| *** markvoelker_ has quit IRC | 00:01 | |
| *** JRobinson__ has quit IRC | 00:11 | |
| *** mss_ has quit IRC | 00:16 | |
| *** harlowja has joined #openstack-ansible | 00:20 | |
| *** gouthamr_ has quit IRC | 00:28 | |
| *** dolpher has joined #openstack-ansible | 00:33 | |
| openstackgerrit | Merged openstack/openstack-ansible: Adding docs/comments for glance config https://review.openstack.org/242918 | 00:34 |
|---|---|---|
| *** jmckind has quit IRC | 00:37 | |
| *** haojing has joined #openstack-ansible | 01:23 | |
| *** markvoelker has joined #openstack-ansible | 01:23 | |
| *** markvoelker has quit IRC | 01:28 | |
| *** devl_ has quit IRC | 01:46 | |
| *** devl_ has joined #openstack-ansible | 02:01 | |
| *** mss has joined #openstack-ansible | 02:06 | |
| *** phschwartz has quit IRC | 02:09 | |
| *** Bjoern_ has quit IRC | 02:09 | |
| *** phschwartz has joined #openstack-ansible | 02:09 | |
| *** dolpher has left #openstack-ansible | 02:11 | |
| *** Bjoern_ has joined #openstack-ansible | 02:48 | |
| *** bapalm has quit IRC | 02:54 | |
| *** bapalm has joined #openstack-ansible | 03:02 | |
| *** cloudtrainme has joined #openstack-ansible | 03:10 | |
| *** cloudtrainme has quit IRC | 03:21 | |
| *** markvoelker has joined #openstack-ansible | 03:24 | |
| *** sdake has quit IRC | 03:25 | |
| *** sdake has joined #openstack-ansible | 03:26 | |
| *** markvoelker has quit IRC | 03:28 | |
| *** Bjoern_ has quit IRC | 03:42 | |
| *** sdake has quit IRC | 03:48 | |
| *** sdake has joined #openstack-ansible | 04:01 | |
| cloudnull | evenings | 04:03 |
| *** sdake has quit IRC | 04:27 | |
| *** sdake has joined #openstack-ansible | 04:29 | |
| bgmccollum | howdy | 04:42 |
| *** tlian2 has quit IRC | 05:13 | |
| *** markvoelker has joined #openstack-ansible | 05:25 | |
| *** markvoelker has quit IRC | 05:29 | |
| *** haojingma has joined #openstack-ansible | 05:31 | |
| *** haojing has quit IRC | 05:33 | |
| *** woodard has quit IRC | 05:43 | |
| *** neillc is now known as neillc_away | 05:45 | |
| *** ggillies_ is now known as ggillies | 06:02 | |
| *** ig0r__ has quit IRC | 06:05 | |
| *** javeriak has joined #openstack-ansible | 06:09 | |
| *** ig0r__ has joined #openstack-ansible | 06:10 | |
| *** devl has joined #openstack-ansible | 06:14 | |
| *** devl_ has quit IRC | 06:15 | |
| *** phiche has joined #openstack-ansible | 06:28 | |
| *** haojing has joined #openstack-ansible | 06:30 | |
| *** haojingma has quit IRC | 06:33 | |
| *** mss has quit IRC | 06:36 | |
| *** August1914 has quit IRC | 06:43 | |
| *** agireud has quit IRC | 06:44 | |
| *** agireud has joined #openstack-ansible | 06:46 | |
| *** phiche has quit IRC | 07:02 | |
| *** javeriak has quit IRC | 07:13 | |
| *** phiche has joined #openstack-ansible | 07:15 | |
| *** javeriak has joined #openstack-ansible | 07:16 | |
| *** javeriak has quit IRC | 07:21 | |
| *** markvoelker has joined #openstack-ansible | 07:26 | |
| *** javeriak has joined #openstack-ansible | 07:30 | |
| *** markvoelker has quit IRC | 07:31 | |
| *** mss has joined #openstack-ansible | 07:33 | |
| *** ashishb has joined #openstack-ansible | 07:37 | |
| *** mss has quit IRC | 07:38 | |
| *** openstackgerrit has quit IRC | 07:46 | |
| *** openstackgerrit has joined #openstack-ansible | 07:47 | |
| *** britthouser has quit IRC | 07:59 | |
| *** javeriak has quit IRC | 08:10 | |
| *** javeriak has joined #openstack-ansible | 08:10 | |
| *** javeriak_ has joined #openstack-ansible | 08:14 | |
| *** javeriak has quit IRC | 08:15 | |
| *** britthouser has joined #openstack-ansible | 08:15 | |
| *** mpavone has joined #openstack-ansible | 08:23 | |
| *** mpavone has quit IRC | 08:23 | |
| *** openstackgerrit has quit IRC | 08:31 | |
| *** devl has quit IRC | 08:31 | |
| *** openstackgerrit has joined #openstack-ansible | 08:31 | |
| *** javeriak_ has quit IRC | 08:34 | |
| *** devl_ has joined #openstack-ansible | 08:39 | |
| *** devl_ has quit IRC | 08:41 | |
| *** tiagogomes_ has joined #openstack-ansible | 08:42 | |
| *** devl_ has joined #openstack-ansible | 08:43 | |
| *** agireud has quit IRC | 08:43 | |
| *** egonzalez has joined #openstack-ansible | 08:51 | |
| *** wtie has joined #openstack-ansible | 09:03 | |
| evrardjp | hello everyone | 09:10 |
| mattt | howdy evrardjp | 09:21 |
| *** gouthamr has joined #openstack-ansible | 09:21 | |
| *** markvoelker has joined #openstack-ansible | 09:27 | |
| *** subscope has joined #openstack-ansible | 09:29 | |
| *** markvoelker has quit IRC | 09:31 | |
| *** mss has joined #openstack-ansible | 09:35 | |
| *** mss has quit IRC | 09:39 | |
| *** sdake has quit IRC | 09:42 | |
| *** sdake has joined #openstack-ansible | 09:42 | |
| *** devl_ has quit IRC | 09:43 | |
| evrardjp | hello mattt! is everything all right on your side? | 09:54 |
| *** haojing has quit IRC | 10:08 | |
| mattt | evrardjp: can't complain, how are you? :) | 10:10 |
| *** devl_ has joined #openstack-ansible | 10:18 | |
| evrardjp | I have minutes to kill, so I'll do a few reviews | 10:18 |
| *** sdake has quit IRC | 10:20 | |
| *** subscope has quit IRC | 10:26 | |
| *** mss has joined #openstack-ansible | 10:35 | |
| *** mss has quit IRC | 10:39 | |
| *** ig0r__ has quit IRC | 10:40 | |
| mattt | evrardjp: thanks! :) | 10:46 |
| *** javeriak has joined #openstack-ansible | 10:51 | |
| *** javeriak has quit IRC | 10:54 | |
| *** gparaskevas has joined #openstack-ansible | 11:02 | |
| *** javeriak has joined #openstack-ansible | 11:11 | |
| *** metral is now known as metral_zzz | 11:12 | |
| evrardjp | do you think the chan name should be changed now? (the repo-rename is, I guess, not really relevant anymore?) | 11:12 |
| evrardjp | I'd be happy to learn more about this neutron_migration stuff | 11:16 |
| evrardjp | I didn't got the chance to check on this neutron_migrations_facts, but I'd need more knowledge to review it | 11:17 |
| *** sdake has joined #openstack-ansible | 11:19 | |
| *** sdake has quit IRC | 11:21 | |
| mattt | evrardjp: chan name change? | 11:24 |
| mattt | evrardjp: to what? | 11:24 |
| evrardjp | mattt: just removing the "repo name from stackforge/os-ansible-deployment ..." | 11:25 |
| evrardjp | I'm not sure it's useful anymore | 11:25 |
| mattt | evrardjp: ah, the topic! yes, that should probably be changed now :P | 11:26 |
| *** markvoelker has joined #openstack-ansible | 11:27 | |
| *** metral_zzz is now known as metral | 11:29 | |
| *** markvoelker has quit IRC | 11:32 | |
| *** subscope has joined #openstack-ansible | 11:41 | |
| *** jaypipes has joined #openstack-ansible | 11:42 | |
| *** javeriak_ has joined #openstack-ansible | 11:46 | |
| *** javeriak has quit IRC | 11:49 | |
| *** wtie has quit IRC | 11:49 | |
| *** sdake has joined #openstack-ansible | 12:32 | |
| *** sdake has quit IRC | 12:33 | |
| *** cloudtrainme has joined #openstack-ansible | 12:34 | |
| *** sdake has joined #openstack-ansible | 12:34 | |
| *** sdake has quit IRC | 12:35 | |
| *** alkari has joined #openstack-ansible | 12:47 | |
| *** admiralboom has quit IRC | 12:48 | |
| *** markvoelker has joined #openstack-ansible | 12:58 | |
| *** admiralboom has joined #openstack-ansible | 13:00 | |
| *** gouthamr has quit IRC | 13:00 | |
| *** markvoelker has quit IRC | 13:03 | |
| *** cloudtrainme has quit IRC | 13:08 | |
| *** gouthamr has joined #openstack-ansible | 13:10 | |
| mhayden | mornin' | 13:11 |
| mhayden | evrardjp: thanks for the comments! | 13:11 |
| evrardjp | hmmm I plead the 5th? | 13:12 |
| mhayden | evrardjp: hah, was talking about the CA spec | 13:13 |
| evrardjp | ;) | 13:13 |
| mhayden | evrardjp: are you suggesting i should create it within my own github namespace, and then reel it in similar to how your code is brought in? | 13:15 |
| evrardjp | It's up to you, maybe we should speak about that with odyssey4me and cloudnull | 13:16 |
| evrardjp | but yeah, I think it's a good idea | 13:17 |
| mhayden | i can definitely see your point there | 13:17 |
| evrardjp | it's seperate stuff but it could be good glue | 13:17 |
| *** subscope has quit IRC | 13:17 | |
| mhayden | i'd like to write one anyway that i could use elsewhere ;) | 13:17 |
| evrardjp | :D | 13:17 |
| mhayden | i keep using XCA for my own CA | 13:17 |
| evrardjp | don't know XCA | 13:17 |
| mhayden | http://xca.sourceforge.net/ | 13:17 |
| mhayden | it's a GUI app for linux/mac/windows for managing a simple CA | 13:17 |
| evrardjp | we don't maintain our ca anymore, so it's easier for us :D | 13:17 |
| evrardjp | cool | 13:18 |
| *** ashishb has quit IRC | 13:18 | |
| mhayden | all of the stuff related to the CA is kept in one encrypted file | 13:18 |
| evrardjp | easier than ejbca I guess? ;) | 13:18 |
| mhayden | it's quite handy | 13:18 |
| mhayden | ah, i hadn't heard of ejbca | 13:18 |
| mhayden | whoa, this thing looks nice | 13:18 |
| evrardjp | it supports hardware tokens | 13:19 |
| evrardjp | :D | 13:19 |
| evrardjp | a lighter tool could be dogtag from redhat | 13:19 |
| evrardjp | it's integrated in freeIPA IIRC (while you could never have heard about dogtag, you could have heard about freeipa) | 13:20 |
| *** markvoelker has joined #openstack-ansible | 13:20 | |
| *** markvoelker has quit IRC | 13:20 | |
| mhayden | ah yes, i've known about that one for a while :P | 13:21 |
| mhayden | it's palendae's favorite | 13:21 |
| evrardjp | for deployers, the easier the better | 13:23 |
| evrardjp | else it would lose the benefit/purpose | 13:24 |
| evrardjp | lose isn't the correct term | 13:24 |
| *** markvoelker has joined #openstack-ansible | 13:25 | |
| *** woodard has joined #openstack-ansible | 13:28 | |
| *** woodard has quit IRC | 13:28 | |
| *** woodard has joined #openstack-ansible | 13:29 | |
| *** wtie has joined #openstack-ansible | 13:33 | |
| *** subscope has joined #openstack-ansible | 13:34 | |
| *** Bjoern_ has joined #openstack-ansible | 13:40 | |
| *** britthouser has quit IRC | 13:51 | |
| *** britthouser has joined #openstack-ansible | 13:51 | |
| *** tlian has joined #openstack-ansible | 13:53 | |
| cloudnull | morning | 14:13 |
| cloudnull | mhayden: evrardjp: ++ I maintain i push things locally all the time and re-push or move the repos when ready for general coonsumption | 14:15 |
| cloudnull | it seems to be a good pattern for me | 14:15 |
| cloudnull | mhayden: can you backport his to kilo https://review.openstack.org/#/c/243226/ ? | 14:29 |
| cloudnull | ha never mind | 14:29 |
| mhayden | cloudnull: isn't that this one? https://review.openstack.org/#/c/243227/ | 14:29 |
| cloudnull | i just dont read good :) | 14:30 |
| mhayden | it's okay | 14:30 |
| mhayden | :) | 14:30 |
| * cloudnull http://25.media.tumblr.com/tumblr_l8wkfpSzOu1qa9a0do1_500.jpg | 14:31 | |
| *** jmckind has joined #openstack-ansible | 14:31 | |
| *** sdake has joined #openstack-ansible | 14:34 | |
| evrardjp | :) | 14:34 |
| *** sdake has quit IRC | 14:35 | |
| *** jwitko has quit IRC | 14:36 | |
| *** jwitko has joined #openstack-ansible | 14:36 | |
| *** mss has joined #openstack-ansible | 14:45 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:45 | |
| *** javeriak_ has quit IRC | 14:48 | |
| *** cloudtrainme has joined #openstack-ansible | 14:51 | |
| openstackgerrit | XiaBing Yao proposed openstack/openstack-ansible: Choose virt type automatically https://review.openstack.org/243098 | 14:55 |
| *** Mudpuppy has joined #openstack-ansible | 15:04 | |
| *** Mudpuppy has quit IRC | 15:05 | |
| *** Mudpuppy has joined #openstack-ansible | 15:06 | |
| *** KLevenstein has joined #openstack-ansible | 15:08 | |
| mhayden | cloudnull: is this one worth backporting? https://review.openstack.org/#/c/242918/ | 15:09 |
| mhayden | it looks like the same functionality is in libery/kilo | 15:09 |
| *** Mudpuppy has quit IRC | 15:09 | |
| *** spotz_zzz is now known as spotz | 15:09 | |
| *** Mudpuppy has joined #openstack-ansible | 15:10 | |
| *** woodard has quit IRC | 15:13 | |
| palendae | mhayden: Eh? | 15:15 |
| mhayden | palendae: i was just messing with you about IPA/dogtag :P | 15:16 |
| palendae | I have no idea what you're talking about with those | 15:16 |
| mhayden | hmm, then perhaps i was having a joke with someone else | 15:16 |
| * mhayden cannot keep his jokes straight | 15:16 | |
| bgmccollum | anyone seen their memcache containers fail to have their facts gathered, because they aren't up yet? seems like after creating the containers, waiting for SSH to be up might be needed. | 15:18 |
| evrardjp | bgmccollum: didn't see that one yet | 15:19 |
| bgmccollum | evrardjp: ah...waiting for SSH was added in lliberty. ill see if i can find the review and backport. | 15:20 |
| bgmccollum | im seeing this in kilo | 15:20 |
| *** javeriak has joined #openstack-ansible | 15:22 | |
| *** javeriak has quit IRC | 15:23 | |
| *** javeriak has joined #openstack-ansible | 15:23 | |
| *** jimchou has joined #openstack-ansible | 15:23 | |
| bgmccollum | https://review.openstack.org/#/c/178374/ has the change, but its part of a larger changeset. whats the best way to backport...or should i just forego using cherry-pick? | 15:26 |
| palendae | bgmccollum: I would take the change you want, but use the gerrit change-id in your commit so it can be tracked | 15:27 |
| bgmccollum | palendae: ok thanks | 15:27 |
| *** javeriak has quit IRC | 15:28 | |
| *** javeriak_ has joined #openstack-ansible | 15:28 | |
| *** sdake has joined #openstack-ansible | 15:29 | |
| *** sdake has quit IRC | 15:30 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Adding docs/comments for glance config https://review.openstack.org/243661 | 15:33 |
| *** fawadkhaliq has joined #openstack-ansible | 15:36 | |
| bgmccollum | palendae: when i use the same change ID, it says its already closed -- (change 178377 closed) | 15:42 |
| palendae | bgmccollum: This one - Change-Id: I1872a32b53b9caecb8ac0803db304fa0ee254475 | 15:42 |
| bgmccollum | yeah...i think a partial change was backported to kilo, using that change ID | 15:43 |
| bgmccollum | its just the part i need wasn't | 15:43 |
| bgmccollum | so...new change ID? | 15:43 |
| palendae | Ahh, ok | 15:43 |
| palendae | Yeah, guess so | 15:43 |
| bgmccollum | thanks | 15:44 |
| openstackgerrit | Byron McCollum proposed openstack/openstack-ansible: Wait for container SSH https://review.openstack.org/243668 | 15:44 |
| *** sdake has joined #openstack-ansible | 15:48 | |
| *** subscope has quit IRC | 15:51 | |
| *** sdake_ has joined #openstack-ansible | 15:51 | |
| *** sdake has quit IRC | 15:53 | |
| *** ysm has joined #openstack-ansible | 15:55 | |
| *** egonzalez has quit IRC | 15:57 | |
| *** egonzalez has joined #openstack-ansible | 15:59 | |
| *** cloudtrainme has quit IRC | 16:01 | |
| *** phiche has quit IRC | 16:03 | |
| mhayden | cloudnull: git clone --depth 1 is totally not recommended | 16:03 |
| mhayden | experiment complete | 16:03 |
| mhayden | :P | 16:03 |
| mattt | no meeting today ? | 16:05 |
| *** egonzalez has quit IRC | 16:06 | |
| *** egonzalez has joined #openstack-ansible | 16:06 | |
| *** cloudtrainme has joined #openstack-ansible | 16:07 | |
| *** egonzalez has quit IRC | 16:07 | |
| cloudnull | mattt: just got back to the desk | 16:07 |
| cloudnull | im happy to run the meeting i think odyssey4me_ is still out | 16:08 |
| palendae | mhayden: Doesn't that just get you the top level directory? | 16:08 |
| cloudnull | cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, mancdaz, dolphm, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva -- bug triage meeting | 16:08 |
| mhayden | only snags the last commit in there | 16:08 |
| mattt | cloudnull: yeah he's not in the office today, not sure if he was wfh | 16:08 |
| palendae | Ahh | 16:08 |
| Sam-I-Am | moo. | 16:09 |
| cloudnull | first issue to triage https://bugs.launchpad.net/openstack-ansible/+bug/1513132 | 16:09 |
| openstack | Launchpad bug 1513132 in openstack-ansible "RabbitMQ should not install when not needed" [Undecided,New] | 16:09 |
| cloudnull | rackertom: around ? | 16:10 |
| rackertom | o/ | 16:10 |
| mattt | o/ | 16:10 |
| *** gparaskevas has quit IRC | 16:11 | |
| cloudnull | rackertom: it seems that you can set affinity for rabbitmq_containers or remove the env bits from the /etc/openstack_deploy/env.d dir | 16:11 |
| cloudnull | that would get rid of the assumed rabbitmq deployment when using the shared-infra group | 16:12 |
| rackertom | Yeah, so we discussed removing the env.d dir, but my worry was what happens when someone later on decides they do want a component that uses rabbit? | 16:12 |
| rackertom | Setting affinity meaning assign the location of the rabbit containers to a location that doesn't exist? | 16:13 |
| cloudnull | thats fair, we could set the affinity for that container type to 0 | 16:13 |
| cloudnull | https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio#L76 | 16:13 |
| rackertom | Ah! | 16:13 |
| rackertom | Well, that might be an interesting way to do it. | 16:13 |
| cloudnull | in that way there'd be no rabbitmq entries in inventory | 16:14 |
| rackertom | I was thinking more along the lines of making the osa install grab components based on dependency, but this is a quicker solution with less disruption. :D | 16:14 |
| palendae | rackertom: affinity being the number of containers made | 16:14 |
| palendae | That seem cleaner to me, yeah | 16:14 |
| palendae | And easier to adjust if rabbitmq becomes necessary | 16:14 |
| palendae | So sounds like a doc-ish fix? | 16:15 |
| cloudnull | so invalid ? or do we think that we want to do a larger decoupling effort ? | 16:15 |
| cloudnull | ++ | 16:15 |
| rackertom | It's still _basically_ cheating the system. But it totally works. | 16:15 |
| cloudnull | documenting more on affinities would likely be a good route too | 16:15 |
| evrardjp | introducing dependencies could be done when doing the role split | 16:15 |
| rackertom | But yeah, for the right here and now, this is easy to solve by filing a docs bug. | 16:15 |
| rackertom | ++ | 16:16 |
| evrardjp | in a later phase | 16:16 |
| cloudnull | sweet | 16:16 |
| rackertom | evrardjp: ++ | 16:16 |
| rackertom | That was my thoughts exactly. | 16:16 |
| rackertom | s/thoughts/thought/ | 16:16 |
| cloudnull | next: https://bugs.launchpad.net/openstack-ansible/+bug/1512883 | 16:17 |
| openstack | Launchpad bug 1512883 in openstack-ansible "Host name of haproxy in openstack_user_config.yml causes recursion error" [Undecided,New] | 16:17 |
| palendae | Yeah, looking at https://www.dropbox.com/s/pldwzs846fc3aqy/osa.png, some things will need to point to rabbit | 16:17 |
| cloudnull | so this is an interesting use case and seemds to cause a recursion error. | 16:18 |
| cloudnull | idk if this is something that anyone can confirm ? | 16:19 |
| evrardjp | I didn't test it | 16:20 |
| palendae | I can spin up an AIO and give my results | 16:20 |
| palendae | I'm curious about other services, too | 16:20 |
| evrardjp | yeah | 16:20 |
| cloudnull | yea that one would be interesting and if confirmed will likely be something we have to adjust in dyn-inv to be more name aware . | 16:21 |
| cloudnull | palendae: you mind rolling with that for now ? | 16:21 |
| palendae | Sure | 16:21 |
| cloudnull | tyvm sir | 16:23 |
| cloudnull | next https://bugs.launchpad.net/openstack-ansible/+bug/1511910 | 16:23 |
| openstack | Launchpad bug 1511910 in openstack-ansible "disable innobackupex --version-check" [Undecided,Incomplete] | 16:23 |
| cloudnull | this issue seems to be related only to mariadb 5.5 and likely only 5.5 from an early Juno deployment | 16:24 |
| evrardjp | mariadb 5.5 is also used for kilo | 16:26 |
| evrardjp | how do you know it's a juno issue? | 16:26 |
| cloudnull | its an assumption . | 16:26 |
| cloudnull | Darren works on the RAX eng team and its likely that the issue came from a deployment and in terms of RAX Juno is the largest deployment base | 16:27 |
| cloudnull | also while Kilo uses 5.5 I've not run into this issue within a Kilo deployment | 16:29 |
| *** woodard has joined #openstack-ansible | 16:29 | |
| cloudnull | but it very well could be there too | 16:29 |
| evrardjp | I'm no expert in percona, but I don't see the file mentionned in the bug under /tmp | 16:30 |
| Bjoern_ | yes it's most likely juno | 16:30 |
| *** Bjoern_ is now known as BjoernT | 16:30 | |
| cloudnull | maybe the best fix here is to doc galera recovery and to note that the file needs to be removed if found ? | 16:30 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: AIO bootstrap in Ansible https://review.openstack.org/239525 | 16:31 |
| BjoernT | you get this issue only if galera is running for quite some tie | 16:31 |
| BjoernT | time | 16:31 |
| BjoernT | this issue was actually fixed in percona, we just running a older version | 16:31 |
| evrardjp | 104 days ? | 16:31 |
| evrardjp | ;) | 16:31 |
| BjoernT | 104 days how cute :-) | 16:32 |
| evrardjp | anyway, the bug lacks description IMO | 16:32 |
| BjoernT | yes update it please | 16:32 |
| cloudnull | BjoernT: could you relay to darrin to update the issue ? | 16:32 |
| evrardjp | I approve cloudnull's comment | 16:32 |
| BjoernT | i prefer to set the bug in incomplete | 16:33 |
| *** sdake_ has quit IRC | 16:33 | |
| BjoernT | he's 2nd shift so I don't see him all the time | 16:33 |
| cloudnull | ok ill leave the issue "incomplete" for now but i posed a comment, we'll see what happens. | 16:35 |
| cloudnull | so thats it today for the new issues | 16:36 |
| cloudnull | anything inflight that we should specifically talk about ? | 16:36 |
| cloudnull | or get eyes on ? | 16:36 |
| *** August1914 has joined #openstack-ansible | 16:37 | |
| sigmavirus24 | BjoernT: there was an ldap issue with kilo that you had filed and I failed to follow up on | 16:37 |
| mhayden | would definitely appreciate some feedback on the AIO bootstrap | 16:37 |
| BjoernT | let me check, i'm not sure what's missing | 16:38 |
| cloudnull | BjoernT: sigmavirus24: https://review.openstack.org/#/c/241809/ | 16:38 |
| cloudnull | Jimmy jumped on that one | 16:38 |
| *** sdake has joined #openstack-ansible | 16:38 | |
| cloudnull | jmccrory: | 16:38 |
| *** subscope has joined #openstack-ansible | 16:38 | |
| sigmavirus24 | Ah okay | 16:38 |
| sigmavirus24 | Nevermind then BjoernT sorry | 16:38 |
| BjoernT | ok | 16:38 |
| cloudnull | sigmavirus24: good looking out though | 16:39 |
| sigmavirus24 | Thanks jmccrory! | 16:39 |
| cloudnull | on open reviews https://review.openstack.org/#/c/241483/ could use some love | 16:39 |
| cloudnull | we'll need that once we start going down the path of full roll removal | 16:39 |
| *** subscope has quit IRC | 16:39 | |
| cloudnull | from the main repo | 16:40 |
| palendae | Along those lines, I should probably file something to get https://www.dropbox.com/s/pldwzs846fc3aqy/osa.png auto-generated and inserted into the docs on builds | 16:40 |
| cloudnull | also https://review.openstack.org/#/c/239505/ would be a nice to have | 16:40 |
| cloudnull | ++ palendae that would be great | 16:40 |
| *** Jezogwza has joined #openstack-ansible | 16:41 | |
| cloudnull | however the logging change there would be a stop gap change until the role work is finalized in https://github.com/evrardjp/ansible-haproxy | 16:42 |
| evrardjp | it's still worth doing | 16:42 |
| evrardjp | IMHO | 16:43 |
| cloudnull | evrardjp: if you'd like I can PR that change into your new proposed role ? | 16:43 |
| evrardjp | I have time tomorrow, I'll be on holiday, so I can even do that myself | 16:43 |
| cloudnull | if you think its worth while and wouldn't conflict with what you have that is . | 16:43 |
| *** subscope has joined #openstack-ansible | 16:44 | |
| cloudnull | so thats all i have today folks . thanks everyone ! | 16:44 |
| cloudnull | mhayden: im testing the aio bits RE: https://review.openstack.org/#/c/239525/ | 16:45 |
| cloudnull | so more on that soon | 16:45 |
| *** mgoddard_ has joined #openstack-ansible | 16:46 | |
| mhayden | cloudnull: awesome | 16:48 |
| mhayden | thanks | 16:48 |
| *** mgoddard has quit IRC | 16:48 | |
| *** August1914 has quit IRC | 16:53 | |
| *** sdake has quit IRC | 16:54 | |
| *** subscope has quit IRC | 16:54 | |
| *** mgoddard has joined #openstack-ansible | 16:55 | |
| *** mgoddard_ has quit IRC | 16:55 | |
| mhayden | if anyone has thoughts on the CA idea, that'd be nice to hear as well | 16:58 |
| mhayden | evrardjp had a good suggestion of developing something generic outside of openstack and then reeling it in during bootstrap | 16:58 |
| mhayden | if that's the case, i might WIP the spec and bring back a newer version of it when a good CA role exists to rope in | 16:59 |
| evrardjp | mhayden: did you by chance, check on github/ansible galaxy if there was a decent CA role? | 17:00 |
| *** fawadkhaliq has quit IRC | 17:00 | |
| *** phiche has joined #openstack-ansible | 17:00 | |
| *** phiche1 has joined #openstack-ansible | 17:00 | |
| evrardjp | cloudnull: I checked the commit https://review.openstack.org/#/c/239505/ | 17:00 |
| evrardjp | I don't think anything from this commit should require a PR | 17:00 |
| evrardjp | in my role | 17:00 |
| evrardjp | the role is already ready for logging changes | 17:01 |
| cloudnull | sweet ! | 17:01 |
| cloudnull | then we're good here :) | 17:01 |
| evrardjp | ofc, the rsyslog changes should be somewhere, but for me, that's part of the glue: it's part of the playbooks or part of the rsyslog role | 17:01 |
| cloudnull | ++ agreeded , we can get that change in the playbook | 17:02 |
| evrardjp | this way haproxy is independant of rsyslog | 17:02 |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible: Make post tasks dependent on is_metal flag. Those udev tasks can't be executed when running on metal. https://review.openstack.org/243727 | 17:02 |
| evrardjp | (btw it's possible to change haproxy logging system to files IIRC) | 17:02 |
| mhayden | evrardjp: no good roles found, but i did find that feller who wrote some handy ansibe libraries | 17:02 |
| mhayden | they make managing a CA via tasks really easy | 17:02 |
| evrardjp | cool | 17:02 |
| evrardjp | I'll check that | 17:03 |
| mhayden | evrardjp: https://github.com/rclayton-the-terrible/Ansible-OpenSSL | 17:03 |
| mhayden | he has some weird voodoo in that build.py script though | 17:04 |
| evrardjp | we should check if it deserves to be some kind of module for openstack-ansible or even ansible upstream | 17:04 |
| *** phiche has quit IRC | 17:04 | |
| mhayden | the other option might be to go low tech and let ansible generate some type of script via template and run that | 17:05 |
| mhayden | to issue certs | 17:05 |
| mhayden | perhaps even use ca.pl | 17:05 |
| mhayden | but alas, the belly calls for good | 17:05 |
| mhayden | so i must depart ;) | 17:06 |
| evrardjp | :D | 17:06 |
| mhayden | s/good/food/ | 17:06 |
| evrardjp | good food? | 17:06 |
| stevelle | life is too short for any other kind | 17:06 |
| evrardjp | I like the ca module idea | 17:07 |
| evrardjp | it brings convenience to something that should be convenient | 17:07 |
| *** ashishb has joined #openstack-ansible | 17:13 | |
| * cloudnull lunching | 17:15 | |
| *** galstrom_zzz is now known as galstrom | 17:18 | |
| *** alkari has quit IRC | 17:20 | |
| *** ashishb has quit IRC | 17:24 | |
| *** ashishb has joined #openstack-ansible | 17:25 | |
| *** ashishb has joined #openstack-ansible | 17:25 | |
| *** ashishb has quit IRC | 17:25 | |
| *** ashishb has joined #openstack-ansible | 17:26 | |
| *** jmckind is now known as jmckind_ | 17:26 | |
| *** ashishb has quit IRC | 17:27 | |
| *** ashishb has joined #openstack-ansible | 17:28 | |
| *** ashishb has quit IRC | 17:29 | |
| *** ashishb has joined #openstack-ansible | 17:30 | |
| *** ashishb has quit IRC | 17:32 | |
| *** ashishb has joined #openstack-ansible | 17:33 | |
| *** ashishb has quit IRC | 17:35 | |
| *** ashishb has joined #openstack-ansible | 17:36 | |
| *** ashishb has quit IRC | 17:38 | |
| *** ashishb has joined #openstack-ansible | 17:38 | |
| *** ashishb has quit IRC | 17:42 | |
| *** ashishb has joined #openstack-ansible | 17:43 | |
| *** ashishb has quit IRC | 17:47 | |
| *** ashishb has joined #openstack-ansible | 17:48 | |
| *** ashishb has quit IRC | 17:51 | |
| mhayden | good lunchtime reading: http://gizmodo.com/what-happens-if-you-drop-your-iphone-and-then-accidenta-1741693373 | 17:51 |
| *** ashishb has joined #openstack-ansible | 17:51 | |
| *** ashishb has quit IRC | 17:53 | |
| *** ashishb has joined #openstack-ansible | 17:53 | |
| mhayden | you know, because that happens to everyone from time to time | 17:54 |
| *** ashishb has quit IRC | 17:54 | |
| *** ashishb has joined #openstack-ansible | 17:54 | |
| *** ashishb has quit IRC | 17:55 | |
| *** ashishb has joined #openstack-ansible | 17:55 | |
| *** ashishb has quit IRC | 17:55 | |
| *** ashishb has joined #openstack-ansible | 17:56 | |
| *** ashishb has quit IRC | 17:57 | |
| *** ashishb has joined #openstack-ansible | 17:57 | |
| *** rady has joined #openstack-ansible | 17:57 | |
| *** ashishb has quit IRC | 18:00 | |
| *** ashishb has joined #openstack-ansible | 18:00 | |
| *** CheKoLyN has joined #openstack-ansible | 18:02 | |
| *** ashishb has quit IRC | 18:02 | |
| *** ashishb has joined #openstack-ansible | 18:02 | |
| *** ashishb has quit IRC | 18:04 | |
| *** ashishb has joined #openstack-ansible | 18:04 | |
| *** ashishb has quit IRC | 18:06 | |
| *** ashishb has joined #openstack-ansible | 18:08 | |
| *** rebase has joined #openstack-ansible | 18:09 | |
| *** cloudtrainme has quit IRC | 18:10 | |
| *** ashishb has quit IRC | 18:10 | |
| *** ashishb has joined #openstack-ansible | 18:10 | |
| * mhayden winks at andyhky | 18:11 | |
| *** ashishb has quit IRC | 18:16 | |
| *** ashishb has joined #openstack-ansible | 18:17 | |
| *** ashishb has quit IRC | 18:18 | |
| *** ashishb has joined #openstack-ansible | 18:18 | |
| *** abitha has joined #openstack-ansible | 18:22 | |
| *** ashishb has quit IRC | 18:23 | |
| *** ashishb has joined #openstack-ansible | 18:23 | |
| *** ashishb has quit IRC | 18:25 | |
| *** ashishb has joined #openstack-ansible | 18:26 | |
| *** ashishb has quit IRC | 18:27 | |
| *** ashishb_ has joined #openstack-ansible | 18:28 | |
| *** ashishb has joined #openstack-ansible | 18:28 | |
| *** ashishb has quit IRC | 18:28 | |
| *** ashishb_ has quit IRC | 18:28 | |
| *** ashishb has joined #openstack-ansible | 18:29 | |
| *** ashishb_ has joined #openstack-ansible | 18:29 | |
| *** ashishb has quit IRC | 18:29 | |
| *** ashishb_ has quit IRC | 18:29 | |
| *** ashishb has joined #openstack-ansible | 18:30 | |
| *** ashishb has quit IRC | 18:31 | |
| *** ashishb has joined #openstack-ansible | 18:31 | |
| *** rromans is now known as robb_afk | 18:33 | |
| *** cloudtrainme has joined #openstack-ansible | 18:35 | |
| *** ashishb_ has joined #openstack-ansible | 18:35 | |
| *** ashishb has quit IRC | 18:36 | |
| *** ysm has quit IRC | 18:40 | |
| *** sdake has joined #openstack-ansible | 18:40 | |
| *** ashishb_ has quit IRC | 18:56 | |
| *** galstrom is now known as galstrom_zzz | 19:25 | |
| *** KLevenstein_ has joined #openstack-ansible | 19:29 | |
| *** KLevenstein has quit IRC | 19:29 | |
| *** KLevenstein_ is now known as KLevenstein | 19:29 | |
| *** devl_ has quit IRC | 19:32 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: AIO bootstrap in Ansible https://review.openstack.org/239525 | 19:35 |
| *** subscope has joined #openstack-ansible | 19:39 | |
| *** ysm has joined #openstack-ansible | 19:40 | |
| *** galstrom_zzz is now known as galstrom | 19:46 | |
| *** cloudtrainme has quit IRC | 19:46 | |
| *** rebase has quit IRC | 19:52 | |
| *** daneyon has joined #openstack-ansible | 19:54 | |
| *** daneyon__ has joined #openstack-ansible | 19:55 | |
| *** daneyon_ has quit IRC | 19:56 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Avoid requesting get-pip.py multiple times https://review.openstack.org/243804 | 19:57 |
| *** daneyon has quit IRC | 19:58 | |
| bgmccollum | does nodepool use SSD flavors? | 19:59 |
| mhayden | for rax, it does | 19:59 |
| bgmccollum | hmpf...needs more speeds | 20:00 |
| *** sdake has quit IRC | 20:00 | |
| *** jimchou has quit IRC | 20:01 | |
| * palendae pushes turbo button | 20:01 | |
| bgmccollum | make it blink faster | 20:01 |
| bgmccollum | i wonder if you'd get better IO from CBS SSD vs local SSD | 20:03 |
| bgmccollum | not that we have control over those instances | 20:03 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Avoid requesting get-pip.py multiple times https://review.openstack.org/243804 | 20:03 |
| *** galstrom is now known as galstrom_zzz | 20:10 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Avoid requesting get-pip.py multiple times https://review.openstack.org/243804 | 20:18 |
| *** jimchou has joined #openstack-ansible | 20:20 | |
| *** jmckind_ is now known as jmckind | 20:26 | |
| *** galstrom_zzz is now known as galstrom | 20:28 | |
| *** phiche1 has quit IRC | 20:28 | |
| *** cloudtrainme has joined #openstack-ansible | 20:31 | |
| *** jmckind is now known as jmckind_ | 20:34 | |
| *** phiche has joined #openstack-ansible | 20:38 | |
| *** javeriak_ has quit IRC | 20:44 | |
| *** cloudtrainme has quit IRC | 20:48 | |
| *** devl_ has joined #openstack-ansible | 20:51 | |
| mhayden | cloudnull: i think i ticket all the boxes on your review except the mongo stuff https://review.openstack.org/#/c/239525/ | 20:52 |
| *** jaybeers has joined #openstack-ansible | 20:54 | |
| mhayden | ooh, just got my first "Jenkins is going to shut down" when i was watching an openstack ci test | 20:55 |
| mhayden | wonder if it's for the java serialization fun | 20:55 |
| *** galstrom is now known as galstrom_zzz | 20:58 | |
| openstackgerrit | Merged openstack/openstack-ansible: Fix race condition for /openstack directories https://review.openstack.org/242291 | 21:00 |
| *** galstrom_zzz is now known as galstrom | 21:01 | |
| jaybeers | I heard that it's possible to group Galera cluster nodes so that each group elects a single node to synchronize with other groups (to improve performance with geographically-distributed clusters); how would I go about configuring that? | 21:03 |
| *** sacharya has joined #openstack-ansible | 21:10 | |
| stevelle | jaybeers: such a feature is unknown to me, I would love to know more. | 21:23 |
| jaybeers | that makes two of us :) | 21:23 |
| jaybeers | I can't find any evidence of such a feature in the InterPhone | 21:24 |
| stevelle | when I looked into it, probably 4 or 5 months ago, I didn't find such a thing. | 21:24 |
| stevelle | But I very easily could have missed it. | 21:24 |
| jaybeers | but an extremely reliable source said it was a thing :) | 21:24 |
| *** rebase has joined #openstack-ansible | 21:26 | |
| *** Jezogwza has quit IRC | 21:27 | |
| stevelle | Please, ping me when you come up with something :) | 21:27 |
| *** cloudtrainme has joined #openstack-ansible | 21:37 | |
| *** KLevenstein has quit IRC | 21:40 | |
| *** KLevenstein_ has joined #openstack-ansible | 21:40 | |
| *** sacharya has quit IRC | 21:44 | |
| *** devl_ has quit IRC | 21:50 | |
| *** woodard_ has joined #openstack-ansible | 21:51 | |
| *** woodard has quit IRC | 21:54 | |
| *** jmckind has joined #openstack-ansible | 21:55 | |
| *** woodard_ has quit IRC | 21:56 | |
| *** jmckind has quit IRC | 21:56 | |
| *** jmckind has joined #openstack-ansible | 21:57 | |
| *** jmckind_ has quit IRC | 21:57 | |
| *** jmckind has quit IRC | 21:58 | |
| *** jmckind has joined #openstack-ansible | 21:58 | |
| *** jmckind has quit IRC | 21:59 | |
| *** jmckind has joined #openstack-ansible | 22:00 | |
| *** jmckind has quit IRC | 22:00 | |
| *** jmckind has joined #openstack-ansible | 22:01 | |
| *** jmckind has quit IRC | 22:03 | |
| *** sdake has joined #openstack-ansible | 22:03 | |
| *** jmckind has joined #openstack-ansible | 22:03 | |
| *** jmckind has quit IRC | 22:04 | |
| *** galstrom is now known as galstrom_zzz | 22:15 | |
| *** galstrom_zzz is now known as galstrom | 22:15 | |
| *** neillc_away is now known as neillc | 22:17 | |
| *** gouthamr has quit IRC | 22:20 | |
| *** galstrom is now known as galstrom_zzz | 22:24 | |
| *** ggillies has quit IRC | 22:27 | |
| *** sdake has quit IRC | 22:28 | |
| *** ggillies has joined #openstack-ansible | 22:29 | |
| *** rebase has quit IRC | 22:37 | |
| *** rebase has joined #openstack-ansible | 22:37 | |
| *** Mudpuppy has quit IRC | 22:42 | |
| *** ysm has quit IRC | 22:45 | |
| *** spotz is now known as spotz_zzz | 22:46 | |
| cloudnull | jaybeers: odyssey4me_ is the point of contact on that , ill have to look more into it however odyssey4me_ would know more. | 22:51 |
| jaybeers | ah, right on. Thanks for the tip! | 22:52 |
| jaybeers | we're not in a huge hurry over it, but if it is possible we'd like to investigate | 22:52 |
| cloudnull | yea we should nail that down as a configurable option | 22:55 |
| jaybeers | other than forking the repository and sending a pull request, is there anything I should do to start contributing to the repository? It's just a small change, but everything has a humble beginning I suppose :) | 22:56 |
| jaybeers | I want to variabilize (yup...that's the word I'm going with) the MariaDB datadir variable | 22:57 |
| cloudnull | in gerrit you need to setup your account then clone the repo, run ``git review -s`` | 22:58 |
| *** Jezogwza has joined #openstack-ansible | 22:58 | |
| jaybeers | ok | 22:58 |
| cloudnull | http://docs.openstack.org/contributor-guide/first-timers-quickstart.html | 22:58 |
| jaybeers | oh. well, that was...cleverly hidden. :P | 22:59 |
| *** phiche has quit IRC | 23:00 | |
| *** cloudtrainme has quit IRC | 23:04 | |
| *** eil397 has joined #openstack-ansible | 23:11 | |
| *** rady has quit IRC | 23:12 | |
| *** Jezogwza has quit IRC | 23:16 | |
| jaybeers | I notice the variable lxc_container_directory, which you could use to override /var/lib/lxc; is it possible to specify that value for only some servers? For instance, our db servers have a (relatively) small root partition, but have a large attached storage partition. It's not a standard directory location, so the other servers don't have it; on those machines, /var/lib/lxc is fine. | 23:24 |
| jaybeers | currently, we're solving the problem by creating a symlink from /var/lib/lxc to the target location; this seems to be working just fine, but it feels more elegant to actually configure LXC to look in the right place from the get-go :) | 23:25 |
| jaybeers | I guess another solution would be to create the directory on all the servers (even if it's just part of "/" and not actually a mount point) and point _all_ the servers there | 23:26 |
| *** alkari has joined #openstack-ansible | 23:29 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 23:29 | |
| jmccrory | you're running db containers on a separate host from other infra services? | 23:39 |
| jaybeers | well, they're on separate hosts than, say, the log server | 23:48 |
| stevelle | you want the db containers to use a different value for lxc_container_directory but the db containers are still on the same infra hosts? | 23:49 |
| jaybeers | so, ok...we have hosts that are running the db containers, and they have large storage attached because data | 23:49 |
| jaybeers | then we have other hosts that are running other containers, that don't have attached storage | 23:50 |
| jaybeers | the other shared infra containers can go in the shared storage on the db nodes, that's fine | 23:50 |
| jaybeers | but then there are entire other hosts with other containers...for example, what's in log_hosts and not in shared-infra_hosts | 23:51 |
| jaybeers | am I making sense? it makes sense to me, but then, I'm familiar with our environment :P | 23:51 |
| stevelle | I'm terrible at this :D | 23:52 |
| stevelle | I'll read a few times and see if I can work it out | 23:52 |
| jaybeers | so, setting aside containers | 23:52 |
| jaybeers | there are shared infra hosts, and logging hosts | 23:52 |
| jaybeers | physical machine | 23:52 |
| stevelle | and os-infra hosts, yes | 23:52 |
| jaybeers | on the shared infra hosts, where the db container is, there is attached storage | 23:53 |
| jaybeers | the servers that host the log server containers don't have shared storage | 23:53 |
| jaybeers | on the latter, we just install the containers to /var/lib/lxc 'cause it doesn't matter | 23:54 |
| *** woodard has joined #openstack-ansible | 23:54 | |
| jaybeers | on the former, the root partition isn't large enough to hold the containers, so we install them to the shared storage | 23:54 |
| jaybeers | in order to do that, we have to manually symlink /var/lib/lxc to the new location, because that directory doesn't exist on any of the other servers | 23:54 |
| jaybeers | which works fine, by the way | 23:55 |
| jaybeers | but if we could configure LXC on a per-group (or per-host, I guess) basis, then lxc.conf could point directly to the storage mount point, on those host machines that have one | 23:55 |
| jaybeers | the other solution would be to create that mount point directory on _all_ the nodes (even the ones that aren't actually mounting a filesystem there) | 23:57 |
| jaybeers | which would look _slightly_ silly, but perhaps less silly than a symlink | 23:57 |
| stevelle | not that this hasn't occurred to you but that would make everything behave consistently so possibly easier to support, and be easier to rig up in the config | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
