was | 09:33 |
| shardy | d0ugal: Ok, well if you want someone to try and break it I'm happy to try ;) | 09:33 |
| * shardy has lots of obscure env stuff to try | 09:33 | |
| d0ugal | shardy: Do you have one you could send me? | 09:33 |
| d0ugal | shardy: I'm 99% sure it will break, so might be easier for me to itterate on it | 09:33 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud: Add constraints needed for ironic and inspector https://review.openstack.org/353865 | 09:34 |
| social | Jokke_: I don't think enabled_share_protocols belongs to cephfs backend | 09:36 |
| social | Jokke_: and you are missing spec test for the class | 09:36 |
| social | Jokke_: I'll comment on the review | 09:36 |
| *** apetrich has joined #tripleo | 09:37 | |
| Jokke_ | social: thanks. Yes that spec testing is one thing I had no idea about. I was trying to look into them, but just couldn't figure it out what I'm supposed to be doing there :( | 09:38 |
| *** _coolsvap_ is now known as coolsvap | 09:38 | |
| openstackgerrit | Merged openstack/tripleo-quickstart: Run dstat on the virthost and the undercloud https://review.openstack.org/349591 | 09:40 |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validation to verify connectivity to overcloud https://review.openstack.org/353934 | 09:40 |
| shardy | d0ugal: http://paste.openstack.org/show/554103/ is a starting point | 09:40 |
| shardy | If those all work the basics are probably OK | 09:41 |
| shardy | also need to test combining multiple environments which are conflicting | 09:41 |
| shardy | to ensure the CLI ordering is persisted into the environment_files list | 09:41 |
| d0ugal | oh interesting, I wouldn't have thought of all of those. I'll let you know how I get on. | 09:42 |
| shardy | d0ugal: sounds good, thanks! | 09:43 |
| shardy | d0ugal: we need to think about that password generation stuff too, but that can be done after the basic flow is fixed | 09:43 |
| d0ugal | shardy: Yeah, at the moment I am leaving that in the CLI. | 09:44 |
| shardy | d0ugal: Yeah, I think we'll need to fix the bug I referenced, then move it to mistral afterwards | 09:45 |
| shardy | d0ugal: FYI I also had a discussion in the heat meeting, where we agreed I can probably remove the restriction about having one parameter_group per parameter | 09:45 |
| shardy | which means we can probably add all *Password parameters to a special "generate_password" group | 09:46 |
| d0ugal | shardy: oh, that would be useful. | 09:46 |
| shardy | to identify those which need creating without the hard-coded list | 09:46 |
| openstackgerrit | Martin André proposed openstack/instack-undercloud: Introduce 'enable_validations' option https://review.openstack.org/322893 | 09:47 |
| d0ugal | ccccccfngfccikhvdktgnejrlndrnjhrhjltkibddrgb | 09:54 |
| d0ugal | gah | 09:54 |
| * d0ugal needs to move his yubikey | 09:54 | |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validation to check ulimits on controller https://review.openstack.org/353943 | 09:54 |
| pino|work | (cat-on-keyboard?) | 09:54 |
| shardy | gah, we made the multinode job voting in the gate pipeline didn't we | 09:58 |
| shardy | which probably means https://review.openstack.org/#/c/353582/ will fail validation | 09:58 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-puppet-elements: Add thias/puppet-selinux module https://review.openstack.org/353945 | 09:59 |
| gfidente | social, ^^ | 10:00 |
| gfidente | is there a problem for opm? | 10:00 |
| gfidente | I think puppet-ceph will have to set dircontext ... | 10:00 |
| *** tosky has joined #tripleo | 10:01 | |
| *** bootsha has joined #tripleo | 10:01 | |
| Jokke_ | social: where am I supposed to hook into that enabled_share_protocols if I move it away from the backend manifest? Just trying to figure out the basic logic of these | 10:02 |
| social | gfidente: not much, create new distgit and poke rdo people with it. but what we need it for? | 10:02 |
| gfidente | social, https://review.openstack.org/#/c/353673/1/manifests/osd.pp | 10:03 |
| gfidente | check the comments | 10:03 |
| *** bootsha has quit IRC | 10:04 | |
| *** dtantsur|bbl is now known as dtantsur | 10:04 | |
| social | uuuh I need to read bit more about this | 10:04 |
| gfidente | selinux::dircontext is going to run the same exec | 10:05 |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validate the IP range for the control plane https://review.openstack.org/353950 | 10:05 |
| gfidente | but hey ... what can I tell | 10:05 |
| social | well we used to have rule to use defaults so it should land in location which already has ceph_var_lib_t and if user wants something else he should set it up sooner | 10:06 |
| gfidente | right, we're in the scenario where 'user should set it up' | 10:06 |
| * shardy shakes fist at gate jobs | 10:06 | |
| social | gfidente: this is ugly even with selinux module, we are just hiding the exec :) | 10:07 |
| gfidente | social, right selinux is doing the exec anyway | 10:07 |
| gfidente | but at least if we manage it in ceph module we can test ! -b $data :) | 10:08 |
| gfidente | goal seems good to me though, we just want the right selinux context applied to the data dir given | 10:08 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP - Implement overcloud upgrade job - Mitaka -> Newton https://review.openstack.org/323750 | 10:08 |
| social | gfidente: if we go down this path we should start setting up selinux context for ports on which we listen for keystone, nova... | 10:09 |
| social | gfidente: eg we should use selinux module everywhere :) | 10:09 |
| social | gfidente: which might help us getting rid of openstack-selinux package but I'm not sure if puppet upstream would like this | 10:10 |
| gfidente | ok I see what you mean | 10:10 |
| *** sshnaidm|afk is now known as sshnaidm | 10:10 | |
| gfidente | so there are two problems | 10:10 |
| gfidente | osd.pp expects you to give a path, not sure we can just default it to /var/lib/ceph | 10:11 |
| gfidente | so it's slightly different than default binding ports | 10:11 |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validate haproxy configuration values https://review.openstack.org/353953 | 10:11 |
| gfidente | and then there is the tripleo specific issue, we used to go in /srv/data | 10:11 |
| social | btw if you set selinux contexts before even creating path puppet will use the correct one and it'll work nicely so it is definetly matter for user to set up if he wants to override | 10:12 |
| gfidente | so we need that dir to be labeled correctly, we can't "migrate" users | 10:12 |
| social | yes | 10:12 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Align node_ips hiera keys with the service name. https://review.openstack.org/353637 | 10:12 |
| social | gfidente: another one for openstack-selinux package :( | 10:12 |
| gfidente | social, yeah I was trying to avoid that, but maybe we can walk both paths in parallel | 10:12 |
| openstackgerrit | Steven Hardy proposed openstack/puppet-tripleo: Align hiera keys with service names https://review.openstack.org/353582 | 10:12 |
| social | gfidente: move the exex and selinux_type to THT | 10:15 |
| gfidente | social, huh how? | 10:15 |
| *** bootsha has joined #tripleo | 10:17 | |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validate the instackenv format https://review.openstack.org/353956 | 10:17 |
| jaosorior | shardy: what had merged that ended up in a merge conflict in those commits? | 10:18 |
| shardy | jaosorior: nothing, we made the multinode jobs gating, so it's impossible to merge anything with red CI now | 10:18 |
| jaosorior | I see | 10:19 |
| jaosorior | shardy: should I workflow them again or do I wait for CI? | 10:21 |
| shardy | jaosorior: I guess we wait for CI again | 10:22 |
| shardy | it can't merge without CI passing anyway | 10:22 |
| jaosorior | alright | 10:23 |
| jaosorior | when it passes CI I'll +A it | 10:23 |
| *** bootsha has quit IRC | 10:23 | |
| *** bootsha has joined #tripleo | 10:24 | |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validate the open-files-limit value in mysql https://review.openstack.org/353961 | 10:24 |
| *** mgould|afk is now known as mgould | 10:26 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Move common includes to init.pp https://review.openstack.org/353962 | 10:26 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Include ::tripleo which imports the common manifests https://review.openstack.org/353964 | 10:29 |
| *** bootsha has quit IRC | 10:31 | |
| *** fragatina has joined #tripleo | 10:31 | |
| sshnaidm | does anybody deal with os_vif module problem in case? | 10:33 |
| *** fragatina has quit IRC | 10:33 | |
| *** Goneri has quit IRC | 10:33 | |
| *** mbound has joined #tripleo | 10:33 | |
| *** fragatina has joined #tripleo | 10:34 | |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validate ntp is running and synchronised https://review.openstack.org/353967 | 10:34 |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validate pacemaker status https://review.openstack.org/353969 | 10:38 |
| *** mbound has quit IRC | 10:38 | |
| openstackgerrit | Merged openstack/tripleo-validations: Validation to verify undercloud disk size reqs https://review.openstack.org/351224 | 10:39 |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validate rabbitmq limits https://review.openstack.org/353970 | 10:40 |
| gfidente | social, those policies are executed at the time the package is installed and as such available from the first moment the node is booting right? | 10:41 |
| *** Guest54563 is now known as beagles | 10:41 | |
| *** dbecker has quit IRC | 10:41 | |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Validate rogue DHCP servers https://review.openstack.org/353973 | 10:43 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Add support for cinder v3 endpoint https://review.openstack.org/353364 | 10:43 |
| yolanda | question... i'm hitting an error with a baremetal deployment just on liberty, that complains about external network conflicting with another network | 10:44 |
| yolanda | i've been checking and seems that neutron needs a flag allow_overlapping_ips set to true | 10:44 |
| yolanda | what shall be the best way to set it? i tried on my env file, on parameter defaults, with ExtraConfig: neutron::allow_overlapping_ips: true, but no luck | 10:45 |
| openstackgerrit | Tomas Sedovic proposed openstack/tripleo-validations: Check for differences between introspected HW https://review.openstack.org/353978 | 10:47 |
| *** Goneri has joined #tripleo | 10:47 | |
| jaosorior | shardy: Your patches for the endpoint map already went into merge conflict. I'm waiting for the manila patch to merge to push a rebase of them | 10:49 |
| *** fragatin_ has joined #tripleo | 10:50 | |
| *** xuao has joined #tripleo | 10:52 | |
| *** paramite is now known as paramite|afk | 10:52 | |
| *** athomas has quit IRC | 10:53 | |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Align node_ips hiera keys with the service name. https://review.openstack.org/353637 | 10:53 |
| *** fragatina has quit IRC | 10:54 | |
| *** athomas has joined #tripleo | 11:01 | |
| *** kullmar has joined #tripleo | 11:02 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Composable Mistral services https://review.openstack.org/323436 | 11:03 |
| *** xuao has quit IRC | 11:06 | |
| *** bkopilov has quit IRC | 11:11 | |
| *** kullmar has quit IRC | 11:19 | |
| openstackgerrit | Steven Hardy proposed openstack/puppet-tripleo: Add ironic inspector as a terminated HAProxy endpoint https://review.openstack.org/345183 | 11:19 |
| jaosorior | shardy: thanks for updating the commit message there | 11:20 |
| shardy | jaosorior: np, when it passes CI I'll approve it | 11:20 |
| jaosorior | great | 11:20 |
| *** iogg has joined #tripleo | 11:23 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Add hieradata for Manila v1 and v2 endpoints https://review.openstack.org/352353 | 11:24 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: DO NOT MERGE - Periodic ha test https://review.openstack.org/346949 | 11:25 |
| *** jaosorior has quit IRC | 11:29 | |
| *** jaosorior has joined #tripleo | 11:30 | |
| openstackgerrit | yolanda.robla proposed openstack/diskimage-builder: Allow to parameterize the number of kernels to be kept https://review.openstack.org/353994 | 11:30 |
| openstackgerrit | mathieu bultel proposed openstack/tripleo-common: Do only a strick grep on nova show status https://review.openstack.org/353996 | 11:31 |
| sshnaidm | FYI, https://bugs.launchpad.net/tripleo/+bug/1612178 , handled by amoralej | 11:32 |
| openstack | Launchpad bug 1612178 in tripleo "Nova compute fails because of "No module named os_vif"" [Critical,In progress] | 11:32 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Remove the incorrect warning caused by the boot_option local dict https://review.openstack.org/354001 | 11:33 |
| shardy | sshnaidm: is there a patch fixing the spec file posted we can follow? | 11:34 |
| sshnaidm | shardy, not yet, I'll post it into bug when it will be ready | 11:34 |
| shardy | sshnaidm: Ok, thanks! | 11:34 |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: NO MERGE Test periodic https://review.openstack.org/354002 | 11:35 |
| *** sshnaidm has quit IRC | 11:35 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Convert EndpointMap to not require per-service VIP parameters https://review.openstack.org/342900 | 11:36 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Convert ServiceNetMap to a nested template https://review.openstack.org/353032 | 11:36 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP - Implement overcloud upgrade job - Mitaka -> Newton https://review.openstack.org/323750 | 11:36 |
| shardy | Anyone know why puppet fails to run like this: http://paste.openstack.org/show/554119/ | 11:40 |
| shardy | Error: Could not run: Could not retrieve facts for overcloud-controller-0.openstacklocal | 11:40 |
| shardy | AFAICS the facts are available via facter, but clearly I'm missing something | 11:41 |
| slagle | shardy: probably more hostname shenanigans | 11:41 |
| gfidente | yeah needs a mapping in /etc/hosts | 11:42 |
| slagle | shardy: try export FACTER_fqdn=overcloud-controller-0 before the puppet apply | 11:42 |
| slagle | if that works, you know it's hostname related | 11:42 |
| *** chlong|mtg has joined #tripleo | 11:43 | |
| shardy | slagle, gfidente: Ah, yeah facter fqdn doesn't match hostname -f, something is broken in the hosts file | 11:44 |
| shardy | thanks! | 11:44 |
| gfidente | though it says no address so looks like it's the ip > name mapping missing | 11:45 |
| shardy | yeah the domain is missing from the hosts entry, not sure why, will debug | 11:45 |
| slagle | you also need to check searchdomain in resolv.conf | 11:46 |
| slagle | as puppet might be pulling domain from there | 11:46 |
| shardy | slagle: Yeah that has search openstacklocal, but openstacklocal is missing from the hosts entry | 11:46 |
| shardy | not sure why that happens but I'm testing some local patches so will have to bisect and see what broke it | 11:47 |
| shardy | thanks for the pointers :) | 11:47 |
| *** cmyster has quit IRC | 11:48 | |
| *** dbecker has joined #tripleo | 11:49 | |
| openstackgerrit | Justin Kilpatrick proposed openstack/tripleo-quickstart: Add documentation for add-on roles https://review.openstack.org/346733 | 11:50 |
| *** lucasagomes is now known as lucas-hungry | 11:52 | |
| panda | pabelanger: do you know if there's any spell check non voting upstream test ? It's useful when reviewing documentation | 11:53 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient: Migrate the deploy command to use the Mistral workflows https://review.openstack.org/342261 | 11:53 |
| *** xuao has joined #tripleo | 11:54 | |
| d0ugal | shardy: ^ Close, but still not there. The path munging is messy. I added a comment about the bit missing if you want to take a look. I gotta run now (see your emails if you've not yet) | 11:56 |
| *** xuao has quit IRC | 11:56 | |
| *** xuao has joined #tripleo | 11:56 | |
| *** milan has quit IRC | 11:56 | |
| amoralej | shardy, sshnaidm, i need to create a new package so first thing to do is to send a package review bz, then add the new package to rdo and finally add the dependency to nova spec | 11:58 |
| shardy | d0ugal: ack, thanks! | 11:58 |
| *** alee has quit IRC | 11:58 | |
| *** rhallisey has joined #tripleo | 11:59 | |
| *** jpena is now known as jpena|lunch | 12:00 | |
| *** milan has joined #tripleo | 12:03 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: Update Ironic composable services https://review.openstack.org/341433 | 12:11 |
| *** weshay has joined #tripleo | 12:12 | |
| *** pradk has joined #tripleo | 12:16 | |
| openstackgerrit | Marios Andreou proposed openstack/puppet-tripleo: Add manila-netapp backend to manila class https://review.openstack.org/354014 | 12:16 |
| *** xuao has quit IRC | 12:16 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-docs: [WIP] Documentation for installing and using Ironic in overcloud https://review.openstack.org/354016 | 12:17 |
| *** xuao has joined #tripleo | 12:18 | |
| *** sshnaidm has joined #tripleo | 12:20 | |
| *** limao_ has quit IRC | 12:20 | |
| *** xuao has quit IRC | 12:20 | |
| *** xuao has joined #tripleo | 12:21 | |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Add integration with NetApp Manila driver https://review.openstack.org/354019 | 12:21 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Composable Mistral services https://review.openstack.org/323436 | 12:21 |
| marios | rhefner: bswartz tbarron ^^^ i poposed a new review, bswartz was right in the end there wasn't much left in terms of files. there was a bit of copy/paste with the params, but it wasn't worth the rebase pain. I've proposed a new review https://review.openstack.org/354019 and the puppet-tripleo side in the depends on. | 12:22 |
| *** jayg|g0n3 is now known as jayg | 12:23 | |
| marios | rhefner: are you ok with abandoning https://review.openstack.org/#/c/188138/40 in this case? i added you as co-author in the new one | 12:23 |
| marios | tbarron: hope that helps as a start regardless of how we proceed... i also want to revisit the manila generic puppet-tripleo side tomorrow since that needs a tidy up - we are duplicating config unnecessarily between tripleo-heat-templates and puppet-tripleo. but we don't need to block manila netap because of that. | 12:25 |
| *** iogg has quit IRC | 12:26 | |
| *** paramite|afk is now known as paramite | 12:26 | |
| openstackgerrit | Merged openstack/tripleo-quickstart: remove hard setting of ssh.config.ansible in roles-deploy https://review.openstack.org/352990 | 12:26 |
| *** dprince has joined #tripleo | 12:27 | |
| *** xuao has quit IRC | 12:27 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Do not merge, checking composable mistral. https://review.openstack.org/352947 | 12:27 |
| *** chem has quit IRC | 12:28 | |
| *** radeks has quit IRC | 12:29 | |
| *** chem has joined #tripleo | 12:29 | |
| openstackgerrit | Merged openstack/os-net-config: Enable os-net-config to support and configure NFVSwitch https://review.openstack.org/345599 | 12:31 |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Add integration with NetApp Manila driver https://review.openstack.org/354019 | 12:32 |
| openstackgerrit | Merged openstack/puppet-tripleo: Fix comparison to control_virtual_ip https://review.openstack.org/305928 | 12:33 |
| *** jeckersb_gone is now known as jeckersb | 12:34 | |
| *** amoralej is now known as amoralej|lunch | 12:35 | |
| *** ohamada has quit IRC | 12:35 | |
| *** ohamada has joined #tripleo | 12:36 | |
| *** iogg has joined #tripleo | 12:37 | |
| rhefner | marios: yep, I'll abandon the old change now | 12:38 |
| *** [1]cdearborn has joined #tripleo | 12:39 | |
| *** jdob has joined #tripleo | 12:39 | |
| marios | rhefner: appreciate thanks | 12:39 |
| rhefner | marios: thank you for working on this! | 12:40 |
| tbarron | marios: thanks, we'll take a look! | 12:40 |
| honza | for tripleo-heat-templates, which environment files are commonly passed to "stack create" or is there one that imports everything? | 12:41 |
| honza | ah, the root environment file i guess :) | 12:47 |
| marios | rhefner: tbarron ack thanks very much, please add yourselves as reviewers if you like I will revisit tomorrow for updates discussion etc. esp Q @ https://review.openstack.org/#/c/354019/2 | 12:47 |
| *** rlandy has joined #tripleo | 12:47 | |
| jaosorior | pabelanger: Hey, I saw your commit about adding bindep to python-tripleo. Can you explain how that works? We might need that for a barbican job | 12:47 |
| *** Ryjedo_ has joined #tripleo | 12:48 | |
| *** egafford has joined #tripleo | 12:48 | |
| *** Ryjedo has quit IRC | 12:48 | |
| *** Ryjedo_ is now known as Ryjedo | 12:48 | |
| tbarron | marios: sure thinng and thanks again | 12:48 |
| shardy | honza: overcloud-resource-registry-puppet.yaml is always passed, most other environment files are optional | 12:54 |
| honza | shardy: thanks | 12:55 |
| *** lucas-hungry is now known as lucasagomes | 12:55 | |
| marios | tbarron: np, in the end wasn't as painful as expected... imo it shows the composable services is working very well. on the tht side, we have the env file for the netapp params, and then all the changes are in the manila-base file | 12:56 |
| *** fultonj has joined #tripleo | 12:58 | |
| tbarron | marios: yeah, i'm thinking we want to move the ManilaGeneric parameters out of manila-base to a manila-generic env | 12:58 |
| *** amoralej|lunch is now known as amoralej | 12:58 | |
| tbarron | marios: related to your Q in the review | 12:58 |
| marios | tbarron: ack thanks will look and fixup but context switch for now | 12:59 |
| tbarron | marios: but we need to hash this out with bswartz (NetApp) and Jokke_ (cephfsnative plugin) | 12:59 |
| marios | tbarron: right OK, so give folks a chance to see it too | 12:59 |
| tbarron | marios: yeah, we need to talk through the approach withh everyone first anyways | 12:59 |
| marios | tbarron: ok, today was just about getting a first pass out so we have something to actually point at. | 13:00 |
| tbarron | marios: yep, perfect! | 13:01 |
| *** lblanchard has joined #tripleo | 13:01 | |
| *** jpena|lunch is now known as jpena | 13:02 | |
| *** limao has joined #tripleo | 13:04 | |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: Removing WARNING: line has more than 140 characters in puppet-tripleo profiles https://review.openstack.org/354054 | 13:06 |
| *** morazi has joined #tripleo | 13:10 | |
| *** rodrigods has quit IRC | 13:10 | |
| *** rodrigods has joined #tripleo | 13:10 | |
| *** bootsha has joined #tripleo | 13:11 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: Update Ironic composable services https://review.openstack.org/341433 | 13:11 |
| *** cdearborn has joined #tripleo | 13:14 | |
| weshay | sshnaidm, panda can one of you guys put in a review to update the tripleo periodic jobs to run every four hours vs. nightly | 13:15 |
| rook | EmilienM: can we get this pulled in? or get some comments on what needs to change? https://review.openstack.org/#/c/350219/ | 13:16 |
| yolanda | hi... so i'm doing a baremetal deployment on liberty. I see that the setting allow_overlapping_ips is hardcoded to false on neutron.conf, while on mitaka and master is just commented. Is there any way i can customize that with tripleo on liberty, to be set as true? | 13:16 |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-docs: [WIP] Documentation for installing and using Ironic in overcloud https://review.openstack.org/354016 | 13:17 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: Move Nova settings out of puppet/compute.yaml https://review.openstack.org/353748 | 13:17 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: Mv Ceilometer settings out of puppet/compute.yaml https://review.openstack.org/354063 | 13:17 |
| *** [1]cdearborn has quit IRC | 13:18 | |
| *** beagles is now known as beagles_brb | 13:18 | |
| jaosorior | yolanda: I think you can do it by setting the appropriate hieradata in the controllerExtraConfig parameter, which you can set by creating an environment file and adding that to parameter_Defaults | 13:20 |
| jaosorior | *parameter_defaults | 13:20 |
| jaosorior | yolanda: however, you would need to investigate if that parameter setting is set, first, somewhere in tripleo-heat-templates; second, somewhere in puppet-neutron (both in the liberty branches) | 13:21 |
| yolanda | jaosorior, i tried something like | 13:22 |
| yolanda | ExtraConfig: | 13:22 |
| yolanda | neutron::allow_overlapping_ips: true | 13:22 |
| yolanda | but it doesn't seem to be applied | 13:22 |
| *** pgadiya has quit IRC | 13:22 | |
| *** julim has joined #tripleo | 13:22 | |
| *** jcoufal has joined #tripleo | 13:23 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/instack-undercloud: Ironic: Remove dnsmasq-ironic.conf https://review.openstack.org/354068 | 13:23 |
| jaosorior | yolanda: can you check if there's an extraconfig.yaml in /etc/puppet/hieradata ? | 13:25 |
| weshay | panda, sshnaidm you there? | 13:25 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP - Implement overcloud upgrade job - Mitaka -> Newton https://review.openstack.org/323750 | 13:25 |
| jaosorior | shardy: is ExtraConfig supposed to still work in master? | 13:25 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Reinstate CloudDomain properties for Controller/Compute roles https://review.openstack.org/354069 | 13:25 |
| shardy | jaosorior: yes | 13:25 |
| panda | weshay: how many hours ? | 13:25 |
| *** tzumainn has joined #tripleo | 13:25 | |
| weshay | panda, 4 | 13:25 |
| jaosorior | shardy: do we pass that through parameter_defaults nowadays? | 13:25 |
| yolanda | jaosorior, in tripleo-heat-templates, i see puppet/hieradata/common.yaml, with neutorn::allow_overlapping_ips: true | 13:25 |
| shardy | slagle, gfidente: ^^ FYI that fixes the hostname issue I was seeing | 13:26 |
| panda | weshay: oh, I read few, not four | 13:26 |
| shardy | turns out it was a custom-roles refactoring patch that broke the hosts file (without breaking CI..) | 13:26 |
| *** bootsha has quit IRC | 13:26 | |
| shardy | jaosorior: yup, everything should be passed via parameter_defaults | 13:26 |
| *** rbrady has joined #tripleo | 13:27 | |
| *** beagles_brb is now known as beagles | 13:27 | |
| shardy | yolanda: I'd double check you don't have ExtraConfig defined in more than one environment file | 13:27 |
| shardy | and grep in /etc/puppet/hieradata on the node for the key you're setting | 13:27 |
| jaosorior | shardy: alright, just making sure. Cause it's been removed from the parameters that overcloud.yaml passes from the controller and compute, but not the rest of the node types | 13:27 |
| yolanda | i don't have that setting on hieradat, looking more in depth | 13:27 |
| shardy | jaosorior: yup, by using parameter_defaults it should still be applied | 13:27 |
| *** alee has joined #tripleo | 13:28 | |
| *** jistr has quit IRC | 13:28 | |
| dtantsur | folks, has anyone seen deployment step 3 failing because of missing sql_connection option in nova.conf? | 13:28 |
| dtantsur | or maybe at least has an idea why it could happen? | 13:28 |
| jaosorior | dtantsur: got a link to the logs? | 13:29 |
| dtantsur | nope, this is locally | 13:29 |
| dtantsur | I can dump the whole "stack failures list" if that helps | 13:30 |
| dtantsur | jaosorior, ^^ | 13:30 |
| *** jistr has joined #tripleo | 13:30 | |
| jaosorior | dtantsur: haven't seen that, but it seems to me that it might be an ordering issue.. | 13:31 |
| dtantsur | do we have any places where ordering matters? | 13:31 |
| jaosorior | dtantsur: was it the db-sync failing? | 13:32 |
| dtantsur | jaosorior, yep. I'm testing https://review.openstack.org/#/c/341433/ locally | 13:32 |
| jaosorior | dtantsur: Let me check puppet-nova | 13:32 |
| dtantsur | before I introduced the environment file, everything was fine :( maybe I've messed some ordering when doing it | 13:32 |
| jaosorior | well... that's weird, nova's db-sync is supposed to be anchored | 13:33 |
| jaosorior | dtantsur: gotta go to a meeting :( | 13:33 |
| dtantsur | yeah.. so I'm getting a very partial nova.conf for some reason | 13:33 |
| *** ohamada has quit IRC | 13:33 | |
| dtantsur | jaosorior, np, thanks | 13:33 |
| *** ohamada has joined #tripleo | 13:33 | |
| jaosorior | dtantsur: but which db-sync was, nova's db-sync or ironic's? | 13:34 |
| lucasagomes | hi all, any chance someone can take a look at https://review.openstack.org/#/c/350627/ ? It has a +2 already :-) | 13:34 |
| dtantsur | jaosorior, /Stage[main]/Nova::Db::Sync/Exec[nova-db-sync] and /Stage[main]/Nova::Db::Sync_api/Exec[nova-db-sync-api] both failed | 13:35 |
| panda | weshay: this may probe difficult, there is a single periodic definition for all the projects in zuul layout. It may mean that periodicities below 1 day are not encouraged. | 13:35 |
| panda | prove* | 13:35 |
| weshay | of course | 13:35 |
| weshay | pabelanger, is that the case re: periodic ^ | 13:35 |
| jaosorior | dtantsur: :( I see... well, I think it's an ordering issue. and it's weird cause there is an anchor set for both of those https://github.com/openstack/puppet-nova/blob/master/manifests/db/sync.pp#L23 | 13:36 |
| jaosorior | anyway, I do have a meeting | 13:36 |
| *** jaosorior is now known as jaosorior_mtg | 13:36 | |
| weshay | panda, go see if there is another trigger defined | 13:36 |
| panda | weshay: I see only periodic and periodic-stable, both once a day | 13:36 |
| weshay | panda, meh.. ok go put in a new definition and see what kind of reviews we get | 13:37 |
| panda | weshay: ok | 13:37 |
| dtantsur | ok, I'll try refreshing the environment, maybe old images or something like that | 13:38 |
| panda | weshay: but it's better to explain exactly why we want to do this, if we'll have to defend the change. | 13:39 |
| bandini | gfidente: do we need to recheck https://review.openstack.org/#/c/254788/ due to the POST_FAILURE? | 13:39 |
| gfidente | bandini, the patch testing that passed but ... | 13:39 |
| yolanda | jaosorior_mtg, shardy , i found that only on one place, on /home/stack/network-environment.yaml. Maybe is not a good place to define that? | 13:40 |
| bandini | gfidente: does it mean I should recheck? | 13:40 |
| * bandini slightly confused | 13:41 | |
| gfidente | bandini, yes | 13:42 |
| bandini | gfidente: thanks! | 13:42 |
| shardy | yolanda: provided it's not malformed and you're including /home/stack/network-environment.yaml that should work - if you'd like it checked can you paste it somewhere? | 13:43 |
| shardy | yolanda: you can also run the overcloud deploy with --debug and tee the results to a file, then check the ExtraConfig you expect is included in the call to heat | 13:43 |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: Update Ironic composable services https://review.openstack.org/341433 | 13:43 |
| yolanda | shardy, here it is http://paste.openstack.org/show/554145/ | 13:44 |
| yolanda | this file is being picked because it gets the network config from there | 13:45 |
| yolanda | shardy, i wonder if i'm missing concepts again... because actually, that value is on neutron.conf of the undercloud | 13:46 |
| shardy | yolanda: Hmm, so you want that to be set for the undercloud neutron? | 13:46 |
| shardy | ExtraConfig will only influence the overcloud services | 13:46 |
| yolanda | that was my suspect... so in undercloud for mitaka/master, that value is just commented | 13:47 |
| yolanda | in liberty, is set to false, then network step of overcloud fails | 13:47 |
| yolanda | where is that value set for the undercloud, you know? | 13:47 |
| shardy | yolanda: for recent undercloud versions you can pass a hieradata_override = option in the undercloud.conf | 13:48 |
| shardy | I'm not sure that got backported to mitaka tho | 13:48 |
| yolanda | it is liberty so even worse | 13:48 |
| yolanda | no hieradata_override for instack-undercloud on liberty | 13:50 |
| *** ayoung has joined #tripleo | 13:51 | |
| *** myoung|afk is now known as myoung | 13:52 | |
| *** limao has quit IRC | 13:54 | |
| sshnaidm | weshay, hi | 13:55 |
| shardy | EmilienM: Hey, testing your hiera/json patch, and I think there's a problem when a file doesn't exist that's defined in the hierarchy | 13:55 |
| shardy | http://paste.openstack.org/show/554148/ | 13:55 |
| shardy | the deployment gets stuck in a loop doing that at step3 | 13:55 |
| *** fultonj has quit IRC | 13:56 | |
| yolanda | shardy, maybe a way can be to have a undercloud-install-post.sh script that updates that setting on neutron? | 13:56 |
| weshay | sshnaidm, can you work w/ EmilienM so that we can add the puppet CI to the tripleo periodic job set that votes | 13:56 |
| *** fultonj has joined #tripleo | 13:56 | |
| shardy | yolanda: you could manually configure it every time the undercloud install is run as a workaround | 13:56 |
| EmilienM | shardy: will look in a few, in a mtg now | 13:57 |
| yolanda | yes, something like that, oooq allows to have that custom post template, that could make the trick, although it looks a bit dirty | 13:57 |
| shardy | EmilienM: np, thanks - it may be a json/yaml issue | 13:57 |
| shardy | as it's looking for the yaml files as foo.json | 13:57 |
| sshnaidm | weshay, EmilienM, sure, which puppet CI job are we talking? | 13:58 |
| slagle | deep dive starting, https://bluejeans.com/176756457/ | 13:58 |
| *** maeca1 has joined #tripleo | 13:59 | |
| *** limao has joined #tripleo | 13:59 | |
| *** leanderthal|afk is now known as leanderthal | 13:59 | |
| hewbrocca | slagle: I hope the pool for the deep dive isn't green... | 14:00 |
| *** limao_ has joined #tripleo | 14:00 | |
| *** limao_ has quit IRC | 14:01 | |
| *** limao_ has joined #tripleo | 14:01 | |
| EmilienM | shardy: I don't see how it's related to my patch in puppet-openstacklib, it must be something in tht, I can look | 14:01 |
| shardy | EmilienM: Ok, np I've got the related t-h-t patch applied so perhaps that's somehow broken | 14:02 |
| shardy | I don't see how either, will dig deeper | 14:02 |
| mandre | EmilienM: just to confirm I'm on the right track for https://review.openstack.org/#/c/344140/ | 14:02 |
| EmilienM | shardy: my patch in puppet translate a string json into the right puppet type | 14:02 |
| mandre | EmilienM: I'll need a couple of patches like this one https://review.openstack.org/#/c/354087/, correct? | 14:03 |
| shardy | EmilienM: Hmm, running it again perhaps those warnings are spurious | 14:03 |
| slagle | dprince: you dropped | 14:03 |
| shardy | sorry, it may be getting stuck somewhere else | 14:03 |
| shardy | will try again, sorry for the noise | 14:03 |
| *** limao has quit IRC | 14:03 | |
| *** dprince_ has joined #tripleo | 14:03 | |
| *** limao has joined #tripleo | 14:04 | |
| *** dprince has quit IRC | 14:04 | |
| EmilienM | mandre: what is the context? | 14:05 |
| mandre | EmilienM: you asked me to submit patches to puppet-* to configure cors settings | 14:06 |
| EmilienM | ah maybe | 14:06 |
| mandre | EmilienM: https://review.openstack.org/#/c/344140/2/elements/puppet-stack-config/puppet-stack-config.pp | 14:06 |
| EmilienM | mandre: good | 14:06 |
| *** limao has quit IRC | 14:06 | |
| EmilienM | mandre: patch needs tests & release note | 14:06 |
| *** limao_ has quit IRC | 14:07 | |
| mandre | EmilienM: so, yeah, I'm a puppet noob so I just wanted to confirm that https://review.openstack.org/#/c/354087/ was the way to go | 14:07 |
| EmilienM | mandre: it is | 14:07 |
| mandre | EmilienM: excellent! I'll add specs and release notes | 14:08 |
| openstackgerrit | Saravanan KR proposed openstack/os-net-config: Add support for OVS DPDK Bridge and Port https://review.openstack.org/338694 | 14:08 |
| panda | weshay: https://review.openstack.org/354098 | 14:11 |
| *** [1]cdearborn has joined #tripleo | 14:11 | |
| *** trozet has joined #tripleo | 14:13 | |
| *** panda is now known as panda|scuba | 14:13 | |
| *** masco has quit IRC | 14:13 | |
| trozet | EmilienM: thanks for reviewing 335173, can you tell me more about what this "HA light" arch is? | 14:15 |
| EmilienM | trozet: https://specs.openstack.org/openstack/tripleo-specs/specs/newton/pacemaker-next-generation-architecture.html | 14:16 |
| *** jistr is now known as jistr|call | 14:16 | |
| *** oshvartz has quit IRC | 14:16 | |
| trozet | EmilienM: thanks | 14:17 |
| *** ebalduf has joined #tripleo | 14:18 | |
| openstackgerrit | Saravanan KR proposed openstack/os-net-config: Add support for OVS DPDK Bridge and Port https://review.openstack.org/338694 | 14:20 |
| *** cdearborn has quit IRC | 14:22 | |
| *** beekneemech is now known as bnemec | 14:23 | |
| openstackgerrit | Saravanan KR proposed openstack/os-net-config: Add support for OVS DPDK Bond https://review.openstack.org/349183 | 14:23 |
| EmilienM | shardy: re - t-i-e, do you want me to branch it and release it? | 14:24 |
| *** jprovazn has quit IRC | 14:26 | |
| *** limao has joined #tripleo | 14:29 | |
| *** chlong|mtg has quit IRC | 14:33 | |
| shardy | EmilienM: I think probably not now, lets decide if it should be branched when we branch everything for Newton? | 14:33 |
| shardy | thanks for offering though | 14:33 |
| EmilienM | shardy: ack | 14:33 |
| *** Vijayendra has quit IRC | 14:35 | |
| *** Vijayendra has joined #tripleo | 14:35 | |
| shardy | EmilienM: so, testing further, it seems that the interpolated value looks correct (json string), but the value which we put into the service templates is not | 14:37 |
| shardy | e.g passing a list into keystone::rabbit_hosts gives rabbit_hosts = 192.0.2.10 | 14:37 |
| trozet | EmilienM: ok I like this new arch. So as long as the service is under overcloud resource-registry, it should just work with HA? | 14:37 |
| shardy | but passing a json string gives rabbit_hosts = ["192.0.2.10"] | 14:37 |
| shardy | so it would seem that the json->list conversion isn't working? | 14:38 |
| shardy | EmilienM: any tips on how to debug the code you added? | 14:38 |
| *** jtomasek_ has quit IRC | 14:40 | |
| EmilienM | shardy: what happens if you don't use my code? | 14:40 |
| *** limao_ has joined #tripleo | 14:42 | |
| shardy | EmilienM: Not tried that yet as it's a combination I didn't expect to work, sec let me try | 14:42 |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: Removing WARNING: line has more than 140 characters in puppet-tripleo profiles https://review.openstack.org/354054 | 14:43 |
| EmilienM | shardy: I also found that my patch might be not finished | 14:44 |
| EmilienM | shardy: Ineed to patch another provider | 14:44 |
| EmilienM | let me iterate on it | 14:44 |
| *** limao has quit IRC | 14:45 | |
| *** chlong|mtg has joined #tripleo | 14:45 | |
| shardy | EmilienM: ack, thanks! I can re-test when you're ready | 14:45 |
| *** numan_ has quit IRC | 14:45 | |
| EmilienM | shardy: please test without my patch and see the result if it's different | 14:46 |
| *** apetrich has quit IRC | 14:47 | |
| *** limao_ has quit IRC | 14:49 | |
| *** limao has joined #tripleo | 14:49 | |
| *** skramaja has quit IRC | 14:49 | |
| *** bana_k has joined #tripleo | 14:50 | |
| EmilienM | shardy: I updated the patch | 14:52 |
| marios | dprince_: thanks dan well done fwiw | 14:52 |
| shardy | thanks dprince_! | 14:53 |
| dprince_ | ack, thanks guys | 14:53 |
| EmilienM | dprince_: ++ | 14:53 |
| EmilienM | dprince_: we want to see the switch now | 14:53 |
| dprince_ | EmilienM: meh, it isn't that exciting I think | 14:54 |
| *** limao has quit IRC | 14:56 | |
| *** limao has joined #tripleo | 14:56 | |
| *** rajinir has joined #tripleo | 14:58 | |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-quickstart: files on undercloud had wrong uid https://review.openstack.org/352398 | 14:59 |
| openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-quickstart: fix teardown (tagging and ansible_user) https://review.openstack.org/352112 | 14:59 |
| jaosorior_mtg | larsks: in quickstart is it possible to only run a specific task depending on what release we're deploying | 15:00 |
| larsks | jaosorior_mtg: you could add specific conditionals to a task to do that, yes. | 15:01 |
| jaosorior_mtg | larsks is there an example somewhere? | 15:01 |
| larsks | jaosorior_mtg: not on a task right now, but in some of the templated shell scripts. On a task you would add something like "when: release in ['mitaka', 'liberty']" | 15:02 |
| jaosorior_mtg | larsks: thanks | 15:02 |
| jrist | dprince_: thanks for the dd, can you add this to the etherpad? https://invis.io/UM87J4NBQ#/179046668_2016-8-1_TripleO_UI37 | 15:07 |
| *** limao has quit IRC | 15:07 | |
| dprince_ | jrist: sure | 15:08 |
| jrist | thanks | 15:09 |
| *** limao has joined #tripleo | 15:09 | |
| EmilienM | I like how UI folks always have fancy tools :-) | 15:09 |
| EmilienM | jrist: I'm jealous ! | 15:09 |
| *** Guest20032 is now known as saneax_AFK | 15:09 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-quickstart: Persist ironic configuration changes https://review.openstack.org/351683 | 15:10 |
| *** panda|scuba is now known as panda | 15:12 | |
| *** penick has quit IRC | 15:12 | |
| *** aufi has quit IRC | 15:14 | |
| jrist | EmilienM: hahaha | 15:17 |
| jrist | EmilienM: like what? Invision? | 15:17 |
| *** jaosorior_mtg is now known as jaosorior | 15:17 | |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: Move Nova settings out of puppet/compute.yaml https://review.openstack.org/353748 | 15:17 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: Mv Ceilometer settings out of puppet/compute.yaml https://review.openstack.org/354063 | 15:17 |
| *** ebarrera has quit IRC | 15:18 | |
| *** limao has quit IRC | 15:19 | |
| EmilienM | jrist: yeah | 15:19 |
| jaosorior | EmilienM: couple of simple commits :D https://review.openstack.org/#/c/353962/ https://review.openstack.org/#/c/353964/ | 15:19 |
| EmilienM | jaosorior: ack | 15:19 |
| *** limao has joined #tripleo | 15:19 | |
| jrist | EmilienM: command line is so well done at this point | 15:19 |
| jrist | what with emacs and all | 15:19 |
| jrist | :-P | 15:20 |
| * jrist starts holy war | 15:20 | |
| jrist | tmux and vim | 15:20 |
| jrist | what else do you need? :) | 15:20 |
| dprince_ | jrist: nothing | 15:20 |
| *** limao has quit IRC | 15:20 | |
| dtantsur | so folks, I guess https://review.openstack.org/#/c/341433/14/puppet/services/haproxy.yaml is not enough for haproxy to start working with ironic, is it? | 15:21 |
| dtantsur | do you know what I can be missing? | 15:21 |
| *** ifarkas is now known as ifarkas_afk | 15:21 | |
| *** jistr|call is now known as jistr | 15:21 | |
| *** tesseract- has quit IRC | 15:23 | |
| shardy | EmilienM: still not working - does the PSON::load work by reference, or should it be new_value = PSON::load(new_value) ? | 15:23 |
| shardy | seems like it's just passing the json string into the service config | 15:23 |
| EmilienM | shardy: so you tried without my patch and you got same results? | 15:23 |
| shardy | EmilienM: yup | 15:24 |
| EmilienM | yeah it's possible | 15:24 |
| shardy | EmilienM: I'll try adding the assignment | 15:24 |
| EmilienM | mhh | 15:24 |
| *** mcornea has quit IRC | 15:25 | |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: Removing WARNING: line has more than 140 characters in puppet-tripleo profiles https://review.openstack.org/354054 | 15:25 |
| jaosorior | EmilienM: What do you mean they won't exist anymore? We still need to have common stuff for all the nodes | 15:25 |
| jaosorior | what's gonna replce that? | 15:26 |
| * shardy wonders if a unit test may be a quicker way to debug this | 15:26 | |
| bswartz | jrist: why does everyone love tmux now? | 15:26 |
| EmilienM | jaosorior: the overcloud_*.pp files won't exist anymore | 15:26 |
| dtantsur | hmm, openstack-ironic-api fails with "address in use", so I guess I need to somehow tell it to move | 15:26 |
| EmilienM | shardy: I'll look in a few, I'm currently doing a CI thing for puppet/tripleo | 15:27 |
| EmilienM | shardy: I figured we don't run tripleo jobs in puppet stable/mitaka CI | 15:27 |
| shardy | EmilienM: Ok, no problem, thanks | 15:27 |
| EmilienM | I'm on it | 15:27 |
| *** fultonj has quit IRC | 15:27 | |
| jaosorior | EmilienM: sure, so what will replace that? will we have a common role or something of the sort? | 15:28 |
| jaosorior | would be better to give an alternative than just say "I won't support it" | 15:28 |
| *** fultonj has joined #tripleo | 15:28 | |
| sshnaidm | EmilienM, hi, regarding puppet CI job as part of promotion jobs, are you familiar with this idea? just checking to be on the same page | 15:28 |
| EmilienM | I'm shutting down IRC, can't focus on my CI problem | 15:29 |
| openstackgerrit | Merged openstack/puppet-tripleo: Handle redirects for keystone https://review.openstack.org/352956 | 15:29 |
| *** zaneb has quit IRC | 15:30 | |
| rook | hey shardy - is there somewhere I can track for custom roles? | 15:35 |
| jrist | bswartz: no idea haha | 15:36 |
| *** bana_k has quit IRC | 15:37 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-heat-templates: Update Ironic composable services https://review.openstack.org/341433 | 15:39 |
| openstackgerrit | mathieu bultel proposed openstack-infra/tripleo-ci: WIP - Implement overcloud upgrade job - Mitaka -> Newton https://review.openstack.org/323750 | 15:39 |
| *** jcoufal has quit IRC | 15:40 | |
| shardy | jaosorior: Common stuff will have to be defined either via composable services (which can be deployed on all roles), or via the allNodesConfig for cluster-wide stuff (which I'm currently working on refactoring) | 15:42 |
| shardy | there will be no per-role manifests tho | 15:42 |
| jaosorior | shardy: sure, that's no biggie. I'm just wondering where are ::tripleo::packages and ::tripleo::firewall gonna be included from? | 15:43 |
| *** links has quit IRC | 15:43 | |
| shardy | jaosorior: I think they'll need to move to a services template | 15:43 |
| jaosorior | shardy: by the way... the keystone endpoints via puppet are being quite problematic. I'm running into a lot of OOM crashes there. | 15:43 |
| jaosorior | either we bump up the memory for the nodes and hope it works there. Or I think we're gonna end up doing those in mistral | 15:44 |
| shardy | jaosorior: oh dear, that doesn't sound good | 15:44 |
| shardy | why would puppet creating the endpoints use much more ram? | 15:44 |
| shardy | I assumed it probably just shells out to openstackclient (maybe openstackclient is the reason?) | 15:45 |
| jaosorior | shardy: I have no ide dude | 15:45 |
| jaosorior | but it does use openstackclient in the background | 15:45 |
| jaosorior | might be | 15:45 |
| jaosorior | shardy: could also be that puppet does things quite concurrently. So doing several of those endpoints at the same time could be problematic due to the memory usage? | 15:46 |
| shardy | jaosorior: possibly, need to profile the memory usage while it's running really I think | 15:46 |
| *** adarazs is now known as adarazs_brb | 15:47 | |
| openstackgerrit | Julie Pichon proposed openstack/tripleo-common: [WIP] Add Mistral actions for boot & root device configuration https://review.openstack.org/354174 | 15:47 |
| openstackgerrit | Paul Belanger proposed openstack/python-tripleoclient: Initial support for bindep https://review.openstack.org/353929 | 15:47 |
| jaosorior | shardy: somehow I don't get OOMs when running that locally though | 15:47 |
| EmilienM | jaosorior: we need to find a mechanism that include ::tripleo class all the time | 15:50 |
| jaosorior | EmilienM: so, for now I think we can have that as a base.yaml or common.yaml | 15:50 |
| jaosorior | in the form of a composable service | 15:50 |
| jaosorior | but we do need a way to make sure that all the node types always include it | 15:51 |
| EmilienM | jaosorior: yeah, I just don't want our users to forget addin git | 15:51 |
| EmilienM | right ! | 15:51 |
| *** milan has quit IRC | 15:51 | |
| jaosorior | else we're gonna shoot ourselves in the foot and we're gonna have really funky bug reports | 15:51 |
| EmilienM | sshnaidm: yes i'm familiar, we need to run puppet-openstack-integration jobs on centos7 | 15:51 |
| *** adarazs_brb is now known as adarazs | 15:52 | |
| EmilienM | jaosorior: ok, let me think about it | 15:52 |
| jaosorior | EmilienM: that's the reason why I was really stubbornly still using those templates | 15:52 |
| jaosorior | I mean | 15:52 |
| jaosorior | manifests | 15:52 |
| EmilienM | yeah but they will be gone very very soon | 15:53 |
| EmilienM | we don't need it anymore | 15:53 |
| EmilienM | jaosorior: maybe in short term we can include ::tripleo in all base profiles in puppet-tripleo? | 15:54 |
| mandre | EmilienM: regarding https://review.openstack.org/#/c/354087/, the spec was failing in my environment, I'm not too sure why | 15:57 |
| *** rhallisey has quit IRC | 15:57 | |
| mandre | EmilienM: you might want to hold on before approving too quickly | 15:57 |
| mandre | EmilienM: how long before the test job returns in puppet-keystone usually? | 15:58 |
| EmilienM | mandre: ~30 min | 15:58 |
| *** rhallisey has joined #tripleo | 15:58 | |
| EmilienM | it depends if osinfra is busy | 15:59 |
| EmilienM | http://status.openstack.org/zuul/ | 15:59 |
| EmilienM | 250 jobs in queue, its busy | 15:59 |
| EmilienM | so maybe 45 min | 15:59 |
| *** jprovazn has joined #tripleo | 16:00 | |
| mandre | EmilienM: here is the exact failure in my environment, http://paste.openstack.org/show/554219/ | 16:01 |
| mandre | EmilienM: I don't understand enough to debug why cors/allowed_origin defaults to nil | 16:01 |
| *** bnemec has quit IRC | 16:04 | |
| *** rwsu has joined #tripleo | 16:09 | |
| *** bnemec has joined #tripleo | 16:12 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/tripleo-docs: [WIP] Documentation for installing and using Ironic in overcloud https://review.openstack.org/354016 | 16:15 |
| *** rcernin has quit IRC | 16:15 | |
| jaosorior | bnemec: Hey dude, what do we do with your commit that bumps the memory for the ovb nodes? https://review.openstack.org/#/c/353655/ | 16:16 |
| bnemec | jaosorior: Really all we need to do is merge it. I've already applied the change to rh1. | 16:16 |
| jaosorior | shardy: could we merge this? https://review.openstack.org/#/c/353655/ it bumps up the memory for the ovb nodes, would surely be useful | 16:17 |
| jaosorior | or slagle, dprince_ ^^ | 16:17 |
| *** dtantsur is now known as dtantsur|afk | 16:18 | |
| *** iogg has quit IRC | 16:18 | |
| dprince_ | jaosorior: +A | 16:20 |
| jaosorior | awww yeah | 16:20 |
| *** zeroshft has joined #tripleo | 16:21 | |
| slagle | jaosorior: they are already bumped up | 16:21 |
| slagle | jaosorior: bnemec did it manually yesterday | 16:21 |
| slagle | if we could find out how much memory we actually need... | 16:22 |
| jaosorior | oh | 16:22 |
| *** egafford has quit IRC | 16:22 | |
| slagle | if creating endpoints via puppet requires another 512mb, then i don't think we want to do it | 16:23 |
| jaosorior | slagle: well... I use 6144 for the overcloud nodes :/ | 16:23 |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: Move neutron settings out of puppet/compute.yaml https://review.openstack.org/354201 | 16:23 |
| jaosorior | slagle: I'll see if it can pass with the current upgrade. and if it doesn't then we gotta do it in mistral then | 16:23 |
| *** apetrich has joined #tripleo | 16:25 | |
| jaosorior | anyway | 16:25 |
| jaosorior | I'm off. Have a good day guys! | 16:25 |
| *** dbecker has quit IRC | 16:25 | |
| *** jaosorior has quit IRC | 16:25 | |
| slagle | ttyl | 16:25 |
| *** hewbrocca is now known as hewbrocca-afk | 16:26 | |
| *** yamahata has joined #tripleo | 16:27 | |
| *** ebarrera has joined #tripleo | 16:29 | |
| bnemec | DB on rh1 crashed. :-( | 16:30 |
| bnemec | Things are going to be bad on the CI front for a while. | 16:30 |
| gfidente | EmilienM, this is only adding files https://review.openstack.org/#/c/311218/ | 16:32 |
| gfidente | there is no way it can ever make CI fail and in fact | 16:32 |
| gfidente | https://review.openstack.org/#/c/304563/ which depends on it passed, can we merge them all? | 16:32 |
| EmilienM | gfidente: yes | 16:32 |
| EmilienM | gfidente: done | 16:33 |
| gfidente | thanks :0 | 16:33 |
| gfidente | ^_^ | 16:33 |
| slagle | bnemec: looks like you restarted mariadb? | 16:35 |
| bnemec | slagle: Yeah. | 16:36 |
| bnemec | Looks like maybe it ran out of fds. | 16:36 |
| bnemec | I also dropped the number of Heat engine workers. | 16:36 |
| bnemec | When it died, we had like 13 heat stacks creating. | 16:36 |
| bnemec | And the cpu was pegged. | 16:36 |
| shardy | michchap: Hi, I think https://blueprints.launchpad.net/tripleo/+spec/refactor-puppet-manifests can be marked implemented now, do you agree? | 16:37 |
| *** jpich has quit IRC | 16:37 | |
| michchap | shardy: I agree. Done. | 16:39 |
| shardy | michchap: thanks! | 16:39 |
| michchap | shardy: sorry for ending up the owner of that - it was other people's work! | 16:39 |
| *** akrivoka has quit IRC | 16:39 | |
| EmilienM | michchap: lol | 16:40 |
| EmilienM | thanks for the 123023 patches ! | 16:40 |
| shardy | michchap: hehe - it was a team effort :) | 16:40 |
| michchap | EmilienM: I submitted like 1 change at the start because I was impatient :D | 16:40 |
| EmilienM | ahah | 16:40 |
| slagle | bnemec: i guess we never fixed the fd limit on the oc | 16:41 |
| bnemec | slagle: Yeah, I looked into it when this first happened and it looked like it was set to 4096 correctly. | 16:42 |
| bnemec | When I was watching the mysql connections they were hovering around 700, which seemed pretty safe. | 16:43 |
| bnemec | But I haven't checked in a couple of days, so it's possible they had increased when we re-enabled jobs and increased the number of envs. | 16:43 |
| *** florianf has quit IRC | 16:44 | |
| slagle | bnemec: i dont think it's the connections, it's the open files limit | 16:46 |
| slagle | maybe they're related | 16:46 |
| bnemec | slagle: Yeah, I think they are. Rabbit has the same issue. | 16:48 |
| *** zeroshft has quit IRC | 16:48 | |
| bnemec | And I see it's limited to 1024: open files (-n) 1024 | 16:49 |
| bnemec | I thought we had fixed that. :-/ | 16:49 |
| *** zeroshft has joined #tripleo | 16:49 | |
| sshnaidm | bnemec, does geard run therE? | 16:49 |
| bnemec | sshnaidm: No, that runs in a VM. This is the host cloud controller. | 16:50 |
| sshnaidm | slagle, it's the same limit for both of them - files and connections | 16:50 |
| shardy | Jokke_, gfidente: should we be either marking https://bugs.launchpad.net/tripleo/+bug/1589922 invalid or wontfix? | 16:50 |
| openstack | Launchpad bug 1589922 in tripleo "OSD data dir expected in /var/lib/ceph/osd" [High,In progress] - Assigned to Erno Kuvaja (jokke) | 16:50 |
| slagle | bnemec: we fixed in the undercloud | 16:50 |
| *** lucasagomes is now known as lucas-dinner | 16:50 | |
| pabelanger | So, regarding tripleo-test-cloud-rh1. What are the changes of getting a new project called openstackzuul created? Reason being, from nodepool we can see all the private infrastructure being run in the openstack-nodepool project which is starting to flood our nodepool debug logs | 16:50 |
| sshnaidm | bnemec, ok, because it was the same problem with geard before | 16:50 |
| shardy | The discussion in the bug seems inconclusive | 16:50 |
| *** tosky has quit IRC | 16:51 | |
| pabelanger | we'd still keep openstack-nodepool just add openstackzuul where we launch our VMs | 16:51 |
| slagle | bnemec: and also in my ha pacemaker env, the limit is 20485 | 16:51 |
| gfidente | shardy, was just working on that, let me add a comment | 16:51 |
| slagle | bnemec: so i think it's just not fixed for non-pacemaker envs | 16:51 |
| shardy | gfidente: Ok, thanks, I'm just trying to groom the newton-3 bugs a bit | 16:51 |
| slagle | since mariadb is started under systemd | 16:51 |
| bnemec | slagle: Oh /facepalm | 16:51 |
| *** tosky has joined #tripleo | 16:51 | |
| *** fragatina has joined #tripleo | 16:51 | |
| slagle | bnemec: quick fix would be to copy /etc/systemd/system/mariadb.service.d/limits.conf from uc to the oc, and restart mariadb | 16:52 |
| *** rhallisey has quit IRC | 16:52 | |
| *** rhallisey has joined #tripleo | 16:54 | |
| *** sshnaidm has quit IRC | 16:54 | |
| *** fragatin_ has quit IRC | 16:54 | |
| bnemec | slagle: Done | 16:54 |
| bnemec | We need to add that to the deployment docs too so this doesn't happen again. | 16:55 |
| *** yamahata has quit IRC | 16:55 | |
| *** yamahata has joined #tripleo | 16:55 | |
| slagle | yea i'll add it to the etherpad | 16:56 |
| slagle | and file a bug to fix in tht | 16:56 |
| pabelanger | bnemec: slagle | 16:56 |
| pabelanger | err | 16:56 |
| pabelanger | bnemec: slagle: does openstack-infra have admin credentials on tripleo-test-cloud-rh1? | 16:57 |
| pabelanger | should they? | 16:57 |
| pabelanger | I would ask, because might be able to drive some of the configuration of the cloud via ansible and system-config | 16:57 |
| pabelanger | but I'd need to confirm with openstack-infra if we wanted to do that | 16:58 |
| bnemec | pabelanger: It's a TripleO-deployed cloud, so we wouldn't want to do that. | 16:58 |
| slagle | pabelanger: afaik, openstack-nodepool is not admin | 16:58 |
| bnemec | It's actually an excellent exercise in dogfooding. | 16:58 |
| pabelanger | Right | 16:58 |
| slagle | pabelanger: what type of configuration? | 16:58 |
| *** julim has quit IRC | 16:58 | |
| bnemec | Because we probably have users who have hit this too. | 16:58 |
| *** apetrich has quit IRC | 16:59 | |
| pabelanger | slagle: basically, were using ansible now to provision our openstack clouds configuration, like projects, security groups, etc. EG: https://review.openstack.org/#/c/353594/ Be relaunched blueblox yesterday and it worked well | 16:59 |
| pabelanger | was going to talk with tripleo and see if you'd be interested in starting to do that too. From what I see, it is just bash scripts for the most part | 17:00 |
| bnemec | Oh, so this is unrelated to the mariadb thing. :-) | 17:00 |
| slagle | yea :) | 17:00 |
| slagle | pabelanger: i guess this might work. you're just configuring things in keystone/neutron/etc? | 17:01 |
| bnemec | I'm fine with using standard infra tooling to configure things. | 17:01 |
| slagle | by talking to api's? | 17:01 |
| pabelanger | slagle: if you see scroll back about the openstackzuul project / user, I plan to use ansible to create our projects settings with it | 17:01 |
| pabelanger | slagle: creating of security groups for example, and flavors | 17:02 |
| *** rcernin has joined #tripleo | 17:02 | |
| pabelanger | things we need to do to make nodepool work but now we are automating it with shade and ansible | 17:02 |
| slagle | ok, i think it sounds fine | 17:03 |
| slagle | what do other clouds do? | 17:03 |
| slagle | do they create an admin user for infra to use? | 17:03 |
| pabelanger | usually not, | 17:03 |
| pabelanger | we only have admin for bluebox | 17:03 |
| pabelanger | so, tripleo doesn't need to share admin creds with openstack-infra | 17:04 |
| pabelanger | but maybe we start moving tripleo-ci away from bash scripts to ansible playbooks | 17:05 |
| pabelanger | _shurgs_ | 17:05 |
| pabelanger | only if people are interested | 17:05 |
| *** rcernin has quit IRC | 17:05 | |
| *** rcernin has joined #tripleo | 17:05 | |
| *** fragatina has quit IRC | 17:06 | |
| *** fragatina has joined #tripleo | 17:07 | |
| slagle | pabelanger: is prepare-ovb-cloud.sh what it replaces? | 17:07 |
| slagle | (from tripleo-ci) | 17:07 |
| pabelanger | slagle: Ya, that was the file I was looking at. | 17:08 |
| slagle | i think it makes sense. happy to use standard infra tooling if we can make it work | 17:08 |
| bnemec | Note that there's tripleo-specific stuff in that file too. | 17:08 |
| slagle | bnemec: yea, ovb stuff | 17:08 |
| bnemec | Infra's not going to configure the OVB flavors, for instance. | 17:09 |
| bnemec | Right | 17:09 |
| slagle | i noticed, we'd have to split that out | 17:09 |
| *** apetrich has joined #tripleo | 17:09 | |
| openstackgerrit | Miles Gould proposed openstack/diskimage-builder: Allow installing proliant tools from an OS package https://review.openstack.org/344214 | 17:09 |
| pabelanger | Lets do this, if you create the openstackzuul user / project (see backscroll for reason), I'll push a patch to review.o.o showing an example configuration | 17:09 |
| bnemec | But yeah, total +1 to using infra tools to create infra bits. | 17:10 |
| *** bana_k has joined #tripleo | 17:10 | |
| slagle | pabelanger: ok. is there any special way I should create it? | 17:11 |
| *** fragatina has quit IRC | 17:12 | |
| pabelanger | slagle: I don't think so. The openstackzuul user needs to be part of the openstackzuul project, everything else we should be able to automate for settings | 17:12 |
| pabelanger | slagle: we need to be on the same network as your private infra and OVB servers | 17:12 |
| pabelanger | as we are moving nodepool vms from openstack-nodepool to openstackzuul | 17:13 |
| pabelanger | we won't make the cut over until we can manually launch a server and test networking bits | 17:13 |
| *** numans has joined #tripleo | 17:15 | |
| bnemec | That shouldn't be a problem. The private network is already shared. | 17:15 |
| *** mgould is now known as mgould|afk | 17:17 | |
| *** jeckersb is now known as jeckersb_gone | 17:17 | |
| openstackgerrit | Saravanan KR proposed openstack/os-net-config: Add support for OVS DPDK Bond https://review.openstack.org/349183 | 17:22 |
| *** fultonj_ has joined #tripleo | 17:23 | |
| *** fragatina has joined #tripleo | 17:27 | |
| *** electrofelix has quit IRC | 17:28 | |
| *** Goneri has quit IRC | 17:28 | |
| *** apetrich has quit IRC | 17:35 | |
| *** egafford has joined #tripleo | 17:37 | |
| *** ebarrera has quit IRC | 17:38 | |
| *** abehl has quit IRC | 17:41 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-docs: Document Ceph repositories to be enabled https://review.openstack.org/352910 | 17:47 |
| *** leanderthal is now known as leanderthal|afk | 17:50 | |
| openstackgerrit | Dan Prince proposed openstack/tripleo-heat-templates: Move neutron settings out of puppet/compute.yaml https://review.openstack.org/354201 | 17:52 |
| dprince_ | bnemec: hey, what do you have in mind for here https://review.openstack.org/#/c/300139/ | 17:54 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Disable sahara services by default https://review.openstack.org/352886 | 17:54 |
| dprince_ | bnemec: I would like to get that in. It has been broken for months now | 17:55 |
| gfidente | EmilienM, slagle ^^ that sahara thing landed in mitaka | 17:56 |
| gfidente | so I am not sure it's good to disable it in newton | 17:56 |
| *** fragatina has quit IRC | 17:57 | |
| openstackgerrit | Merged openstack/tripleo-quickstart: files on undercloud had wrong uid https://review.openstack.org/352398 | 17:57 |
| *** ohamada has quit IRC | 17:57 | |
| openstackgerrit | Merged openstack/tripleo-quickstart: fix teardown (tagging and ansible_user) https://review.openstack.org/352112 | 17:57 |
| slagle | gfidente: why? we dont need sahara by default | 17:57 |
| gfidente | slagle, right I am just saying people upgrading have it enabled though | 17:58 |
| slagle | upgrading from mitaka? | 17:58 |
| gfidente | from mitaka yes | 17:58 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-docs: Document Ceph repositories to be enabled https://review.openstack.org/352910 | 18:00 |
| slagle | gfidente: ok, well darn | 18:00 |
| *** jeckersb_gone is now known as jeckersb | 18:00 | |
| gfidente | slagle, yeah in general we don't have a 'ensure disabled' mechanism though it sounds like what we need really | 18:01 |
| slagle | gfidente: can u comment on the review? | 18:02 |
| slagle | gfidente: my motivation with disabling it is mainly for CI | 18:02 |
| slagle | but i can do that via an environment | 18:02 |
| gfidente | ok | 18:02 |
| slagle | i'd like to optimize what we test in the gate to what is required | 18:02 |
| *** tosky has quit IRC | 18:03 | |
| *** julim has joined #tripleo | 18:04 | |
| EmilienM | gfidente: https://review.openstack.org/#/c/353673/2/manifests/osd.pp | 18:12 |
| EmilienM | oh dear... | 18:12 |
| EmilienM | again an exec? | 18:12 |
| gfidente | I think we'd need an exec to check if {data} is a dir or a block device anyway | 18:12 |
| gfidente | but I have a local copy using selinux::dircontext | 18:12 |
| gfidente | meantime as per comment in the BZ I think we can make '/srv/data' labeled correctly by ceph-selinux | 18:13 |
| *** numans has quit IRC | 18:14 | |
| EmilienM | gfidente: I -1 for stylistic things but I won't block the exec if we really need it | 18:15 |
| gfidente | thanks so I will fix those | 18:16 |
| gfidente | but also let's wait if we can get ceph-selinux to adopt us | 18:16 |
| gfidente | though in general the puppet-ceph feature looks sensible | 18:16 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/os-cloud-config: Updated from global requirements https://review.openstack.org/350676 | 18:16 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/os-collect-config: Updated from global requirements https://review.openstack.org/350905 | 18:16 |
| gfidente | what did you mean here "can't we do it in packaging side? do we really need this exec?" ? | 18:17 |
| gfidente | which alternative were you thinking about? | 18:17 |
| *** ayoung has quit IRC | 18:24 | |
| EmilienM | gfidente: in ceph-selinux? | 18:24 |
| EmilienM | I don't know.. | 18:24 |
| gfidente | ah right | 18:24 |
| EmilienM | all those execs scare me | 18:24 |
| EmilienM | lunch now | 18:24 |
| gfidente | yeah the probblem is if people passes any non-default path they will get denials | 18:24 |
| gfidente | so we *can* fix the default path in ceph-selinux but in general the problem remains in the module | 18:24 |
| EmilienM | ok let's ffix it in the module then | 18:25 |
| gfidente | we're doing things in ceph-selinux as well, so we have less pressure on module part | 18:27 |
| openstackgerrit | Sarath Kumar Sankaran Kutty proposed openstack/os-net-config: Cleanups and new unit test for IVS https://review.openstack.org/346161 | 18:27 |
| *** alee has quit IRC | 18:31 | |
| *** fragatina has joined #tripleo | 18:35 | |
| *** chlong|mtg has quit IRC | 18:37 | |
| *** chlong|mtg has joined #tripleo | 18:38 | |
| *** alee has joined #tripleo | 18:44 | |
| *** amoralej is now known as amoralej|off | 18:47 | |
| *** lblanchard has quit IRC | 18:54 | |
| *** apetrich has joined #tripleo | 18:54 | |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/tripleo-quickstart: Add undercloud reboot https://review.openstack.org/350489 | 18:56 |
| *** fragatin_ has joined #tripleo | 19:00 | |
| *** fragatina has quit IRC | 19:03 | |
| *** ebarrera has joined #tripleo | 19:04 | |
| jtomasek | d0ugal, shardy: here is what the example mistral environment looks like: http://paste.openstack.org/show/554117/ | 19:05 |
| jtomasek | d0ugal, shardy: that way is totally fine for GUI | 19:06 |
| jtomasek | d0ugal: GetCapabilitiesAction combines environments from here with the capabilities map structure | 19:07 |
| *** athomas has quit IRC | 19:07 | |
| jtomasek | d0ugal: IOW, GetParametersAction goes through the capabilities map and adds enabled: true to the environments there if they are present in mistral environment. | 19:08 |
| *** dprince_ has quit IRC | 19:08 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Add missing deprecated param https://review.openstack.org/354345 | 19:10 |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: Removing WARNING: line has more than 140 characters in puppet-tripleo profiles https://review.openstack.org/354054 | 19:11 |
| openstackgerrit | Stephanie Miller proposed openstack/diskimage-builder: Spec for adding diskimage_builder module in-chroot https://review.openstack.org/336664 | 19:14 |
| *** oshvartz has joined #tripleo | 19:23 | |
| *** sarath has joined #tripleo | 19:26 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: WIP - Implement undercloud upgrade job - Mitaka -> Newton https://review.openstack.org/346995 | 19:31 |
| *** ayoung has joined #tripleo | 19:40 | |
| *** oshvartz has quit IRC | 19:43 | |
| *** jpena is now known as jpena|off | 19:45 | |
| *** jprovazn has quit IRC | 19:50 | |
| *** fultonj has quit IRC | 19:52 | |
| *** fultonj_ has quit IRC | 19:52 | |
| *** fultonj has joined #tripleo | 19:53 | |
| *** jprovazn has joined #tripleo | 20:02 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Remove deprecated net_ip_subnet_map output https://review.openstack.org/346055 | 20:05 |
| *** fragatin_ has quit IRC | 20:15 | |
| *** fragatina has joined #tripleo | 20:16 | |
| rlandy | bnemec: if these are the values tripleo-quickstart uses for undercloud and overcloud vms, https://github.com/openstack/tripleo-quickstart/blob/master/roles/common/defaults/main.yml#L10, how are you getting away with using such small flavors for baremetal and undercloud in OVB testing? | 20:19 |
| bnemec | rlandy: I have nothing to do with quickstart. | 20:20 |
| bnemec | Our CI has always run on the smallest vms possible. | 20:20 |
| rlandy | that's admirable | 20:20 |
| bnemec | For resource reasons, and to avoid us bloating the memory usage accidentally. | 20:20 |
| rlandy | I am not having a lot of success dropping the flavor requirements | 20:21 |
| bnemec | rlandy: undercloud, overcloud, or both? | 20:21 |
| bnemec | Note that we just bumped the overcloud vms to 5.5 GB because we were seeing intermittent ooms in jobs. | 20:22 |
| *** Goneri has joined #tripleo | 20:22 | |
| rlandy | bnemec: I used smaller flavors for both undercloud and overcloud ... | 20:22 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: Only set if it is not already defined https://review.openstack.org/354365 | 20:23 |
| bnemec | rlandy: How big? | 20:23 |
| rlandy | in the past, with a smaller undercloud, I was seeing issues with keystone dying - probably the same oom errors | 20:23 |
| rlandy | now, with a m1.medium undercloud, deploy is failing out | 20:23 |
| bnemec | rlandy: medium is only 4 GB. That's not enough, unfortunately. | 20:24 |
| bnemec | We run 6 in CI. | 20:24 |
| rlandy | the main issue is that I don't have access to create custom flavors (relic of the aim to optimize general cloud usage - not my personal quota) | 20:24 |
| rlandy | so I land up loosing 2,4,8, 16 | 20:25 |
| bnemec | | 18e29a9b-8b32-4b61-bc76-6346dc7468f2 | undercloud | 6144 | 40 | 0 | | 2 | 1.0 | True | | 20:25 |
| bnemec | | 2aaef5e1-a585-4bb9-b1e8-134d82f55982 | bmc | 512 | 20 | 0 | | 1 | 1.0 | True | | 20:25 |
| bnemec | | 49a7fbb0-2c0b-4441-a298-d69085760516 | baremetal | 5632 | 41 | 0 | | 1 | 1.0 | True | | 20:25 |
| bnemec | rlandy: You're probably stuck with 8 then. | 20:25 |
| rlandy | I can accept that baremetal may be big enough (if I use your requirements) when we move to public cloud | 20:25 |
| rlandy | but the undercloud size surprises me | 20:26 |
| bnemec | If they won't give you custom flavors, then the extra memory usage is the cost, I guess. | 20:26 |
| bnemec | The other problem is that when you add vcpus to the flavor, it increases the memory usage too. | 20:26 |
| bnemec | Since I think m1.large has 4 vcpus, that might make 8 GB required anyway. | 20:27 |
| rlandy | yeah - I am probably doing a useless test here - trying to get minimum requirements for a public cloud using a ' no custom flavor' private one | 20:27 |
| rlandy | still - that aside, I had oom issue with a 8G undercloud | 20:28 |
| bnemec | Strange, I periodically use an 8 GB undercloud if I don't need to build images. | 20:30 |
| rlandy | I can get away with it for mitaka - not master | 20:32 |
| *** [1]cdearborn has quit IRC | 20:36 | |
| *** ebalduf has quit IRC | 20:38 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Set coordination url for gnocchi storage https://review.openstack.org/354378 | 20:38 |
| *** beagles is now known as beagles_biab | 20:38 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Remove deprecated net_ip_uri_map output https://review.openstack.org/354381 | 20:41 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: Only set if it is not already defined https://review.openstack.org/354365 | 20:42 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Remove deprecated net_ip_uri_map output https://review.openstack.org/354381 | 20:43 |
| *** ebarrera has quit IRC | 20:44 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Set the default gnocchi driver to swift https://review.openstack.org/354385 | 20:44 |
| *** yamahata has quit IRC | 20:45 | |
| *** jprovazn has quit IRC | 20:49 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: Only set if it is not already defined https://review.openstack.org/354365 | 20:51 |
| *** pradk has quit IRC | 20:54 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Remove unneeded net_ip_map dump into hieradata https://review.openstack.org/354391 | 20:57 |
| dtrainor | Hi. I'm playing around with the UI and when I try to do node assignment, I see some failures that I don't recognize. I think I'm missing an allowable method somewhere but I'm not certain where http://paste.openstack.org/show/554262/ | 20:57 |
| dtrainor | if i had to guess, it's ironic-inspector not allowing a PATCH method I think... | 20:58 |
| *** yamahata has joined #tripleo | 21:06 | |
| dtrainor | ahh, i see it.... "Client-side error" is a red herring. The real problem is with the UI preforming introspection. | 21:07 |
| *** rhallisey has quit IRC | 21:07 | |
| *** weshay has quit IRC | 21:09 | |
| stevebaker | bnemec: does bmc-image.qcow2 get rebuilt? | 21:16 |
| * stevebaker is *so* lazy | 21:16 | |
| bnemec | stevebaker: No, it won't. I uploaded it manually. | 21:17 |
| bnemec | I should upload a patched one though. | 21:17 |
| *** ayoung has quit IRC | 21:18 | |
| *** zeroshft_ has joined #tripleo | 21:21 | |
| *** shardy has quit IRC | 21:21 | |
| *** alee has quit IRC | 21:21 | |
| *** coolsvap has quit IRC | 21:21 | |
| *** zeroshft has quit IRC | 21:23 | |
| *** weshay has joined #tripleo | 21:24 | |
| *** rlandy has quit IRC | 21:24 | |
| *** pradk has joined #tripleo | 21:26 | |
| *** Goneri has quit IRC | 21:28 | |
| stevebaker | yeah, it would reduce problems for others | 21:29 |
| *** ebalduf has joined #tripleo | 21:29 | |
| *** social has quit IRC | 21:30 | |
| *** zaneb has joined #tripleo | 21:30 | |
| *** zaneb has quit IRC | 21:30 | |
| *** zaneb has joined #tripleo | 21:31 | |
| *** jayg is now known as jayg|g0n3 | 21:31 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Pass ServiceNetMap to services https://review.openstack.org/354414 | 21:41 |
| *** Goneri has joined #tripleo | 21:43 | |
| slagle | hmm? we promoted. how? | 21:48 |
| slagle | and now the promoted repo is breaking the multinode job with the same cinder error | 21:48 |
| *** maeca1 has quit IRC | 21:51 | |
| gfidente | slagle, cinder error was fixed in cinder | 21:51 |
| slagle | gfidente: https://bugs.launchpad.net/tripleo/+bug/1612434 | 21:52 |
| openstack | Launchpad bug 1612434 in tripleo "multinode jobs failing during pingtest with "Resource CREATE failed: ResourceInError: resources.volume1: Went to status error due to "Unknown""" [Undecided,New] | 21:52 |
| slagle | gfidente: maybe a repo got promoted that did not yet have the fix | 21:52 |
| slagle | b/c now i'm seeing the error in the check queue | 21:52 |
| gfidente | is that possible at all? | 21:53 |
| slagle | i dont know it could have happened | 21:53 |
| slagle | i'm asking in rdo if it got done manually | 21:53 |
| slagle | perhaps accidentally | 21:53 |
| openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci: Don't deploy net-iso env when we won't use it https://review.openstack.org/347503 | 21:54 |
| bnemec | Yeah, I'm seeing a bunch of cinder errors killing the ping test. | 21:55 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Now fly and set cinder::api::bind_host within service template https://review.openstack.org/318344 | 22:01 |
| bnemec | lol, nice commit message | 22:02 |
| bnemec | Fly, you fools! | 22:02 |
| *** egafford has quit IRC | 22:12 | |
| *** alee has joined #tripleo | 22:16 | |
| *** maeca1 has joined #tripleo | 22:19 | |
| *** tbarron is now known as tbarron|gone | 22:19 | |
| *** weshay has quit IRC | 22:20 | |
| *** Goneri has quit IRC | 22:21 | |
| *** morazi has quit IRC | 22:37 | |
| *** panda is now known as panda|zz | 22:41 | |
| *** saneax_AFK is now known as saneax | 22:46 | |
| *** saneax is now known as Guest91909 | 22:46 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-docs: Document Ceph repositories to be enabled https://review.openstack.org/352910 | 23:00 |
| *** maeca1 has quit IRC | 23:01 | |
| *** Guest91909 is now known as saneax_AFK | 23:02 | |
| *** thrash is now known as thrash|g0ne | 23:15 | |
| *** zeroshft_ has quit IRC | 23:30 | |
| openstackgerrit | Giulio Fidente proposed openstack/python-tripleoclient: Use non-bracketed version of addresses for keystone.initialize https://review.openstack.org/354443 | 23:35 |
| openstackgerrit | Giulio Fidente proposed openstack/python-tripleoclient: Use non-bracketed version of addresses for keystone.initialize https://review.openstack.org/354443 | 23:37 |
| *** StevenK_ is now known as StevenK | 23:38 | |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: DO NOT MERGE - Periodic test https://review.openstack.org/346949 | 23:44 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: DO NOT MERGE - Periodic test https://review.openstack.org/346949 | 23:47 |
| *** beagles_biab is now known as beagles | 23:49 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
