| *** yamahata has quit IRC | 00:04 | |
| *** matsuhashi has joined #tripleo | 00:21 | |
| *** michchap has joined #tripleo | 00:50 | |
| *** nati_ueno has quit IRC | 00:53 | |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-image-elements: Add os-is-bootstrap-host element and script https://review.openstack.org/86435 | 01:09 |
|---|---|---|
| *** jml has quit IRC | 01:16 | |
| *** nati_ueno has joined #tripleo | 01:19 | |
| *** jml has joined #tripleo | 01:20 | |
| *** nati_ueno has quit IRC | 01:20 | |
| *** yamahata has joined #tripleo | 01:23 | |
| *** nosnos has joined #tripleo | 01:43 | |
| *** pcrews_ has joined #tripleo | 01:49 | |
| *** pcrews has quit IRC | 01:52 | |
| *** yamahata has quit IRC | 01:54 | |
| *** yamahata has joined #tripleo | 01:54 | |
| *** pcrews_ has quit IRC | 02:16 | |
| *** julim has quit IRC | 02:26 | |
| *** untriaged-bot has joined #tripleo | 03:00 | |
| untriaged-bot | Untriaged bugs so far: | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1325762 | 03:00 |
| uvirtbot | Launchpad bug 1325762 in tripleo "intermittent tftp issues using physical hardware" [Undecided,New] | 03:00 |
| *** untriaged-bot has quit IRC | 03:00 | |
| *** tzumainn has quit IRC | 03:03 | |
| *** eghobo has joined #tripleo | 03:16 | |
| *** nosnos has quit IRC | 03:18 | |
| *** greghaynes has quit IRC | 03:51 | |
| *** greghaynes has joined #tripleo | 03:52 | |
| *** greghaynes has quit IRC | 03:52 | |
| *** greghaynes has joined #tripleo | 03:52 | |
| *** matsuhashi has quit IRC | 04:04 | |
| *** nosnos has joined #tripleo | 04:06 | |
| *** matsuhashi has joined #tripleo | 04:13 | |
| *** noslzzp has joined #tripleo | 04:16 | |
| *** rameshg87 has joined #tripleo | 04:58 | |
| *** sseago has quit IRC | 04:59 | |
| StevenK | lifeless: You comment "You have it spot on" on my os-cloud-config logging branch is on line 1 of the base patchset -- are you agreeing with Tomas, or trying to say something else? | 05:01 |
| lifeless | agreeing with Tomas | 05:03 |
| lifeless | hmm, gertty fail I guess | 05:03 |
| *** akuznetsov has joined #tripleo | 05:06 | |
| *** ramishra has joined #tripleo | 05:17 | |
| rameshg87 | hello tripleo-reviewers | 05:28 |
| rameshg87 | request you to take a look at the reviews: https://review.openstack.org/97143 https://review.openstack.org/97301 | 05:29 |
| *** rakesh_hs has joined #tripleo | 05:32 | |
| *** jprovazn has joined #tripleo | 05:33 | |
| *** ramishra has quit IRC | 05:34 | |
| *** ramishra has joined #tripleo | 05:37 | |
| *** flashgordon is now known as jogo | 05:46 | |
| *** xuhaiwei_ has joined #tripleo | 05:47 | |
| *** sseago has joined #tripleo | 05:49 | |
| *** akuznetsov has quit IRC | 05:53 | |
| *** jcoufal has joined #tripleo | 05:57 | |
| *** rameshg87 has left #tripleo | 05:59 | |
| *** rdopiera has joined #tripleo | 06:01 | |
| *** akuznetsov has joined #tripleo | 06:13 | |
| *** eghobo has quit IRC | 06:14 | |
| *** eghobo has joined #tripleo | 06:14 | |
| *** eguz has joined #tripleo | 06:23 | |
| *** eghobo has quit IRC | 06:27 | |
| *** akuznets_ has joined #tripleo | 06:28 | |
| *** akuznetsov has quit IRC | 06:29 | |
| *** akuznetsov has joined #tripleo | 06:34 | |
| *** akuznets_ has quit IRC | 06:37 | |
| *** eguz has quit IRC | 06:54 | |
| *** nosnos has quit IRC | 06:57 | |
| *** lsmola_ has joined #tripleo | 07:05 | |
| *** e0ne has joined #tripleo | 07:16 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Silence the deprecation warning about with statements https://review.openstack.org/98703 | 07:20 |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Silence the deprecation warning about with statements https://review.openstack.org/98703 | 07:22 |
| *** akuznets_ has joined #tripleo | 07:25 | |
| *** ifarkas has joined #tripleo | 07:26 | |
| *** akuznetsov has quit IRC | 07:27 | |
| *** ramishra has quit IRC | 07:29 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Don't use deprecated keystone module https://review.openstack.org/98705 | 07:39 |
| SpamapS | 7/win 13 | 07:43 |
| rdopiera | \/msg #tripleo /lose 21 | 07:44 |
| *** akuznets_ has quit IRC | 07:51 | |
| *** akuznetsov has joined #tripleo | 07:51 | |
| *** akuznets_ has joined #tripleo | 07:53 | |
| *** markmc has joined #tripleo | 07:53 | |
| *** akuznetsov has quit IRC | 07:56 | |
| *** pblaho has joined #tripleo | 07:57 | |
| openstackgerrit | Clint 'SpamapS' Byrum proposed a change to openstack/tripleo-heat-templates: Replace nagios3 template with HOT rewrite https://review.openstack.org/90665 | 07:57 |
| SpamapS | ^^ We might want to consider moving templates like that out of t-h-t into some other utility repo (or a contrib dir or something). It is mostly for our own infrastructure needs. | 07:58 |
| *** jistr has joined #tripleo | 08:00 | |
| SpamapS | hrm. | 08:01 |
| SpamapS | I'd love to test https://review.openstack.org/#/c/97703/ locally.. | 08:01 |
| *** akuznets_ has quit IRC | 08:01 | |
| SpamapS | but I can't even build a working seed | 08:01 |
| *** russell_h has quit IRC | 08:01 | |
| *** russell_h has joined #tripleo | 08:02 | |
| *** openstackgerrit has quit IRC | 08:02 | |
| *** athomas has joined #tripleo | 08:04 | |
| *** martyntaylor has joined #tripleo | 08:07 | |
| *** openstackgerrit has joined #tripleo | 08:07 | |
| elynn | Hi all, do you have time to review this patch? https://review.openstack.org/#/c/90887/ | 08:07 |
| SpamapS | elynn: any reason you don't just put that image in ~/.cache/image-create ? | 08:08 |
| *** derekh_afk is now known as derekh_ | 08:09 | |
| elynn | hi SpamapS , ubuntu element only support tar.gz now, not support qcow2 image. | 08:09 |
| SpamapS | elynn: also you can use a file:// path already | 08:09 |
| SpamapS | oh qcow2? | 08:09 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-heat-templates: Enable multi-node rabbitmq cluster https://review.openstack.org/98709 | 08:10 |
| SpamapS | why would you want to use that? | 08:10 |
| elynn | because my custom image is in qcow2 format... | 08:10 |
| *** anteaya has quit IRC | 08:10 | |
| SpamapS | elynn: ok, I'll leave my feedback in the review. thanks for clarifying. | 08:11 |
| elynn | thx for your review, SpamapS : ) | 08:12 |
| *** russell_h has quit IRC | 08:13 | |
| *** ramishra has joined #tripleo | 08:13 | |
| *** andreaf has joined #tripleo | 08:16 | |
| *** russell_h has joined #tripleo | 08:16 | |
| *** lucasagomes has joined #tripleo | 08:20 | |
| derekh_ | lifeless: did you spot the patch for https://bugs.launchpad.net/tripleo/+bug/1327090 | 08:20 |
| uvirtbot | Launchpad bug 1327090 in tripleo "can't deploy ci-overclouds on Ubuntu - ensure-bridge wipes out /e/n/i" [High,In progress] | 08:20 |
| *** russell_h has quit IRC | 08:21 | |
| lifeless | derekh_: I have been (gosh, golly) reviewing today | 08:22 |
| lifeless | derekh_: but I saw it and had it on my to-try list | 08:23 |
| lifeless | let me kick that now | 08:23 |
| lifeless | first off I have to test this new (hopefully fixed) node | 08:23 |
| *** russell_h has joined #tripleo | 08:24 | |
| lifeless | actually, looks like i don't need to test anything there, righto onto the patch | 08:25 |
| lifeless | derekh_: btw have you seen tripleo-end-to-end/repo_refs.yaml ? | 08:25 |
| derekh_ | lifeless: cool, btw, I had noticed we weren't getting elastic recheck comments for signatures that I thought we should have been, so went digging https://review.openstack.org/#/c/98154/ | 08:26 |
| derekh_ | will talk to sdague later when he is on | 08:26 |
| derekh_ | lifeless: no hadn't seen it | 08:26 |
| *** ramishra has quit IRC | 08:27 | |
| lifeless | when you import the seed.hp1.. rc file it automerges etc | 08:28 |
| lifeless | nifty and the inspiration for a spec ;) | 08:28 |
| openstackgerrit | Gonéri Le Bouder proposed a change to openstack/tripleo-image-elements: indent using 4 spaces (1/3) https://review.openstack.org/93206 | 08:29 |
| derekh_ | lifeless: the spec about multiple refs in source-repositories ? will take a look | 08:29 |
| lifeless | rightyup | 08:29 |
| lifeless | 40 node cloud deploy kicked off | 08:30 |
| lifeless | it will need manual intervention to 'ensure-bridge br-untagged' | 08:30 |
| lifeless | I don't want to take this cloud to production while such manual steps remain | 08:30 |
| *** russell_h has quit IRC | 08:31 | |
| derekh_ | lifeless: ok, makes sense | 08:31 |
| lifeless | how did you deploy the rh region without manual intervention for that step? | 08:32 |
| lifeless | just needed the one bridge I guess? | 08:32 |
| *** sseago has quit IRC | 08:33 | |
| *** russell_h has joined #tripleo | 08:33 | |
| SpamapS | No valid host was found. Reason: No conductor service registered which supports driver null. (HTTP 400) | 08:33 |
| SpamapS | argh | 08:33 |
| *** sseago has joined #tripleo | 08:33 | |
| *** jtomasek has quit IRC | 08:33 | |
| *** jtomasek has joined #tripleo | 08:34 | |
| *** martyntaylor has left #tripleo | 08:34 | |
| SpamapS | doh | 08:34 |
| SpamapS | 0 nodes | 08:34 |
| derekh_ | lifeless: iirc, we did have manual intervention | 08:35 |
| *** rameshg87 has joined #tripleo | 08:35 | |
| *** proffalken has quit IRC | 08:35 | |
| *** proffalken has joined #tripleo | 08:36 | |
| lifeless | SpamapS: still lagged? | 08:38 |
| *** russell_h has quit IRC | 08:38 | |
| *** akrivoka has joined #tripleo | 08:38 | |
| *** yamahata has quit IRC | 08:41 | |
| *** russell_h has joined #tripleo | 08:42 | |
| *** rlandy has joined #tripleo | 08:46 | |
| *** e0ne has quit IRC | 08:47 | |
| SpamapS | lifeless: -ish .. | 08:50 |
| SpamapS | lifeless: much better now.. just mostly trying to get ahead of the flood of monday patches/reviews/etc. :-P | 08:50 |
| lifeless | cool | 08:54 |
| *** rdopiera has quit IRC | 08:56 | |
| lifeless | mordred: https://review.openstack.org/#/c/97583/ <- I think thats a bug in pbr | 08:57 |
| *** russell_h has quit IRC | 08:58 | |
| *** russell_h has joined #tripleo | 08:59 | |
| lifeless | mordred: i'm happy to put up a patch | 08:59 |
| lifeless | mordred: if you like the idea I just put in the review | 08:59 |
| *** untriaged-bot has joined #tripleo | 09:00 | |
| untriaged-bot | Untriaged bugs so far: | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1325762 | 09:00 |
| uvirtbot | Launchpad bug 1325762 in tripleo "intermittent tftp issues using physical hardware" [Undecided,New] | 09:00 |
| *** untriaged-bot has quit IRC | 09:00 | |
| *** pelix has joined #tripleo | 09:01 | |
| *** russell_h has quit IRC | 09:02 | |
| *** IvanBerezovskiy has joined #tripleo | 09:02 | |
| *** pblaho has quit IRC | 09:02 | |
| *** rdopiera has joined #tripleo | 09:03 | |
| *** russell_h has joined #tripleo | 09:04 | |
| *** akrivoka_ has joined #tripleo | 09:11 | |
| *** russell_h has quit IRC | 09:14 | |
| *** akrivoka has quit IRC | 09:14 | |
| *** russell_h has joined #tripleo | 09:14 | |
| lifeless | hmm | 09:15 |
| lifeless | thats amusing | 09:15 |
| lifeless | now we're building all the computes | 09:15 |
| lifeless | before we try to spawn the control | 09:15 |
| openstackgerrit | Gonéri Le Bouder proposed a change to openstack/diskimage-builder: map-services: add openvswitch https://review.openstack.org/96749 | 09:17 |
| openstackgerrit | Gonéri Le Bouder proposed a change to openstack/diskimage-builder: map-services: add openvswitch https://review.openstack.org/96749 | 09:18 |
| *** russell_h has quit IRC | 09:18 | |
| *** russell_h has joined #tripleo | 09:20 | |
| *** pblaho has joined #tripleo | 09:21 | |
| *** yfaykya has quit IRC | 09:25 | |
| *** russell_h has quit IRC | 09:26 | |
| *** athomas has quit IRC | 09:26 | |
| *** ci-overcloud has joined #tripleo | 09:28 | |
| ci-overcloud | ************** ci-overcloud complete status=1 ************ | 09:28 |
| *** ci-overcloud has quit IRC | 09:28 | |
| *** yfaykya has joined #tripleo | 09:30 | |
| *** matsuhashi has quit IRC | 09:31 | |
| *** matsuhashi has joined #tripleo | 09:31 | |
| *** yfaykya has quit IRC | 09:35 | |
| *** yfaykya has joined #tripleo | 09:36 | |
| *** athomas has joined #tripleo | 09:37 | |
| *** akuznetsov has joined #tripleo | 09:38 | |
| lifeless | hmm | 09:38 |
| lifeless | | NovaCompute31 | 40c19b44-b33f-4040-af63-09a7ee4ae2ac | Error: Creation of server ci-overcloud-NovaCompute31-sfsoeoirxccn failed. | CREATE_FAILED | 2014-06-09T09:28:10Z | | 09:39 |
| *** matsuhashi has quit IRC | 09:39 | |
| *** matsuhashi has joined #tripleo | 09:40 | |
| lifeless | derekh_: tag :) - I need to sleep; in principle its just diagnose and fix the issue, then | 09:40 |
| lifeless | USE_CACHE=1 tripleo-image-elements/elements/tripleo-cd/deploy-ci-overcloud | 09:40 |
| lifeless | + login to the controller before things timeout and run ensure-bridge br-untagged eth2 after checking that os-collect-config --print returns sane data etc. | 09:40 |
| *** yfaykya has quit IRC | 09:41 | |
| lifeless | derekh_: / | 09:41 |
| lifeless | derekh_: the cached images have your patch in them | 09:42 |
| derekh_ | lifeless: ok, cool, will jump in in a few minutes | 09:42 |
| *** yfaykya has joined #tripleo | 09:42 | |
| lifeless | derekh_: I haven't looked at what caused that server to fail to deploy; if you put stuff into maintenance please add a note to machine-information-tab.txt | 09:42 |
| derekh_ | lifeless: will do | 09:43 |
| lifeless | derekh_: and if it needs physical/DC intervention get ng or ghe or $whomever to file a ticket for you | 09:43 |
| derekh_ | ok | 09:43 |
| lifeless | cool, thanks! | 09:43 |
| derekh_ | np | 09:43 |
| *** akrivoka_ has quit IRC | 09:46 | |
| *** yfaykya has quit IRC | 09:47 | |
| *** yfaykya has joined #tripleo | 09:48 | |
| *** yfaykya has quit IRC | 09:53 | |
| *** yfaykya has joined #tripleo | 09:54 | |
| *** e0ne has joined #tripleo | 09:55 | |
| *** yfaykya has quit IRC | 09:59 | |
| *** yfaykya has joined #tripleo | 10:01 | |
| *** russell_h has joined #tripleo | 10:01 | |
| *** yfaykya has quit IRC | 10:06 | |
| *** yfaykya has joined #tripleo | 10:07 | |
| *** anteaya has joined #tripleo | 10:11 | |
| *** yfaykya has quit IRC | 10:11 | |
| *** yfaykya has joined #tripleo | 10:13 | |
| *** yfaykya has quit IRC | 10:17 | |
| *** yfaykya has joined #tripleo | 10:19 | |
| *** yfaykya has quit IRC | 10:23 | |
| *** yfaykya has joined #tripleo | 10:25 | |
| *** e0ne has quit IRC | 10:29 | |
| *** yfaykya has quit IRC | 10:29 | |
| *** xuhaiwei_ has quit IRC | 10:30 | |
| *** yfaykya has joined #tripleo | 10:31 | |
| lxsli | lifeless: happy with https://review.openstack.org/#/c/94910/ ? | 10:33 |
| *** yfaykya has quit IRC | 10:36 | |
| *** yfaykya has joined #tripleo | 10:37 | |
| *** yfaykya has quit IRC | 10:42 | |
| *** yfaykya has joined #tripleo | 10:43 | |
| *** yfaykya has quit IRC | 10:48 | |
| *** yfaykya has joined #tripleo | 10:50 | |
| *** matsuhashi has quit IRC | 10:51 | |
| *** rakesh_hs has quit IRC | 10:52 | |
| *** yfaykya has quit IRC | 10:54 | |
| *** yfaykya has joined #tripleo | 10:56 | |
| *** rameshg87 has left #tripleo | 10:56 | |
| *** matsuhashi has joined #tripleo | 10:58 | |
| *** yfaykya has quit IRC | 11:00 | |
| *** e0ne has joined #tripleo | 11:00 | |
| *** yfaykya has joined #tripleo | 11:02 | |
| *** e0ne has quit IRC | 11:04 | |
| *** e0ne has joined #tripleo | 11:06 | |
| *** yfaykya has quit IRC | 11:07 | |
| *** yfaykya has joined #tripleo | 11:08 | |
| *** akrivoka has joined #tripleo | 11:11 | |
| *** yfaykya has quit IRC | 11:12 | |
| *** yfaykya has joined #tripleo | 11:14 | |
| *** yfaykya has quit IRC | 11:18 | |
| *** matsuhashi has quit IRC | 11:18 | |
| *** matsuhashi has joined #tripleo | 11:19 | |
| *** yfaykya has joined #tripleo | 11:20 | |
| *** matsuhashi has quit IRC | 11:24 | |
| *** yfaykya has quit IRC | 11:25 | |
| *** yfaykya has joined #tripleo | 11:26 | |
| *** ramishra has joined #tripleo | 11:28 | |
| *** yfaykya has quit IRC | 11:31 | |
| *** e0ne has quit IRC | 11:32 | |
| *** yfaykya has joined #tripleo | 11:32 | |
| *** weshay has joined #tripleo | 11:37 | |
| *** yfaykya has quit IRC | 11:37 | |
| *** yfaykya has joined #tripleo | 11:39 | |
| *** DuncanT has quit IRC | 11:41 | |
| *** rlandy has quit IRC | 11:42 | |
| *** yfaykya has quit IRC | 11:43 | |
| *** morazi has quit IRC | 11:43 | |
| *** yfaykya has joined #tripleo | 11:45 | |
| *** yfaykya has quit IRC | 11:49 | |
| *** yfaykya has joined #tripleo | 11:51 | |
| *** lucasagomes is now known as lucas-hungry | 11:51 | |
| *** rlandy has joined #tripleo | 11:55 | |
| *** yfaykya has quit IRC | 11:56 | |
| *** yfaykya has joined #tripleo | 11:57 | |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Factor out get kernel image https://review.openstack.org/98756 | 11:58 |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Add get-kernel-image element https://review.openstack.org/98757 | 11:58 |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Use common element get-image-kernel https://review.openstack.org/98758 | 11:58 |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Improve kernel detection in debian family systems https://review.openstack.org/98759 | 11:58 |
| *** tzumainn has joined #tripleo | 12:00 | |
| *** yfaykya has quit IRC | 12:01 | |
| *** tzumainn has quit IRC | 12:02 | |
| *** tzumainn has joined #tripleo | 12:02 | |
| *** dprince has joined #tripleo | 12:02 | |
| *** yfaykya has joined #tripleo | 12:03 | |
| *** jdob has joined #tripleo | 12:07 | |
| *** giulivo has joined #tripleo | 12:07 | |
| *** yfaykya has quit IRC | 12:08 | |
| *** yfaykya has joined #tripleo | 12:09 | |
| *** rbrady has joined #tripleo | 12:13 | |
| *** yfaykya has quit IRC | 12:14 | |
| *** yfaykya has joined #tripleo | 12:15 | |
| *** yfaykya has quit IRC | 12:20 | |
| *** yfaykya has joined #tripleo | 12:21 | |
| *** yfaykya has quit IRC | 12:26 | |
| *** yfaykya has joined #tripleo | 12:27 | |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Set glance registry host from heat metadata https://review.openstack.org/98773 | 12:31 |
| *** yfaykya has quit IRC | 12:32 | |
| *** yfaykya has joined #tripleo | 12:34 | |
| *** yfaykya has quit IRC | 12:38 | |
| *** yfaykya has joined #tripleo | 12:40 | |
| *** ccrouch has left #tripleo | 12:44 | |
| *** yfaykya has quit IRC | 12:45 | |
| *** yfaykya has joined #tripleo | 12:46 | |
| *** yfaykya has quit IRC | 12:51 | |
| *** lucas-hungry is now known as lucasagomes | 12:52 | |
| *** yfaykya has joined #tripleo | 12:52 | |
| d0ugal | tzumainn: hi | 12:52 |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-heat-templates: Use virtual ip for OS services https://review.openstack.org/98778 | 12:52 |
| tzumainn | d0ugal, hiya! | 12:52 |
| openstackgerrit | A change was merged to openstack-infra/tripleo-ci: Create a dummy device on TE hosts https://review.openstack.org/91333 | 12:53 |
| d0ugal | tzumainn: I'm just getting my head around your comment. | 12:53 |
| tzumainn | d0ugal, it's just my understanding of what's going on, it's very possible others disagree | 12:53 |
| d0ugal | tzumainn: So - provider resources are fine, configurations files should be changed to environment files? and I need to add the master template concept? | 12:53 |
| tzumainn | d0ugal, I *think* so, but maybe discuss with jdob | 12:54 |
| tzumainn | I just wanted to state my understanding in one place rather than have it scattered in a hundred different comments | 12:54 |
| d0ugal | Sure | 12:54 |
| tzumainn | i.e., give people one place to disagree : ) | 12:54 |
| d0ugal | and I want to make sure I understand your understanding because I figure out what my understanding is :) | 12:54 |
| tzumainn | lol | 12:54 |
| jdob | yes, the configuration is stored in the environment file and there's a master template that creates resources for the providers defined in the environment | 12:54 |
| jdob | also realize though that, strictly speaking, the environment isn't just a config file as you may be referring to it | 12:55 |
| jdob | (not sure of the exact comment in question here) | 12:55 |
| tzumainn | jdob, https://review.openstack.org/#/c/97553/ | 12:55 |
| d0ugal | jdob: bottom of https://review.openstack.org/#/c/97553/ | 12:55 |
| d0ugal | ah, too slow | 12:55 |
| tzumainn | er, how do I link to a specific comment? | 12:55 |
| tzumainn | jdob, anyway, it's three from the bottom | 12:56 |
| jdob | you misunderstood, I dont want to know the exact comment, I want to get out of this conversation | 12:56 |
| jdob | :) | 12:56 |
| d0ugal | jdob: there is no escape | 12:56 |
| tzumainn | lol | 12:56 |
| *** yfaykya has quit IRC | 12:56 | |
| tzumainn | jdob, you've gotten spicy in your old age | 12:57 |
| jdob | i shouldn't have gone to the doctor and just rode the black plague out from last week | 12:57 |
| d0ugal | lol | 12:57 |
| jdob | i'm behind, four days of pretty much no interaction with people and all the shit talking is coming out at once | 12:57 |
| tzumainn | is it bad that I enjoy your agony? | 12:57 |
| jdob | one issue in this comment: "provider resource template" | 12:57 |
| jdob | its not just a single template | 12:58 |
| jdob | check out the tht repo right now, the config and resource def are split into two files | 12:58 |
| tzumainn | ah, okay | 12:58 |
| d0ugal | always two files? | 12:58 |
| *** yfaykya has joined #tripleo | 12:58 | |
| jdob | i dont think so | 12:58 |
| tzumainn | or is there the potential for more? | 12:58 |
| tzumainn | exciting! | 12:58 |
| tzumainn | makes sense though | 12:58 |
| jdob | i'm trying to read the tht conversion spec now, i'm hoping it's covered in there | 12:58 |
| d0ugal | So, this is the relationship thing? | 12:58 |
| d0ugal | so, this might be when we want shadower to join us :) | 12:59 |
| *** julim has joined #tripleo | 13:01 | |
| shadower | m'ere | 13:02 |
| shadower | d0ugal: I think I'm missing some context | 13:02 |
| tzumainn | shadower, we're trying to understand what it means to store "provider resource template(s)" | 13:02 |
| tzumainn | what exactly might that entail? jdob does some clarification above ^ | 13:03 |
| *** yfaykya has quit IRC | 13:03 | |
| d0ugal | shadower: I think the main issue is that we could have more than one and we are not sure how they are related. | 13:03 |
| *** yfaykya has joined #tripleo | 13:04 | |
| shadower | right, so | 13:04 |
| shadower | in general, you'd have 1 provider template per role. So e.g. nova-compute.yaml, controller.yaml, object-store.yaml, etc. | 13:05 |
| shadower | then you have one template that references puts tham all together (overcloud.yaml) | 13:05 |
| shadower | and environment.yaml which is where the role -> resource type mapping (and possibly parameters) go | 13:06 |
| shadower | s/parameters/parameter values/ | 13:06 |
| tzumainn | shadower, but jdob is saying that a role might be captured in multiple templates | 13:06 |
| shadower | oh | 13:06 |
| shadower | missed that bit | 13:06 |
| jdob | shadower: referring to the split between the resource def and the software config | 13:06 |
| shadower | jdob: you mean one role could have multiple configs and we might want to put each config to its own file for better reuse? | 13:07 |
| jdob | not exactly, I'm referring to the difference between swift-deploy.yaml and swift-source.yaml | 13:08 |
| *** yfaykya has quit IRC | 13:09 | |
| shadower | I'm not fond of that setup. I'd prefer something like swift-source with the OS::Heat::Server and a corresponding deploy resource(s) and swift-config.yaml that contains the software config (and nothing else) | 13:10 |
| *** yfaykya has joined #tripleo | 13:10 | |
| shadower | jdob: I don't see any value in splitting deploys and servers (since they're closely tied to each other). Reusable configs are a different matter | 13:11 |
| tzumainn | shadower, but for purposes of storage - would it be correct to say that a role could be captured in multiple template files? | 13:11 |
| openstackgerrit | Alexis Lee proposed a change to openstack/tripleo-incubator: Allow overcloud fixed range to be configurable https://review.openstack.org/97626 | 13:11 |
| shadower | tzumainn: yeah I would say so | 13:12 |
| jdob | shadower: ok, so what I see in tht with respect to that is changing? | 13:12 |
| d0ugal | and you'd want to pull them all out together? i.e. give me everything for 'swift'? | 13:12 |
| shadower | tzumainn: or rather, depends how you define "role" | 13:12 |
| shadower | :-) | 13:12 |
| tzumainn | shadower, compute, controller, object-store, etc :P | 13:12 |
| tzumainn | shadower, is there one "master template" in those cases that would reference everything else being used? | 13:13 |
| tzumainn | or would the various template files be independent? | 13:13 |
| shadower | tzumainn: I mean: you could say that role is just compute.yaml and then store configs separately. But the role would still reference & use the configs | 13:13 |
| tzumainn | shadower, okay, that makes sense | 13:13 |
| jdob | i'm hoping that the tuskar definition of a role is conceptual, but maps to all of the real pieces underneath | 13:14 |
| tzumainn | so it sounds like in general, when we talk about a template, it has the potential to refer to other templates, those other templates may refer to more templates, etc | 13:14 |
| tzumainn | like a tree | 13:14 |
| jdob | so if I refer to the compute "role" as a user, Tuskar will translate if that means 1 file or 7 | 13:14 |
| tzumainn | but there's always one primary template from which everything else flows? | 13:14 |
| shadower | tzumainn: each role should be standalone (heat stack-create -f compute.yaml + the assorted files should work) but we'd have a master template for the overcloud | 13:14 |
| shadower | tzumainn: yea, it can be a tree | 13:14 |
| shadower | tzumainn: but I don't think the storage system needs to capture that logic | 13:14 |
| shadower | i.e. you should be able to just store a bunch of files | 13:15 |
| tzumainn | shadower, so you don't think it needs to capture the relationship(s) between the templates? | 13:15 |
| *** yfaykya has quit IRC | 13:15 | |
| shadower | tzumainn: I don't see why it should. The logic is captured within the templates themselves | 13:15 |
| jdob | shadower: but the problem comes in asking for them | 13:15 |
| jdob | so if I ask for a list of roles | 13:15 |
| shadower | the storage could store the whole bundle as .tar.gz for all we care | 13:15 |
| jdob | that's a lot of inspection | 13:16 |
| shadower | hm | 13:16 |
| shadower | so | 13:16 |
| jdob | that's why I think the tuskar storage layer isn't just a generic file storage mechanism | 13:16 |
| jdob | because it has to be optimized for certain questions | 13:16 |
| jdob | and since it's not a DB, it's not as simple as writing queries | 13:16 |
| *** yfaykya has joined #tripleo | 13:16 | |
| d0ugal | If only it was a db... oh wait | 13:17 |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Don't use deprecated keystone module https://review.openstack.org/98705 | 13:17 |
| shadower | I think the canonical operation should be "give me the overcloud I set up last Thursday" which should spit out all the templates you need for launching that thing | 13:17 |
| shadower | immutable & versioned (ideally) | 13:17 |
| shadower | if different overclouds reference the same file in the background, that's fine | 13:17 |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Silence the deprecation warning about with statements https://review.openstack.org/98703 | 13:17 |
| d0ugal | jdob: It would be useful to build up a list of expected queries/questions | 13:18 |
| tzumainn | so. . . can a role be stored in some sort of folder construct | 13:18 |
| jdob | that's what I thought your spec was doing (not to sound like an ass) | 13:18 |
| jdob | but that's why I was expecting to see domain-specific APIs | 13:18 |
| jdob | that defined those questions | 13:18 |
| tzumainn | and we just stick all the templates into that folder? | 13:18 |
| jdob | list_roles instead of get(filename) | 13:18 |
| d0ugal | jdob: well, yes, but I mean from you as somebody that plans to use it | 13:18 |
| shadower | tzumainn: it could be but it's not at the moment and I'm not sure it should | 13:19 |
| shadower | (stored in a folder structure) | 13:19 |
| openstackgerrit | therese-mchale proposed a change to openstack/tripleo-image-elements: Use Percona Packages for Ubuntu https://review.openstack.org/90134 | 13:19 |
| *** morazi has joined #tripleo | 13:19 | |
| jdob | d0ugal: i hear ya, it's just a matter of time. faster for you to think it through and me to bitch it's not what I expected :) | 13:19 |
| tzumainn | shadower, what would be a reasonable alternative? | 13:19 |
| jdob | in all seriousness, I see your point, i just dont totally have the bandwidth to write it out all spec-like | 13:19 |
| d0ugal | jdob: sure, that's where I'm at I guess. | 13:19 |
| d0ugal | jdob: I didn't mean spec-like, just an ether or whatever - doesn't matter, it was just an idea. | 13:20 |
| jdob | no no, it was a valid one, don't let my punchiness today get in the way of that | 13:20 |
| *** yfaykya has quit IRC | 13:21 | |
| *** yfaykya has joined #tripleo | 13:23 | |
| *** jang1 has quit IRC | 13:24 | |
| *** matty_dubs|gone is now known as matty_dubs | 13:24 | |
| shadower | tzumainn: treat a deployment unit (e.g. the undercloud) as a set of templates that go together. No further structure needed (as far as I can see) | 13:25 |
| *** ccrouch has joined #tripleo | 13:25 | |
| tzumainn | shadower, the structure is needed because a role might be used in multiple overclouds | 13:25 |
| tzumainn | so we want to separate out "role templates" so we can stick them in the correct deployment units | 13:25 |
| shadower | tzumainn: but roles are a tripleo-specific concept. It's not necessarily useful to anyone else | 13:26 |
| shadower | the current direction seems to be towards tuskar being useful for describing othre things than openstack | 13:26 |
| jdob | not sure I'd say it's only a tripleo-specfic concept | 13:27 |
| shadower | tzumainn: you should still be able to reuse files without introducing higher-level constructs such as roles | 13:27 |
| jdob | in any large-scale app, you might want to split your templates into multiple files like that | 13:27 |
| *** yfaykya has quit IRC | 13:27 | |
| jdob | and have a logical concept to refer to | 13:27 |
| shadower | yeah I guess | 13:27 |
| shadower | tzumainn: btw I misspoke earlier, it's not a tree it's a directed acyclic graph | 13:28 |
| shadower | (two files could reference the same file) | 13:28 |
| tzumainn | shadower, yeah, you're right | 13:28 |
| jdob | that sounds so much more intimidating than "tree" :) | 13:28 |
| d0ugal | lol | 13:28 |
| tzumainn | sounds like it's time for jdob to pretend to be sick again! | 13:29 |
| shadower | hence my preference to store them just as bundles | 13:29 |
| shadower | works for git, should work for us, too | 13:29 |
| *** yfaykya has joined #tripleo | 13:29 | |
| tzumainn | shadower, right, I guess my argument is just that it's useful to store the bundles at a 'role' level | 13:29 |
| d0ugal | tzumainn: Why? | 13:29 |
| jdob | how else would you store them? | 13:29 |
| jdob | the nice part about what tzumainn is saying is that the list of bundles is effectively the list of roles | 13:30 |
| jdob | i want the "compute" role, Tuskar pulls the compute bundle and starts handling it | 13:30 |
| d0ugal | bundle being a number of templates? | 13:30 |
| *** ci-overcloud has joined #tripleo | 13:31 | |
| ci-overcloud | ************** ci-overcloud complete status=127 ************ | 13:31 |
| *** ci-overcloud has quit IRC | 13:31 | |
| jdob | to be honest, I dunno | 13:31 |
| jdob | it kinda depends on what THT does | 13:31 |
| *** howleyt has quit IRC | 13:32 | |
| jdob | originally, I assumed each "role" would be split into a single template | 13:32 |
| jdob | but that's not what I'm seeing in the tht repo right now | 13:32 |
| shadower | yea me too | 13:32 |
| jdob | the tht spec shadower wrote seems to allude to a single file | 13:32 |
| jdob | and I want to hash out in that spec if that's the plan or not | 13:32 |
| jdob | a lot is contingent on that | 13:32 |
| shadower | jdob: I think I'll have to amend it | 13:32 |
| d0ugal | ok | 13:32 |
| jdob | which is why I was so obnoxious in pushing for that spec to exist | 13:32 |
| * shadower doesn't think we can rely on role/template having a 1-1 mapping | 13:33 | |
| jdob | because it's hard to guess at given the current THT repo | 13:33 |
| *** yfaykya has quit IRC | 13:33 | |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Use common element get-image-kernel https://review.openstack.org/98758 | 13:33 |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Improve kernel detection in debian family systems https://review.openstack.org/98759 | 13:33 |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Factor out get kernel image https://review.openstack.org/98756 | 13:34 |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Add get-kernel-image element https://review.openstack.org/98757 | 13:34 |
| d0ugal | Okay, I'm going to go and spend a bit of time properly grokking that spec | 13:34 |
| *** ramishra has quit IRC | 13:34 | |
| shadower | fwiw, my concern is this: all the structure necessary to figure out which templates to send to heat is in the template files themselves | 13:35 |
| *** yfaykya has joined #tripleo | 13:35 | |
| shadower | in theory all you need is the root template | 13:35 |
| shadower | so I'm wary of duplicating or restricting that logic higher up | 13:35 |
| jdob | understood, and that's awesome | 13:35 |
| jdob | the only reason I'd say to duplicate | 13:35 |
| jdob | and i wouldnt even use that word | 13:35 |
| d0ugal | shadower: How does the root template refer to them? does it assume they are all on local disk? | 13:35 |
| jdob | cache might be more appropriate | 13:35 |
| jdob | is so that we can more easily ask the sort of tuskar questions we know we'll need to regularly ask | 13:36 |
| shadower | d0ugal: via URIs | 13:36 |
| jdob | without loading everything up and parsing it every time | 13:36 |
| shadower | d0ugal: they could theoretically be at different servers | 13:36 |
| shadower | jdob: I'm fine with caching. But having a standalone Role resource (resource as in REST) at the *storage* level would limit things ihmo | 13:37 |
| shadower | I'm perfectly fine with roles in Tuskar | 13:37 |
| shadower | but I'd like the storage to look more like git | 13:38 |
| shadower | i.e. a bunch of templates that reference each other | 13:38 |
| d0ugal | shadower: Lots of one to one relationships? | 13:38 |
| shadower | caching is fine but it'd be a transparent implementation detail kind of thing, rather than a bit of logic that we have to explicitly deal with | 13:38 |
| shadower | d0ugal: ya | 13:38 |
| shadower | which is what t-h-t does now and will do in the future | 13:39 |
| d0ugal | Interesting. | 13:39 |
| jdob | hrm, ok, that's fair. I'm approaching this very much as the dude writing the service code that needs to call into storage | 13:39 |
| shadower | d0ugal: in theory, you wouldn't even need to store the 1-1 relationships (if you could live with parsing the templates) | 13:39 |
| shadower | but that's where the caching comes in I think | 13:39 |
| d0ugal | shadower: That's the relationship model being added to Glance Artifacts | 13:39 |
| *** yfaykya has quit IRC | 13:39 | |
| *** pcrews_ has joined #tripleo | 13:40 | |
| shadower | jdob, d0ugal for the record, chances are that I won't be touching either side of that code much (storage and api) so whatever works for you guys is ultimately fine | 13:40 |
| d0ugal | ha :) | 13:40 |
| shadower | just saying what feels right to me wrt the way heat and t-h-t work | 13:40 |
| jdob | i'd like to take a similar approach, I like the idea of the service saying what it needs and storage figures it out however they want | 13:41 |
| jdob | shadower: still, totally appreciate your input | 13:41 |
| *** yfaykya has joined #tripleo | 13:41 | |
| jdob | you have a better idea of the templates so it's valuable to hear | 13:41 |
| d0ugal | shadower: yup, thanks! | 13:41 |
| *** yfaykya has quit IRC | 13:45 | |
| *** yfaykya has joined #tripleo | 13:47 | |
| *** ifarkas has quit IRC | 13:47 | |
| *** e0ne has joined #tripleo | 13:51 | |
| openstackgerrit | Jan Provaznik proposed a change to openstack/tripleo-image-elements: Update openstack services to listen on stunnel connect port https://review.openstack.org/61376 | 13:51 |
| *** yfaykya has quit IRC | 13:51 | |
| *** yfaykya has joined #tripleo | 13:53 | |
| *** howleyt has joined #tripleo | 13:55 | |
| *** yfaykya has quit IRC | 13:57 | |
| *** yfaykya has joined #tripleo | 13:59 | |
| *** akrivoka has quit IRC | 14:00 | |
| *** beekneemech is now known as bnemec | 14:02 | |
| ccrouch | ping giulivo | 14:04 |
| *** yfaykya has quit IRC | 14:04 | |
| *** rwsu has joined #tripleo | 14:04 | |
| *** yfaykya has joined #tripleo | 14:05 | |
| openstackgerrit | Matthew Macdonald-Wallace proposed a change to openstack/tripleo-image-elements: Install check_mk server https://review.openstack.org/87226 | 14:06 |
| openstackgerrit | Matthew Macdonald-Wallace proposed a change to openstack/tripleo-image-elements: Auto-collect local checks https://review.openstack.org/91324 | 14:09 |
| *** akrivoka has joined #tripleo | 14:09 | |
| *** yfaykya has quit IRC | 14:10 | |
| *** yfaykya has joined #tripleo | 14:11 | |
| *** yfaykya has quit IRC | 14:16 | |
| *** yfaykya has joined #tripleo | 14:18 | |
| *** ramishra has joined #tripleo | 14:18 | |
| *** e0ne has quit IRC | 14:20 | |
| *** yfaykya has quit IRC | 14:22 | |
| *** ci-overcloud has joined #tripleo | 14:23 | |
| ci-overcloud | ************** ci-overcloud complete status=127 ************ | 14:23 |
| *** ci-overcloud has quit IRC | 14:23 | |
| *** yfaykya has joined #tripleo | 14:24 | |
| *** jcoufal has quit IRC | 14:24 | |
| *** ramishra has quit IRC | 14:25 | |
| *** spiffxp_ has joined #tripleo | 14:25 | |
| *** SlickN1k has joined #tripleo | 14:26 | |
| openstackgerrit | Alexis Lee proposed a change to openstack-infra/tripleo-ci: Split very long line in toci_devtest.sh https://review.openstack.org/97429 | 14:27 |
| *** ramishra has joined #tripleo | 14:27 | |
| *** yfaykya has quit IRC | 14:29 | |
| *** jistr has quit IRC | 14:29 | |
| *** rdopiera has quit IRC | 14:30 | |
| *** yfaykya has joined #tripleo | 14:30 | |
| *** akrivoka has quit IRC | 14:31 | |
| *** akrivoka has joined #tripleo | 14:31 | |
| *** jistr has joined #tripleo | 14:32 | |
| *** ramishra has quit IRC | 14:34 | |
| *** yfaykya has quit IRC | 14:34 | |
| *** yfaykya has joined #tripleo | 14:36 | |
| *** yfaykya has quit IRC | 14:41 | |
| *** yfaykya has joined #tripleo | 14:42 | |
| *** yfaykya has quit IRC | 14:47 | |
| *** yfaykya has joined #tripleo | 14:48 | |
| *** yfaykya has quit IRC | 14:53 | |
| *** e0ne has joined #tripleo | 14:53 | |
| *** yfaykya has joined #tripleo | 14:54 | |
| vinsh | hey.. whats the "key" to get the cluster virtualIP .. | 14:58 |
| openstackgerrit | therese-mchale proposed a change to openstack/tripleo-image-elements: Use Percona Packages for Ubuntu https://review.openstack.org/90134 | 14:58 |
| vinsh | say you are on the undercloud.. or the tripleo-root.. and want to find the overcloud virtual ip | 14:58 |
| vinsh | so that you can point devtest_overcloud at that.. vs the overcloud controller ip | 14:58 |
| *** yfaykya has quit IRC | 14:59 | |
| *** untriaged-bot has joined #tripleo | 15:00 | |
| untriaged-bot | Untriaged bugs so far: | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1325762 | 15:00 |
| *** jistr has quit IRC | 15:00 | |
| uvirtbot | Launchpad bug 1325762 in tripleo "intermittent tftp issues using physical hardware" [Undecided,New] | 15:00 |
| *** untriaged-bot has quit IRC | 15:00 | |
| *** tzumainn has quit IRC | 15:00 | |
| *** jistr has joined #tripleo | 15:00 | |
| *** yfaykya has joined #tripleo | 15:00 | |
| *** tzumainn has joined #tripleo | 15:01 | |
| *** ci-overcloud has joined #tripleo | 15:02 | |
| ci-overcloud | ************** ci-overcloud complete status=1 ************ | 15:02 |
| *** ci-overcloud has quit IRC | 15:02 | |
| *** yfaykya has quit IRC | 15:05 | |
| *** yfaykya has joined #tripleo | 15:07 | |
| vinsh | greghaynes, related to change 93554... haproxy.. any idea? | 15:08 |
| vinsh | how do we query the virtual IP from the overcloud? | 15:08 |
| vinsh | I want to wedge that into devtest_overcloud.. vs the controller IP | 15:08 |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Remove temporary nova revert https://review.openstack.org/98816 | 15:10 |
| *** yfaykya has quit IRC | 15:11 | |
| *** yfaykya has joined #tripleo | 15:13 | |
| *** jprovazn is now known as jprovazn_afk | 15:16 | |
| *** yfaykya has quit IRC | 15:17 | |
| jprovazn_afk | vinsh, this is addressed in this patch: https://review.openstack.org/#/c/94354/1/scripts/devtest_overcloud.sh | 15:18 |
| openstackgerrit | Andrea Frittoli proposed a change to openstack/tripleo-image-elements: Configurable tests2skip file https://review.openstack.org/98817 | 15:18 |
| *** morganfainberg_Z is now known as morganfainberg | 15:18 | |
| vinsh | jprovazn_afk, BINGO :) ty. | 15:19 |
| *** yfaykya has joined #tripleo | 15:19 | |
| jprovazn_afk | vinsh, np | 15:19 |
| openstackgerrit | Andrea Frittoli proposed a change to openstack/tripleo-image-elements: Configurable tests2skip file https://review.openstack.org/98817 | 15:20 |
| *** yfaykya has quit IRC | 15:23 | |
| *** julim has quit IRC | 15:24 | |
| *** yfaykya has joined #tripleo | 15:25 | |
| openstackgerrit | Ben Nemec proposed a change to openstack/tripleo-specs: TripleO on OpenStack https://review.openstack.org/92642 | 15:28 |
| *** rlandy_ has joined #tripleo | 15:29 | |
| *** yfaykya has quit IRC | 15:30 | |
| *** yfaykya has joined #tripleo | 15:31 | |
| *** rlandy has quit IRC | 15:32 | |
| *** russell_h has quit IRC | 15:32 | |
| *** russell_h has joined #tripleo | 15:32 | |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: Add keepalive option to rabbitmq config https://review.openstack.org/81920 | 15:33 |
| *** yfaykya has quit IRC | 15:36 | |
| *** yfaykya has joined #tripleo | 15:37 | |
| openstackgerrit | therese-mchale proposed a change to openstack/tripleo-image-elements: Use Percona Packages for Ubuntu https://review.openstack.org/90134 | 15:40 |
| *** jcoufal has joined #tripleo | 15:41 | |
| *** yfaykya has quit IRC | 15:42 | |
| *** yfaykya has joined #tripleo | 15:43 | |
| *** geerdest has joined #tripleo | 15:44 | |
| *** athomas has quit IRC | 15:47 | |
| *** yfaykya has quit IRC | 15:48 | |
| *** yfaykya has joined #tripleo | 15:49 | |
| *** e0ne has quit IRC | 15:53 | |
| *** athomas has joined #tripleo | 15:53 | |
| *** e0ne has joined #tripleo | 15:54 | |
| *** yfaykya has quit IRC | 15:55 | |
| *** yfaykya has joined #tripleo | 15:56 | |
| *** matty_dubs is now known as matty_dubs|lunch | 15:57 | |
| *** e0ne has quit IRC | 15:58 | |
| *** andreaf has quit IRC | 16:00 | |
| *** yfaykya has quit IRC | 16:00 | |
| *** yfaykya has joined #tripleo | 16:02 | |
| *** IvanBerezovskiy has left #tripleo | 16:04 | |
| *** yfaykya has quit IRC | 16:06 | |
| *** yfaykya has joined #tripleo | 16:08 | |
| *** yamahata has joined #tripleo | 16:10 | |
| *** yfaykya has quit IRC | 16:12 | |
| *** yfaykya has joined #tripleo | 16:14 | |
| *** eghobo has joined #tripleo | 16:17 | |
| *** yfaykya has quit IRC | 16:19 | |
| *** yfaykya has joined #tripleo | 16:20 | |
| *** yamahata has quit IRC | 16:22 | |
| *** yamahata has joined #tripleo | 16:22 | |
| *** ifarkas has joined #tripleo | 16:24 | |
| vinsh | Interesting.. that the keystone url returned by heat output-show... isn't using the virtualIP in my case | 16:24 |
| *** yfaykya has quit IRC | 16:24 | |
| *** sballe has joined #tripleo | 16:25 | |
| *** spzala has joined #tripleo | 16:25 | |
| *** pblaho has quit IRC | 16:26 | |
| *** yfaykya has joined #tripleo | 16:26 | |
| vinsh | oh neat.. I see where heat output-show comes from now. right on. | 16:27 |
| vinsh | "Outputs:" | 16:27 |
| *** akuznetsov has quit IRC | 16:31 | |
| *** yfaykya has quit IRC | 16:31 | |
| *** yfaykya has joined #tripleo | 16:32 | |
| *** yamahata has quit IRC | 16:35 | |
| *** yamahata has joined #tripleo | 16:35 | |
| *** yfaykya has quit IRC | 16:37 | |
| *** yfaykya has joined #tripleo | 16:38 | |
| *** akuznetsov has joined #tripleo | 16:39 | |
| *** lucasagomes has quit IRC | 16:39 | |
| *** lucasagomes has joined #tripleo | 16:39 | |
| *** jcoufal has quit IRC | 16:40 | |
| *** rcarrill` has joined #tripleo | 16:42 | |
| *** akuznetsov has quit IRC | 16:42 | |
| lifeless | o/ | 16:42 |
| lifeless | vinsh: hi | 16:43 |
| lifeless | vinsh: I missed you before, haproxy change? | 16:43 |
| vinsh | Hey there | 16:43 |
| *** yfaykya has quit IRC | 16:43 | |
| vinsh | I think things make good enough sense here.. just that my output from heat output-show for the keystone url.. does not match what I expect | 16:44 |
| *** rcarrillocruz has quit IRC | 16:44 | |
| vinsh | where for example that same variable "ControllerVirtualIP" when referenced by the haproxy element does infact lay down the virtual ip | 16:44 |
| *** jistr has quit IRC | 16:44 | |
| vinsh | that same output from heat output-show though.. shows it to be the controller ip | 16:44 |
| *** yfaykya has joined #tripleo | 16:45 | |
| vinsh | Not sure why.. as in the overcloud-source heat template.. they are called the same way for both items. | 16:45 |
| lifeless | hmm could be a bug in the template | 16:45 |
| *** matty_dubs|lunch is now known as matty_dubs | 16:46 | |
| vinsh | also.. the net_binds changes to haproxy can be a nightmare :) | 16:46 |
| vinsh | really.. you use net_binds or not | 16:46 |
| vinsh | there is not clean way to override like the readme claims | 16:46 |
| derekh_ | lifeless: I managed to get the heat stack to go into complete state, but br-untagged doesn't seem to be getting an IP address, I've added some details to the bottom of the etherpad | 16:47 |
| vinsh | Though, I am doing evil things here.. with controller nodes that are not homogeneous (hosting different api services on each one) | 16:47 |
| vinsh | so, no net_binds for me. | 16:48 |
| derekh_ | lifeless: the undercloud doesn't seem to be seeing dhcp request from the ci-overcloud , I havn't been able to figure out why | 16:48 |
| lifeless | vinsh: I haven't read the recent haproxy changes | 16:49 |
| *** yfaykya has quit IRC | 16:49 | |
| vinsh | better for your sanity :) | 16:49 |
| lifeless | vinsh: we should still totally be able to do haproxy->other-node balancing | 16:49 |
| vinsh | oh I am | 16:49 |
| lifeless | vinsh: if we can't, thats a huge bug | 16:50 |
| vinsh | without use of net_binds | 16:50 |
| vinsh | net_binds causes mustache overload | 16:50 |
| vinsh | ive got haproxy on control node, sending traffic to another control node that hosts ceilometer. and also sending traffic to swift proxy nodes | 16:51 |
| *** yfaykya has joined #tripleo | 16:51 | |
| vinsh | which are spit out. | 16:51 |
| vinsh | Just futzing with this keystone url.. which isn't right yet. | 16:51 |
| *** spzala has quit IRC | 16:54 | |
| *** yfaykya has quit IRC | 16:55 | |
| *** yfaykya has joined #tripleo | 16:57 | |
| *** derekh_ has quit IRC | 16:59 | |
| *** yfaykya has quit IRC | 17:02 | |
| *** yfaykya has joined #tripleo | 17:03 | |
| *** akrivoka has quit IRC | 17:03 | |
| *** yfaykya has quit IRC | 17:07 | |
| *** marun has joined #tripleo | 17:07 | |
| *** yfaykya has joined #tripleo | 17:09 | |
| *** marun has quit IRC | 17:11 | |
| *** yfaykya has quit IRC | 17:14 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: indent using 4 spaces (1/3) https://review.openstack.org/93206 | 17:14 |
| *** yfaykya has joined #tripleo | 17:15 | |
| *** yfaykya has quit IRC | 17:20 | |
| *** yfaykya has joined #tripleo | 17:21 | |
| *** yfaykya has quit IRC | 17:26 | |
| lifeless | mordred: while you are around: https://review.openstack.org/#/c/97583/ | 17:27 |
| *** eghobo has quit IRC | 17:27 | |
| *** nati_ueno has joined #tripleo | 17:27 | |
| *** yfaykya has joined #tripleo | 17:27 | |
| *** spzala has joined #tripleo | 17:29 | |
| *** yfaykya has quit IRC | 17:32 | |
| *** yfaykya has joined #tripleo | 17:33 | |
| *** russell_h has quit IRC | 17:36 | |
| lifeless | yayayaya the nova revert landed | 17:38 |
| *** yfaykya has quit IRC | 17:38 | |
| *** russell_h has joined #tripleo | 17:39 | |
| *** ramishra has joined #tripleo | 17:39 | |
| *** yfaykya has joined #tripleo | 17:40 | |
| *** rcarrill` has quit IRC | 17:40 | |
| *** rcarrillocruz has joined #tripleo | 17:41 | |
| *** russell_h has quit IRC | 17:41 | |
| *** russell_h has joined #tripleo | 17:41 | |
| *** lucasagomes is now known as lucas-afk | 17:44 | |
| *** yfaykya has quit IRC | 17:44 | |
| *** eghobo has joined #tripleo | 17:45 | |
| *** yfaykya has joined #tripleo | 17:46 | |
| *** ramishra has quit IRC | 17:48 | |
| *** ramishra has joined #tripleo | 17:49 | |
| *** yfaykya has quit IRC | 17:50 | |
| *** yfaykya has joined #tripleo | 17:52 | |
| *** ramishra has quit IRC | 17:53 | |
| *** yfaykya has quit IRC | 17:57 | |
| *** yfaykya has joined #tripleo | 17:58 | |
| *** saurabhs has joined #tripleo | 17:59 | |
| *** ramishra has joined #tripleo | 18:00 | |
| *** jang has joined #tripleo | 18:01 | |
| *** e0ne has joined #tripleo | 18:01 | |
| *** yfaykya has quit IRC | 18:03 | |
| *** yfaykya has joined #tripleo | 18:04 | |
| *** ramishra has quit IRC | 18:05 | |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Replace nagios3 template with HOT rewrite https://review.openstack.org/90665 | 18:06 |
| *** e0ne has quit IRC | 18:07 | |
| *** eguz has joined #tripleo | 18:07 | |
| *** eguz has quit IRC | 18:07 | |
| *** eguz has joined #tripleo | 18:07 | |
| *** yfaykya has quit IRC | 18:09 | |
| *** yfaykya has joined #tripleo | 18:10 | |
| *** eghobo has quit IRC | 18:11 | |
| *** yfaykya has quit IRC | 18:15 | |
| *** yfaykya has joined #tripleo | 18:16 | |
| *** athomas has quit IRC | 18:16 | |
| openstackgerrit | Ghe Rivero proposed a change to openstack-infra/tripleo-ci: Add /mnt/state/var/log to the host state files https://review.openstack.org/93627 | 18:19 |
| *** yfaykya has quit IRC | 18:21 | |
| *** yfaykya has joined #tripleo | 18:22 | |
| *** pelix has quit IRC | 18:26 | |
| *** e0ne has joined #tripleo | 18:26 | |
| *** yfaykya has quit IRC | 18:27 | |
| *** yfaykya has joined #tripleo | 18:29 | |
| *** nati_ueno has quit IRC | 18:31 | |
| *** yfaykya has quit IRC | 18:33 | |
| *** nati_uen_ has joined #tripleo | 18:34 | |
| *** yfaykya has joined #tripleo | 18:35 | |
| *** yfaykya has quit IRC | 18:39 | |
| *** e0ne has quit IRC | 18:39 | |
| *** e0ne has joined #tripleo | 18:40 | |
| *** yfaykya has joined #tripleo | 18:41 | |
| *** yfaykya has quit IRC | 18:45 | |
| *** yfaykya has joined #tripleo | 18:47 | |
| *** spzala has quit IRC | 18:48 | |
| *** yfaykya has quit IRC | 18:52 | |
| *** yfaykya has joined #tripleo | 18:53 | |
| *** weshay has quit IRC | 18:54 | |
| *** lucas-afk is now known as lucasagomes | 18:56 | |
| *** yfaykya has quit IRC | 18:58 | |
| *** yfaykya has joined #tripleo | 18:59 | |
| *** nati_ueno has joined #tripleo | 19:02 | |
| *** yfaykya has quit IRC | 19:03 | |
| *** yfaykya has joined #tripleo | 19:05 | |
| *** nati_uen_ has quit IRC | 19:06 | |
| *** rob_s_parker has quit IRC | 19:07 | |
| *** weshay has joined #tripleo | 19:09 | |
| *** rcarrill` has joined #tripleo | 19:10 | |
| *** yfaykya has quit IRC | 19:10 | |
| *** yfaykya has joined #tripleo | 19:11 | |
| *** rcarrillocruz has quit IRC | 19:12 | |
| *** jang1 has joined #tripleo | 19:13 | |
| *** spzala has joined #tripleo | 19:15 | |
| *** nati_ueno has quit IRC | 19:15 | |
| *** dprince has quit IRC | 19:15 | |
| *** yfaykya has quit IRC | 19:16 | |
| *** yfaykya has joined #tripleo | 19:18 | |
| jogo | lifeless: wow http://logs.openstack.org/08/96308/8/check-tripleo/check-tripleo-undercloud-precise/a879410/ logs are super sparse | 19:20 |
| jogo | nova logs are useless | 19:20 |
| greghaynes | I think thats a somewhat recent change - just noticed when the nova bm scheduler issue came up that I didnt have tracebacks without turning debugging on | 19:21 |
| greghaynes | definitely wasnt the case a while ago | 19:21 |
| lifeless | jogo: yes, and this is why we run default, so you well, know. | 19:22 |
| jogo | greghaynes: hrm, I think sdague looked into that recently | 19:22 |
| *** yfaykya has quit IRC | 19:22 | |
| jogo | lifeless: I added this to the list to investigate this | 19:23 |
| *** yfaykya has joined #tripleo | 19:24 | |
| *** chuckC has quit IRC | 19:24 | |
| *** akuznetsov has joined #tripleo | 19:25 | |
| *** chuckC has joined #tripleo | 19:25 | |
| *** chuckC has quit IRC | 19:25 | |
| *** akuznetsov has quit IRC | 19:25 | |
| greghaynes | dshulyak: jprovazn_afk: Any idea if https://review.openstack.org/#/c/92550/ is planning on being resurrected? Seems like it has some pretty bad conflicts going on... | 19:26 |
| *** akuznetsov has joined #tripleo | 19:26 | |
| *** akuznetsov has quit IRC | 19:27 | |
| *** e0ne has quit IRC | 19:27 | |
| *** akuznetsov has joined #tripleo | 19:28 | |
| *** yfaykya has quit IRC | 19:28 | |
| *** yfaykya has joined #tripleo | 19:30 | |
| *** yfaykya has quit IRC | 19:34 | |
| *** e0ne has joined #tripleo | 19:35 | |
| *** yfaykya has joined #tripleo | 19:36 | |
| *** yfaykya has quit IRC | 19:41 | |
| *** yfaykya has joined #tripleo | 19:42 | |
| vinsh | lifeless, dunno what the deal was.. but on another run.. heat output-show gave me the correct ip for keystone (virtualip). | 19:46 |
| *** yfaykya has quit IRC | 19:46 | |
| vinsh | Maybe a one-of failure. maybe not. | 19:46 |
| greghaynes | vinsh: Youre trying to run the ha overcloud patch stack? | 19:46 |
| vinsh | Ive got most of it yeah | 19:46 |
| greghaynes | whats the patches youre using? | 19:47 |
| vinsh | only am working on it in the split out of the CE context. | 19:47 |
| greghaynes | ah | 19:47 |
| vinsh | I'm using the latest haproxy patches from dimitri | 19:47 |
| greghaynes | what about the one I asked about above? | 19:48 |
| greghaynes | Im getting the feeling that is starting to be superceded by some smaller patches | 19:48 |
| vinsh | ah, havn't hit that yet | 19:48 |
| *** yfaykya has joined #tripleo | 19:48 | |
| vinsh | Yeah, issue I have now.. is that haproxy binds to say.. glance api-port | 19:48 |
| greghaynes | yep | 19:48 |
| vinsh | on virtualip:port | 19:48 |
| *** noslzzp has quit IRC | 19:48 | |
| vinsh | so.. I think that might prevent glance from binding on anything:port | 19:49 |
| vinsh | for example. | 19:49 |
| *** noslzzp has joined #tripleo | 19:49 | |
| greghaynes | yep, the services need specific ip:port binds | 19:49 |
| greghaynes | as well | 19:49 |
| vinsh | Looks like the story is to use stunnel for that issue | 19:50 |
| vinsh | right? | 19:50 |
| vinsh | OR modify all of the elements to change their ports. (extra work) | 19:50 |
| greghaynes | vinsh: Have you read the spec on this/ | 19:51 |
| vinsh | I have it up now | 19:51 |
| greghaynes | so services need to bind on localhost only | 19:51 |
| greghaynes | and then stunnel does the vip <-> localhost mapping | 19:51 |
| greghaynes | er not vip | 19:51 |
| greghaynes | node <-> localhost | 19:51 |
| vinsh | OK, got that | 19:52 |
| *** yfaykya has quit IRC | 19:53 | |
| vinsh | where do the ports come from to fill in 92550? | 19:53 |
| vinsh | The ports I see in overcloud-source .. are for haproxy to bind to.. (standard ports) | 19:54 |
| vinsh | for 92550 we expect all service to bind to localhost only | 19:54 |
| vinsh | i need to read that more. | 19:54 |
| *** yfaykya has joined #tripleo | 19:54 | |
| greghaynes | oh derp, ty for pointing that out | 19:54 |
| greghaynes | so that patch should probably be abandonded | 19:54 |
| vinsh | maybe not though.. | 19:55 |
| vinsh | {{#nova.ec2.port}}31 | 19:55 |
| vinsh | ec2_listen_port={{.}}32 | 19:55 |
| vinsh | {{/nova.ec2.port}} | 19:55 |
| *** marun has joined #tripleo | 19:55 | |
| vinsh | oh ya. | 19:55 |
| openstackgerrit | Jay Dobies proposed a change to openstack/tripleo-specs: Spec for Tuskar Juno REST API changes https://review.openstack.org/94720 | 19:55 |
| vinsh | I was thinking IP | 19:55 |
| lifeless | yup | 19:56 |
| vinsh | so what next? | 19:57 |
| lifeless | haproxy is endpoint <-> node; stunnel is node<->localhost | 19:57 |
| lifeless | haproxy should know all nodes | 19:57 |
| lifeless | stunnel only knows localhost | 19:57 |
| lifeless | services only bind to localhost | 19:57 |
| *** lucasagomes is now known as lucas-dinner | 19:57 | |
| greghaynes | AFAICT were lacking a bing all the things to localhost patch | 19:57 |
| vinsh | does a service bound to localhost:port prevent haproxy from binding to virtualip:sameport ? | 19:58 |
| greghaynes | s/bing/bind | 19:58 |
| greghaynes | no | 19:58 |
| vinsh | k | 19:58 |
| lifeless | haproxy should be binding to *:sameport | 19:58 |
| *** yfaykya has quit IRC | 19:58 | |
| vinsh | that would bind it to virtual ip.. and node ip | 19:59 |
| vinsh | and local host? | 19:59 |
| lifeless | it binds it to * | 20:00 |
| lifeless | which is any incoming traffic without a more specific bind, IIRC. | 20:00 |
| *** yfaykya has joined #tripleo | 20:00 | |
| greghaynes | yea, was just looking up to confirm that, but I thought the same | 20:00 |
| greghaynes | that specific binds override wildcard binds | 20:01 |
| vinsh | for the stunnel pipe/piece here.. is that the open review by Stuart? or do you guys have another WIP? | 20:01 |
| vinsh | https://blueprints.launchpad.net/tripleo/+spec/tripleo-ssl-overcloud | 20:02 |
| greghaynes | https://review.openstack.org/#/c/95663 | 20:02 |
| greghaynes | is important | 20:02 |
| vinsh | From what I understand, once a patch comes in to bind services to localhost. the stunnel/ssl stuff is needed as well | 20:02 |
| vinsh | OIC | 20:03 |
| lifeless | greghaynes: speaking of important, where is the is-bootstrap-host tht patch at ? | 20:03 |
| *** julim has joined #tripleo | 20:03 | |
| greghaynes | https://review.openstack.org/#/c/86435/ | 20:03 |
| greghaynes | looks like about to merge once ci checks | 20:03 |
| greghaynes | only 25 revisions later! | 20:04 |
| vinsh | must be bullet proof now 0_o | 20:04 |
| greghaynes | oh, I need to do some rebasing | 20:04 |
| greghaynes | of things depending on it | 20:04 |
| *** spzala has quit IRC | 20:05 | |
| *** yfaykya has quit IRC | 20:05 | |
| vinsh | Would it be helpful for the ha-story if I were to generate a review for moving element api's to bind to localhost? | 20:05 |
| vinsh | or is that duplicate effort? | 20:05 |
| *** morganfainberg is now known as morganfainberg_Z | 20:06 | |
| vinsh | Suppose I will do it to test my local conifg anywho | 20:06 |
| greghaynes | sounds like plan is to bind them in any addr | 20:06 |
| greghaynes | which would be nice if possible - more defaults in configs ftw | 20:06 |
| *** yfaykya has joined #tripleo | 20:06 | |
| vinsh | lifeless, was saying "services only bind to localhost" | 20:07 |
| greghaynes | oh, got that flipped | 20:07 |
| * greghaynes rereads spec | 20:07 | |
| greghaynes | hrm, spec says we bind all the things on specific ip's. | 20:08 |
| *** boris-42 has quit IRC | 20:08 | |
| vinsh | the spec contradicts what we said here. | 20:11 |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-image-elements: Extract mysql reset files from reset-db https://review.openstack.org/93104 | 20:11 |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-image-elements: Add os-is-bootstrap-host element and script https://review.openstack.org/86435 | 20:11 |
| vinsh | has haproxy only bound to a vip, then it sends traffic out to node ip | 20:11 |
| *** yfaykya has quit IRC | 20:11 | |
| vinsh | stunnel listens only on nodeip and sends to localhost | 20:11 |
| vinsh | services only listen on localhost | 20:11 |
| vinsh | no haproxy bound to *. (i'm probably being thick here) | 20:12 |
| *** boris-42 has joined #tripleo | 20:12 | |
| vinsh | Time to test and see! :) | 20:12 |
| jprovazn_afk | greghaynes, I think ports customization is not priority ATM (https://review.openstack.org/#/c/92550/) | 20:12 |
| *** yfaykya has joined #tripleo | 20:13 | |
| greghaynes | yea, looks like that just isnt needed at all now with the spec we went with? | 20:13 |
| *** chuckC has joined #tripleo | 20:14 | |
| devananda | ya'll may want to update the status of https://bugs.launchpad.net/tripleo/+bug/1326289 now that the fix/revert has landed in nova | 20:14 |
| uvirtbot | Launchpad bug 1326289 in nova "Failing to launch instances : Filter ComputeCapabilitiesFilter returned 0 hosts" [Critical,Fix committed] | 20:14 |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-image-elements: Improve percona cluster initialization https://review.openstack.org/93032 | 20:15 |
| jprovazn_afk | greghaynes, yes | 20:15 |
| *** jcoufal has joined #tripleo | 20:15 | |
| greghaynes | jprovazn_afk: also, do you have any idea if theres a patch up for making services bind on localhost: | 20:17 |
| vinsh | Or a patch to add stunnel to overcloud-source.yaml? | 20:17 |
| jprovazn_afk | greghaynes, sure, that should be one of mine :) | 20:17 |
| *** yfaykya has quit IRC | 20:17 | |
| jprovazn_afk | https://review.openstack.org/#/c/61376/ | 20:18 |
| greghaynes | wow, thats an old one | 20:18 |
| vinsh | I wish all of these were linked off the ha blue print :) | 20:18 |
| jprovazn_afk | vinsh, there is not stunnel patch for overcloud template | 20:18 |
| *** akuznetsov has quit IRC | 20:18 | |
| *** yfaykya has joined #tripleo | 20:19 | |
| jprovazn_afk | vinsh, right, I'll update commit messages tomorrow | 20:19 |
| vinsh | Ok. I get it.. stunnel is modified to talk nodeip-localhost on any needed port. so no config is needed to add stunnel? | 20:19 |
| vinsh | oh. the stunnel element expects heat meatadata... | 20:20 |
| vinsh | so? | 20:20 |
| greghaynes | sounds like unfinished work :) | 20:21 |
| greghaynes | I also dont see a patch for that | 20:21 |
| jprovazn_afk | vinsh, there will be required a change to stunnel element (and/or overcloud template) to bind stunnel to the specified ports | 20:21 |
| vinsh | Ok, I am on the same page, understood. | 20:21 |
| greghaynes | I think https://review.openstack.org/#/c/95663 is all the needed element change | 20:21 |
| greghaynes | (also should link to blueprint) | 20:22 |
| vinsh | some greg guy -1'ed it though. | 20:22 |
| jprovazn_afk | which is different patch than setting binding addr for os services ;) | 20:22 |
| greghaynes | vinsh: yea, that jerk | 20:22 |
| * greghaynes debates just editing commit msgs to add bp for these patches | 20:22 | |
| vinsh | DOIT DOIT DOIT | 20:23 |
| *** yfaykya has quit IRC | 20:23 | |
| *** jdob has quit IRC | 20:25 | |
| *** jdob has joined #tripleo | 20:25 | |
| *** yfaykya has joined #tripleo | 20:25 | |
| jprovazn_afk | FWIW, with pending patches, you should be able to run 3 control nodes overcloud and start instances in it | 20:25 |
| greghaynes | \O/ | 20:25 |
| jprovazn_afk | (w/o stunnel ;) | 20:25 |
| jprovazn_afk | just haproxy <-> service which worked for me | 20:26 |
| jprovazn_afk | but still bunch of work remains :/ | 20:26 |
| SpamapS | jprovazn_afk: are these 3 nodes partition tolerant? | 20:26 |
| jprovazn_afk | SpamapS, partition tolerant? | 20:27 |
| lifeless | jprovazn_afk: if there is a network break and we end up with two LANs, one with 2 nodes and one with 1, what happens | 20:27 |
| lifeless | jprovazn_afk: btw for someone afk, you're very lively | 20:28 |
| *** jprovazn_afk is now known as jprovazn | 20:28 | |
| lifeless | :) | 20:28 |
| jprovazn | SpamapS, lifeless: I did not get to further testing yet, just at the point of being able to start an instance in overcloud | 20:28 |
| lifeless | jprovazn: still, thats awesome | 20:29 |
| lifeless | SpamapS: I believe someone in HP said that keepalived doesn't have quorum support | 20:29 |
| *** yfaykya has quit IRC | 20:29 | |
| lifeless | SpamapS: and that pacemaker or corosync (I forget which and its not paged in) is needed in concert to deliver that | 20:29 |
| hewbrocca | pacemaker + corosync are next on our agenda :) | 20:30 |
| lifeless | hewbrocca: cool | 20:31 |
| *** yfaykya has joined #tripleo | 20:31 | |
| greghaynes | hrm, keepalived.conf does have a quorum field... no idea what tha does though :p | 20:31 |
| greghaynes | oh, looks like you set a min weight | 20:31 |
| greghaynes | er, min total weight | 20:31 |
| hewbrocca | Fabio Di Nitto would never forgive me if I didn't make sure there was at least a pacemker *option* in tripleo | 20:31 |
| SpamapS | So without those, I believe we'll just have a loss-of-quorum failure on Galera | 20:31 |
| SpamapS | IIRC keepalived does not have any way to detect split brain. | 20:32 |
| SpamapS | Which is perfectly fine for haproy | 20:32 |
| SpamapS | haproxy | 20:32 |
| SpamapS | we don't have any state to preserve there | 20:33 |
| greghaynes | this quorum field looks like it does that as long as you set it high enough | 20:33 |
| greghaynes | if you have 3 nodes which can each provide a weight of 5, set quorum to 10 | 20:33 |
| greghaynes | you need a total weight of 10 do do anything | 20:33 |
| *** julim has quit IRC | 20:34 | |
| SpamapS | anyway, We still have unresolved issues with leader election and failure detection, I'm sure. | 20:34 |
| lifeless | ok so we just set that to floor(node_len/2 + 1) and weight one ? | 20:34 |
| greghaynes | yerp | 20:34 |
| SpamapS | but just being able to spint it up is a nice accomplishment | 20:34 |
| greghaynes | lifeless: Thats how the doc reads | 20:34 |
| jprovazn | regarding pacemaker... | 20:35 |
| *** yfaykya has quit IRC | 20:35 | |
| jprovazn | we will need it anyway because of ceilometer which was merged recently | 20:36 |
| jprovazn | and ceilometer central agent *might be* A/A in Juno, there is a bp for it, but I' mscpetic it might be in near future | 20:37 |
| *** yfaykya has joined #tripleo | 20:37 | |
| greghaynes | jprovazn: for the ceilometer bit - how do we pick which node we run that on at the moment? | 20:38 |
| jprovazn | greghaynes, pacemaker does it for us, I suppose | 20:39 |
| *** jcoufal has quit IRC | 20:39 | |
| *** nati_ueno has joined #tripleo | 20:40 | |
| jprovazn | and with pacemaker it's more fun as fedora switched from crm to pcs (management tool), ubuntu/debian sticks (so far) with crm | 20:40 |
| greghaynes | \O/ | 20:40 |
| greghaynes | jprovazn: what do you think of making fedora use mariadb by deafult? | 20:41 |
| greghaynes | something that came up at summit - were not actually testing elements like that one which you have to opt in to | 20:42 |
| *** yfaykya has quit IRC | 20:42 | |
| greghaynes | and sounds like crm / pcs will be yet another of this | 20:42 |
| jprovazn | greghaynes, that makes sense, I was waiting for mariadb to be pulled into fedora repos | 20:42 |
| greghaynes | ah, ok | 20:42 |
| *** yfaykya has joined #tripleo | 20:43 | |
| jprovazn | greghaynes, maybe with pcs/crm, it would worth try to use only one of them on all platforms if possible (=available on all platforms) | 20:44 |
| openstackgerrit | Haneef Ali proposed a change to openstack/tripleo-image-elements: Enable token-cleanup for debian https://review.openstack.org/98887 | 20:45 |
| greghaynes | ah, that would work too | 20:45 |
| greghaynes | Im not very experienced with those tools so id have to research some, but anything to keep from adding elements we dont CI :) | 20:46 |
| *** ifarkas has quit IRC | 20:46 | |
| jprovazn | greghaynes, +! | 20:46 |
| jprovazn | +1 actually | 20:46 |
| greghaynes | -> noms | 20:47 |
| vinsh | btw, in my multi control stack.. I have a separate node that runs only ceilometer.. (is build with that specifically) but shares the database on other controlnodes | 20:47 |
| *** yfaykya has quit IRC | 20:47 | |
| vinsh | its a static version of leader election. :-/ | 20:48 |
| greghaynes | yea, IMO we should just use pacemaker or $leader_election to pick a node at runtime it runs on if it really cant do multinode | 20:48 |
| *** yfaykya has joined #tripleo | 20:49 | |
| vinsh | hi-tech. one day. | 20:50 |
| *** e0ne has quit IRC | 20:53 | |
| *** yfaykya has quit IRC | 20:54 | |
| *** yfaykya has joined #tripleo | 20:55 | |
| *** jprovazn has quit IRC | 20:57 | |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Use common element get-image-kernel https://review.openstack.org/98758 | 20:57 |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Improve kernel detection in debian family systems https://review.openstack.org/98759 | 20:57 |
| openstackgerrit | Ghe Rivero proposed a change to openstack/diskimage-builder: Add get-kernel-image element https://review.openstack.org/98757 | 20:57 |
| openstackgerrit | Adam Gandelman proposed a change to openstack/tripleo-incubator: Run apt-get update before installing UCA keyring https://review.openstack.org/98896 | 20:59 |
| *** jdob has quit IRC | 20:59 | |
| *** yfaykya has quit IRC | 21:00 | |
| *** untriaged-bot has joined #tripleo | 21:00 | |
| untriaged-bot | Untriaged bugs so far: | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1328261 | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1325762 | 21:00 |
| *** untriaged-bot has quit IRC | 21:00 | |
| uvirtbot | Launchpad bug 1328261 in tripleo "Keystone token cleanup is not enabled for debian" [Undecided,In progress] | 21:00 |
| uvirtbot | Launchpad bug 1325762 in tripleo "intermittent tftp issues using physical hardware" [Undecided,New] | 21:00 |
| *** yfaykya has joined #tripleo | 21:02 | |
| *** giulivo has quit IRC | 21:03 | |
| *** matty_dubs is now known as matty_dubs|gone | 21:05 | |
| *** yfaykya has quit IRC | 21:06 | |
| *** yfaykya has joined #tripleo | 21:08 | |
| *** morganfainberg_Z is now known as morganfainberg | 21:10 | |
| *** yfaykya has quit IRC | 21:12 | |
| *** yfaykya has joined #tripleo | 21:14 | |
| greghaynes | vinsh: have you been running into issues with the apache2 ports.conf? | 21:15 |
| *** yfaykya has quit IRC | 21:18 | |
| *** yfaykya has joined #tripleo | 21:20 | |
| *** jtomasek has quit IRC | 21:24 | |
| *** yfaykya has quit IRC | 21:25 | |
| *** yfaykya has joined #tripleo | 21:26 | |
| *** yfaykya has quit IRC | 21:31 | |
| *** yfaykya has joined #tripleo | 21:32 | |
| *** yfaykya has quit IRC | 21:37 | |
| *** yfaykya has joined #tripleo | 21:38 | |
| *** yfaykya has quit IRC | 21:43 | |
| *** yfaykya has joined #tripleo | 21:44 | |
| *** hewbrocca has quit IRC | 21:48 | |
| *** yfaykya has quit IRC | 21:49 | |
| *** yfaykya has joined #tripleo | 21:51 | |
| *** yfaykya has quit IRC | 21:55 | |
| *** andreaf has joined #tripleo | 21:57 | |
| *** yfaykya has joined #tripleo | 21:57 | |
| *** rlandy_ has quit IRC | 21:59 | |
| *** eguz has quit IRC | 22:01 | |
| *** yfaykya has quit IRC | 22:01 | |
| *** eghobo has joined #tripleo | 22:02 | |
| *** yfaykya has joined #tripleo | 22:03 | |
| *** jang1 has quit IRC | 22:06 | |
| *** yfaykya has quit IRC | 22:07 | |
| *** nati_ueno has quit IRC | 22:08 | |
| *** yfaykya has joined #tripleo | 22:09 | |
| vinsh | greghaynes, where do those issues manifest? | 22:09 |
| greghaynes | in os-r-c, it seems like stunnel is defined in the current heat templates but stunnel.connect-ip is not. This results in most of the services having odd configurations | 22:10 |
| vinsh | I need to pull in those template changes as well as, 95663 . and 61376 | 22:12 |
| vinsh | Have not seen that case yet.. bet I will soon! | 22:12 |
| *** yfaykya has quit IRC | 22:13 | |
| *** yfaykya has joined #tripleo | 22:15 | |
| greghaynes | I wonder if jprovazn is just running without the ssl tempalte | 22:17 |
| *** yfaykya has quit IRC | 22:19 | |
| *** yfaykya has joined #tripleo | 22:21 | |
| vinsh | wonder how ports are configured in that case? | 22:23 |
| greghaynes | hrm? | 22:23 |
| vinsh | greghaynes, which template are you using.. just to be sure? | 22:23 |
| lifeless | templates ;) | 22:23 |
| vinsh | hah.. same :) | 22:23 |
| greghaynes | vinsh: https://review.openstack.org/#/c/93554/ | 22:24 |
| greghaynes | rebased off master | 22:24 |
| vinsh | ah yeah. I know this one all to well. | 22:25 |
| vinsh | I need to rebase off master though. | 22:25 |
| *** yfaykya has quit IRC | 22:26 | |
| vinsh | I take it your next step is stubbing out those stunnel connect to ports? | 22:26 |
| greghaynes | I also left a comment on https://review.openstack.org/#/c/61376/ with my issue | 22:27 |
| vinsh | ack. | 22:27 |
| greghaynes | uhm, Id like to ensure that the fallback (no stunnel) behavior isnt horribly broken first | 22:27 |
| vinsh | pshh | 22:27 |
| *** yfaykya has joined #tripleo | 22:27 | |
| greghaynes | because all signs are pointing to yes on that one | 22:27 |
| *** michchap has quit IRC | 22:28 | |
| vinsh | Isn't this sort of an all or nothing ? | 22:29 |
| vinsh | you want haproxy using the spec.. you need the fleshed out template and the stunnel patches | 22:29 |
| *** weshay has quit IRC | 22:29 | |
| lifeless | we need to get there from here | 22:30 |
| vinsh | What I mean is: what is the flow /use case WITHOUT stunnel? | 22:30 |
| lifeless | no SSL | 22:30 |
| lifeless | there is a patch to make throwaway SSL certs for CI | 22:30 |
| lifeless | we need that to do SSL always | 22:30 |
| lifeless | and these patches need to land one at a time | 22:30 |
| greghaynes | The templates have a {#stunnel}stuff with stunnel.connect_ip{^stunnel}other stuff{/stunnel}. The problem is AFAICT we have stunnel defined in ther overcloud templates, but no stunnel.connect_ip | 22:31 |
| *** cody-somerville has joined #tripleo | 22:31 | |
| *** cody-somerville has joined #tripleo | 22:31 | |
| *** yfaykya has quit IRC | 22:31 | |
| greghaynes | so we need to either fix the configs to check for stunnel.connect_ip or to have a way to undefine stunnel all together | 22:32 |
| *** yfaykya has joined #tripleo | 22:33 | |
| vinsh | without stunnel its haproxy bound to * <--> service bound to node IP right? | 22:34 |
| greghaynes | haproxy bound to vip, service bound to node_ip | 22:34 |
| vinsh | k. | 22:34 |
| *** rainya has quit IRC | 22:37 | |
| *** cody-somerville has quit IRC | 22:38 | |
| *** yfaykya has quit IRC | 22:38 | |
| vinsh | Hrmm.. [ALERT] 159/223813 (27312) : Starting proxy glance_api: cannot bind socket | 22:39 |
| *** yfaykya has joined #tripleo | 22:40 | |
| vinsh | listen glance_api | 22:40 |
| vinsh | bind 192.0.2.33:9292 | 22:40 |
| vinsh | mode tcp | 22:40 |
| vinsh | balance source | 22:40 |
| vinsh | option tcpka | 22:40 |
| vinsh | option tcplog | 22:40 |
| vinsh | server controller0 192.0.2.23:9292 check inter 2000 rise 2 fall 5 | 22:40 |
| vinsh | haproxy bound to the virtualip here ^^^ | 22:40 |
| vinsh | glance bound to the node ip | 22:40 |
| vinsh | haproxy unable to bind due to that | 22:40 |
| vinsh | I'm missing something. | 22:40 |
| lifeless | netstat -anp ? | 22:43 |
| *** rainya has joined #tripleo | 22:43 | |
| vinsh | netstat -anp | grep 9292 | 22:44 |
| vinsh | tcp 0 0 0.0.0.0:9292 0.0.0.0:* LISTEN 7482/python | 22:44 |
| vinsh | ? | 22:44 |
| *** yfaykya has quit IRC | 22:44 | |
| vinsh | ah. is this due to 0.0.0.0 vs specific ip for the service bind? | 22:44 |
| TheJulia | it is | 22:44 |
| TheJulia | I'm fairly sure you can't have both at the same time on the same system | 22:44 |
| lifeless | so our plan may need to have haproxy bound to the vip and restart it on vip failovers | 22:45 |
| lifeless | but thats doable | 22:45 |
| SpamapS | Correct, 0.0.0.0 pwns all | 22:45 |
| TheJulia | heh | 22:45 |
| *** yfaykya has joined #tripleo | 22:46 | |
| *** sballe_ has joined #tripleo | 22:46 | |
| *** nati_ueno has joined #tripleo | 22:47 | |
| lifeless | righto | 22:47 |
| vinsh | lifeless, in my case. haproxy IS bound to the vip. its the face that the service is bound to 0.0.0.0 that sucks. | 22:47 |
| lifeless | we need to update ze spec then | 22:47 |
| *** sballe has quit IRC | 22:47 | |
| lifeless | vinsh: yes, the service needs fixing | 22:47 |
| vinsh | +1 I follow. | 22:47 |
| lifeless | so I think a first set of patches is bind to node IP for the services | 22:47 |
| lifeless | that gets haproxy in | 22:47 |
| greghaynes | lifeless: The spec does say everything binds to a specific ip, nothing to anyaddr | 22:48 |
| lifeless | it will break our current public ip support temporarily | 22:48 |
| lifeless | greghaynes: does it? cool. | 22:48 |
| lifeless | greghaynes: not sure you can bind to a vip you don't own though | 22:48 |
| greghaynes | you can :) | 22:48 |
| vinsh | you can | 22:48 |
| greghaynes | theres a patch for that too, somewhere | 22:49 |
| vinsh | https://review.openstack.org/#/c/93121/ its merged | 22:49 |
| greghaynes | lifeless: https://review.openstack.org/#/c/61376/ is the services bind to node_ip patch youre thinking of | 22:49 |
| vinsh | just add that sauce to your template "net.blah.." | 22:49 |
| *** yfaykya has quit IRC | 22:50 | |
| *** vinsh is now known as vinsh_away | 22:51 | |
| greghaynes | SpamapS: Curious what you think of my comment on ^ (looks like you added stunnel and +2'd that patch) | 22:51 |
| *** yfaykya has joined #tripleo | 22:52 | |
| SpamapS | greghaynes: makes sense. We can probably play some games with conditional blocks to be able to operate fine without connect_ip. | 22:53 |
| greghaynes | sweet | 22:54 |
| SpamapS | fine in that, we can go on living without haproxy working right too ;) | 22:55 |
| *** yfaykya has quit IRC | 22:56 | |
| *** yfaykya has joined #tripleo | 22:58 | |
| *** yfaykya has quit IRC | 23:02 | |
| *** yfaykya has joined #tripleo | 23:04 | |
| *** yfaykya has quit IRC | 23:08 | |
| *** yfaykya has joined #tripleo | 23:10 | |
| *** yfaykya has quit IRC | 23:15 | |
| *** yfaykya has joined #tripleo | 23:16 | |
| *** yfaykya has quit IRC | 23:21 | |
| *** nati_ueno has quit IRC | 23:22 | |
| *** yfaykya has joined #tripleo | 23:22 | |
| *** morganfainberg is now known as morganfainberg_Z | 23:25 | |
| *** lucas-dinner has quit IRC | 23:26 | |
| *** geerdest has quit IRC | 23:27 | |
| *** yfaykya has quit IRC | 23:27 | |
| *** nati_ueno has joined #tripleo | 23:28 | |
| *** yfaykya has joined #tripleo | 23:28 | |
| *** nati_uen_ has joined #tripleo | 23:31 | |
| *** nati_ueno has quit IRC | 23:32 | |
| *** yfaykya has quit IRC | 23:33 | |
| *** yfaykya has joined #tripleo | 23:35 | |
| *** xuhaiwei has joined #tripleo | 23:37 | |
| *** yfaykya has quit IRC | 23:40 | |
| *** yfaykya has joined #tripleo | 23:41 | |
| *** morganfainberg_Z is now known as morganfainberg | 23:42 | |
| *** yfaykya has quit IRC | 23:45 | |
| StevenK | lifeless, tchaypo: Looks like we have a winner for the sprint dates: July 21-25 | 23:46 |
| StevenK | Which makes me sad, since I'll miss Sarah's birthday | 23:47 |
| *** yfaykya has joined #tripleo | 23:47 | |
| greghaynes | oh, really? | 23:47 |
| greghaynes | :/ | 23:47 |
| *** eguz has joined #tripleo | 23:49 | |
| *** yfaykya has quit IRC | 23:51 | |
| StevenK | greghaynes: http://lists.openstack.org/pipermail/openstack-dev/2014-June/036974.html | 23:52 |
| *** eghobo has quit IRC | 23:53 | |
| *** yfaykya has joined #tripleo | 23:53 | |
| SpamapS | I may miss the whole thing entirely | 23:53 |
| greghaynes | :c | 23:54 |
| SpamapS | July is just bong for me. | 23:54 |
| SpamapS | And that first week of August is when all three of my kids are starting brand new schools in a span of 3 days. | 23:54 |
| StevenK | I told Sarah the dates before she ran out the door, and she's like "So, situation normal, then." | 23:54 |
| SpamapS | we need to plan these further out | 23:54 |
| greghaynes | SpamapS: ++ | 23:55 |
| SpamapS | I'd rather have it the week before feature freeze than only have 6 weeks to plan. | 23:55 |
| StevenK | So a session at the summit about sprint dates? :-) | 23:55 |
| SpamapS | StevenK: and a session after that where we invite Captain Obvious to detail his plans for curing Polio. | 23:55 |
| SpamapS | I think a simple ML thread pleading with people "Please plan these 2 months in advance at least." | 23:56 |
| pleia2 | StevenK: I enjoy the nice jewelry and delivered flowers that come from the husband being out to down for $important_day | 23:56 |
| pleia2 | (fortunately husband birthday is near xmas, and no one schedules stuff around then, woo) | 23:57 |
| SpamapS | haha.. yeah the summit was right up against mother's day.. I was there for mother's day, but had to leave at 14:00 .. so there were extra flowers in the morning. ;) | 23:57 |
| StevenK | Yeah, I flew out on Mother's Day, so my mother wasn't really pleased | 23:57 |
| *** yfaykya has quit IRC | 23:57 | |
| pleia2 | yeah, I was planning on flying on mother's day, so I sent her jewelry in advance | 23:57 |
| StevenK | pleia2: Hmmmmm, flowers won't work for Sarah, but jewelry you say ... :-) | 23:58 |
| pleia2 | StevenK: http://www.boutiqueacademia.com/ go to town :) | 23:59 |
| StevenK | jogo: "Hacking 0.9.1 has just been released (hacking 0.9.1 had a minor bug)." | 23:59 |
| SpamapS | beautiful | 23:59 |
| *** yfaykya has joined #tripleo | 23:59 | |
| StevenK | pleia2: Oooooh | 23:59 |
| SpamapS | "Hacking 0.9.forklobster has been released to address the bugs in hacking 0.9.1 and hacking 0.9.1." | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!