| *** markvoelker has quit IRC | 00:02 | |
| *** sdake_ has joined #openstack-ansible | 00:04 | |
| *** shoutm has quit IRC | 00:06 | |
| *** sdake has quit IRC | 00:07 | |
| *** sdake has joined #openstack-ansible | 00:20 | |
| *** sdake_ has quit IRC | 00:23 | |
| *** lkoranda has quit IRC | 01:01 | |
| *** lkoranda has joined #openstack-ansible | 01:02 | |
| *** woodard has joined #openstack-ansible | 01:07 | |
| *** woodard has quit IRC | 01:11 | |
| *** markvoelker has joined #openstack-ansible | 01:58 | |
| *** markvoelker has quit IRC | 02:03 | |
| *** logan2 has quit IRC | 02:18 | |
| *** markvoelker has joined #openstack-ansible | 02:31 | |
| *** Mudpuppy has quit IRC | 02:32 | |
| *** logan2 has joined #openstack-ansible | 02:57 | |
| *** JRobinson__ is now known as JRobinson__afk | 03:02 | |
| *** Mudpuppy has joined #openstack-ansible | 03:26 | |
| *** Mudpuppy has quit IRC | 03:30 | |
| *** JRobinson__afk has quit IRC | 04:08 | |
| *** britthouser has joined #openstack-ansible | 04:11 | |
| *** britthou_ has joined #openstack-ansible | 04:12 | |
| *** britthouser has quit IRC | 04:15 | |
| *** openstack has joined #openstack-ansible | 04:20 | |
| *** markvoelker has quit IRC | 04:31 | |
| *** JRobinson__afk has joined #openstack-ansible | 04:46 | |
| *** JRobinson__afk is now known as JRobinson__ | 04:46 | |
| *** promethe1nfire is now known as prometheanfire | 04:50 | |
| *** neillc has joined #openstack-ansible | 04:59 | |
| neillc | hello | 05:00 |
|---|---|---|
| *** daneyon_ has quit IRC | 05:13 | |
| *** daneyon has joined #openstack-ansible | 05:14 | |
| *** shausy has joined #openstack-ansible | 05:24 | |
| *** markvoelker has joined #openstack-ansible | 05:32 | |
| *** markvoelker has quit IRC | 05:36 | |
| *** shausy has quit IRC | 05:38 | |
| *** shausy has joined #openstack-ansible | 05:39 | |
| *** javeriak has joined #openstack-ansible | 06:30 | |
| *** sdake_ has joined #openstack-ansible | 06:44 | |
| *** sdake has quit IRC | 06:47 | |
| *** javeriak has quit IRC | 06:49 | |
| *** javeriak_ has joined #openstack-ansible | 06:49 | |
| *** javeriak has joined #openstack-ansible | 06:56 | |
| *** javeriak_ has quit IRC | 06:57 | |
| *** shoutm has joined #openstack-ansible | 07:01 | |
| odyssey4me | morning all | 07:01 |
| odyssey4me | svg master is unfortunately failing 7/10 times on build - usually in the tempest tests... I'm trying to figure out the cause. It may be resource related, or it may be code related (the underlying code is constantly in flux)... anyway, just so you know that you're not the only one getting failed builds | 07:02 |
| -openstackstatus- NOTICE: Gerrit is currently under very high load and may be unresponsive. infra are looking into the issue. | 07:05 | |
| *** sdake_ has quit IRC | 07:08 | |
| evrardjp | good morning everyone | 07:31 |
| *** markvoelker has joined #openstack-ansible | 07:33 | |
| *** markvoelker has quit IRC | 07:37 | |
| *** vdo has joined #openstack-ansible | 07:47 | |
| *** gparaskevas_ has joined #openstack-ansible | 07:52 | |
| odyssey4me | o/ evrardjp | 07:53 |
| *** shackit has joined #openstack-ansible | 07:56 | |
| *** JRobinson__ has quit IRC | 07:59 | |
| *** gparaskevas_ has quit IRC | 08:06 | |
| *** shausy has quit IRC | 08:10 | |
| *** shausy has joined #openstack-ansible | 08:10 | |
| *** javeriak has quit IRC | 08:26 | |
| *** javeriak has joined #openstack-ansible | 08:30 | |
| *** marekd_404 has quit IRC | 08:54 | |
| *** shausy has quit IRC | 09:06 | |
| odyssey4me | andymccr are you happy with the changes to https://review.openstack.org/212452 from mattt ? | 09:07 |
| mattt | whenever we bump rabbit we should put a copy in http://mirror.rackspace.com/rackspaceprivatecloud/downloads/ | 09:08 |
| mattt | rabbitmq downloads are currently failing from the official site | 09:08 |
| odyssey4me | mattt oh? | 09:09 |
| odyssey4me | mattt there is a review in to bump the version | 09:09 |
| mattt | 3.5.3 which we install in master isn't in /rackspaceprivatecloud/downloads/ tho | 09:10 |
| *** javeriak has quit IRC | 09:10 | |
| *** javeriak has joined #openstack-ansible | 09:10 | |
| odyssey4me | mattt does the task remove the deb? if not then download manually, rename and run the tasks | 09:12 |
| mattt | i can't download manually :P | 09:12 |
| svg | good morning | 09:12 |
| odyssey4me | mattt or override the version/URL with one that is on the rax mirror? | 09:12 |
| odyssey4me | svg morning! | 09:12 |
| svg | odyssey4me: thanks. yes, when i see it's about some timeout, i expect I'm not the culprit :) | 09:13 |
| odyssey4me | svg nope - it seems to be some sort of issue in cinder/nova, but I've yet to dig deep enough to isolate the issue | 09:13 |
| svg | sounds like fun | 09:13 |
| odyssey4me | FYI: http://lists.openstack.org/pipermail/openstack-dev/2015-August/072185.html | 09:18 |
| evrardjp | this brings more questions to how mirror.rackspace.com is managed with private cloud efforts, right? | 09:19 |
| evrardjp | I don't know if it's related but we could be mirror of your rpc-repo.rackspace.com/container_images/ too | 09:21 |
| evrardjp | or for everything under http://rpc-repo.rackspace.com for that matter | 09:21 |
| odyssey4me | evrardjp yes, that it is | 09:22 |
| odyssey4me | http://mirror.rackspace.com/rackspaceprivatecloud is a mirror of http://rpc-repo.rackspace.com | 09:22 |
| evrardjp | the rpc-repo is I guessed managed by a small team, that has other stuff to do than mirrors, compared to mirror.rackspace.com? | 09:23 |
| evrardjp | (just to understand) | 09:23 |
| odyssey4me | I've been meaning to get rid of our reliance on the rackspace mirror for ubuntu packages as that's hurting more than helping - we often hit the mirror during package updates. we should rather be using whatever mirror is provided on the host image. We implemented a common mirror to try to work around issues we get in hpcloud-b4, but it hasn't helped. | 09:24 |
| evrardjp | k | 09:24 |
| odyssey4me | evrardjp the RPC dev team manages the RPC repo | 09:24 |
| evrardjp | that makes sense | 09:24 |
| odyssey4me | it's primarily a place to host the python wheels for RPC | 09:24 |
| evrardjp | yeah I understand what it does | 09:25 |
| evrardjp | I was wondering how it was managed | 09:25 |
| odyssey4me | ie it's where the repo-build play gets run for each tagged version - and RPC uses the repo-sync to pull the wheels down | 09:25 |
| odyssey4me | ah, mostly by hand at this point | 09:25 |
| odyssey4me | the plan is to get rid of the dependency on it down the line | 09:25 |
| evrardjp | it would be far more secure | 09:26 |
| odyssey4me | evrardjp this one on dependency removal: https://blueprints.launchpad.net/openstack-ansible/+spec/remove-upstream-repo-dependency | 09:26 |
| odyssey4me | another is to implement the image building in a role instead of downloading an upstream image | 09:26 |
| evrardjp | yeah there are tools for it | 09:27 |
| evrardjp | and auditing is better | 09:27 |
| evrardjp | but it takes longer to deploy | 09:27 |
| evrardjp | there are lots of work to do everywhere ;) | 09:28 |
| odyssey4me | yes, there's a ton of work | 09:28 |
| *** markvoelker has joined #openstack-ansible | 09:33 | |
| *** markvoelker has quit IRC | 09:38 | |
| *** c0m0 has joined #openstack-ansible | 09:42 | |
| -openstackstatus- NOTICE: review.openstack.org (aka gerrit) is going down for an emergency restart | 10:19 | |
| *** ChanServ changes topic to "review.openstack.org (aka gerrit) is going down for an emergency restart" | 10:19 | |
| svg | ok, gerrit is officially not very stable now | 10:20 |
| *** marekd has joined #openstack-ansible | 10:39 | |
| *** marekd is now known as marekd_404 | 10:40 | |
| *** ashishb has joined #openstack-ansible | 10:43 | |
| *** ChanServ changes topic to "Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible" | 10:46 | |
| -openstackstatus- NOTICE: Gerrit restart has resolved the issue and systems are back up and functioning | 10:46 | |
| *** daneyon has quit IRC | 10:59 | |
| *** daneyon has joined #openstack-ansible | 11:00 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Use slurp to get the content of the ceph.conf file https://review.openstack.org/213170 | 11:15 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Implement /usr/bin/env as the shebang in all bash scripts https://review.openstack.org/211885 | 11:20 |
| *** markvoelker has joined #openstack-ansible | 11:34 | |
| *** markvoelker has quit IRC | 11:39 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Implement /usr/bin/env as the shebang in all bash scripts https://review.openstack.org/211885 | 11:52 |
| *** benwh4 has joined #openstack-ansible | 11:59 | |
| benwh4 | hello everybody ! I have a question ... I try to set up a openstack cloud using OSAD but after the creation of the containers I have an error message saying that ssh could not be use into the containers but I could ssh into my target hosts | 12:00 |
| benwh4 | and I wonder if I should fill the file into etc/network for my target hosts or I should on the tagert hosts set up the interfaces files manually ? | 12:01 |
| evrardjp | what do you mean by "ssh could not be use"? | 12:03 |
| evrardjp | could you copy paste the output, this way we could help you better? | 12:03 |
| evrardjp | (maybe on a pastebin, like the openstack one) | 12:04 |
| evrardjp | your target hosts should be network configured | 12:04 |
| evrardjp | these are the host configuration guidelines: http://osad.readthedocs.org/en/latest/install-guide/targethosts.html | 12:04 |
| evrardjp | sometimes ansible fails because of the amount of connections, don't hesitate to rerun your playbooks | 12:05 |
| benwh4 | fatal: [infra1_keystone_container-2d97108f -> infra1] => SSH Error: data could not be sent to the remote host. Make sure this host can be reached over ssh | 12:05 |
| evrardjp | ok | 12:06 |
| evrardjp | this happens sometimes | 12:06 |
| evrardjp | just rereun your playbook | 12:06 |
| evrardjp | I meant: run your playbook again | 12:06 |
| evrardjp | sorry for my english | 12:06 |
| svg | I confirm, that happens quite a lot. a rerun and then it just works. And in the next task it could fail again. | 12:07 |
| benwh4 | but I didn't reboot the target hosts after configuring the network but I wonder if the configurations must be set up on the /etc/net/interfaces on the hosts directly or into the directory on the os-ansible-deployment | 12:07 |
| svg | I never managed to understand why that happens. | 12:07 |
| evrardjp | svg: me neither | 12:08 |
| evrardjp | ControlMaster maybe | 12:08 |
| svg | why? | 12:08 |
| svg | it is fishy though, that should not happen | 12:08 |
| evrardjp | yup | 12:08 |
| evrardjp | did you try fireball mode? | 12:08 |
| benwh4 | so what is the /opt/os-ansible-deployment/etc/network for ? | 12:08 |
| evrardjp | I should try | 12:08 |
| evrardjp | this is used as an example | 12:09 |
| evrardjp | IIRC | 12:09 |
| evrardjp | if you follow the guide on the link I sent you, you'll eventually arrive here: | 12:09 |
| benwh4 | ok it is not meant to be fill with the same changes as on the target hosts ? | 12:09 |
| evrardjp | http://osad.readthedocs.org/en/latest/install-guide/targethosts-network.html | 12:09 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Add nova_libvirt_live_migration_flag variable https://review.openstack.org/212452 | 12:10 |
| evrardjp | by default, you shouldn't edit your /opt/os-ansible-deployment/ | 12:10 |
| evrardjp | the code arrives there, so it's not meant for configuration... that happens on the /etc/openstack_deploy/ | 12:11 |
| evrardjp | (I meant the configuration stands in /Etc/openstack_deploy/ folder) | 12:12 |
| evrardjp | but you need to contact your future cloud hosts on a network, and that is what you have to define manually... | 12:13 |
| evrardjp | which means: your future cloud hosts have to be network configured | 12:13 |
| evrardjp | this way you can ssh onto them and then deploy the cloud | 12:13 |
| evrardjp | using ansible | 12:13 |
| evrardjp | provided by the project | 12:14 |
| svg | That setup needs to be run by root. Why? | 12:15 |
| evrardjp | svg: ? | 12:16 |
| *** woodard has joined #openstack-ansible | 12:17 | |
| mgariepy | good morning everyone | 12:17 |
| evrardjp | good morning mgariepy | 12:17 |
| evrardjp | are the rabbitmq code/binaries uploaded on your repos, dear rackspace guys? :) | 12:18 |
| evrardjp | latest* | 12:18 |
| evrardjp | svg: I've a question for you, for ansible... In roles/repo_server/tasks/repo_sync_manager.yml there is a when | 12:20 |
| evrardjp | under a notify | 12:20 |
| svg | (deploying and working with osad assumes you run it with the root user) | 12:21 |
| evrardjp | it means the task runs when the "when" condition is met, right? Not that the "when" triggers for the notify event | 12:21 |
| svg | evrardjp: you mean there is a task with a condigitional and a notify | 12:22 |
| benwh4 | another thing ... why there is a aio1 configuration running when the playbooks are set up for a normal deployment ? | 12:22 |
| evrardjp | svg: yes, so the when is only for the task, not for the notify | 12:22 |
| svg | yes, the conditional controls task exectution, the changed status controls the notify | 12:22 |
| evrardjp | is there a way to have that only for notify event? | 12:22 |
| evrardjp | except by using a wrapper script | 12:22 |
| svg | you can put a cnditional in the handler | 12:22 |
| evrardjp | that's what I meant | 12:23 |
| evrardjp | so when in the handler section works | 12:23 |
| evrardjp | cool | 12:23 |
| svg | or you can use changed_when: as that is what controls the changed status which will fire the handler | 12:23 |
| svg | you can even add a notify in a task | 12:23 |
| evrardjp | ofc | 12:24 |
| svg | ... in a handler | 12:24 |
| svg | I mean | 12:24 |
| evrardjp | I understood ;) | 12:24 |
| svg | and have a chain reaction | 12:24 |
| svg | handlers are tasks basically | 12:24 |
| evrardjp | that are triggered at the end | 12:24 |
| evrardjp | Ok | 12:24 |
| svg | yes | 12:24 |
| evrardjp | I think I'll remove the haproxy restart logic | 12:24 |
| evrardjp | it's a pain | 12:25 |
| evrardjp | it's not only done at the end | 12:25 |
| evrardjp | I guess when internal and external vip are different, there are post_ tasks that run after the notify event of the role has been sent | 12:26 |
| evrardjp | (IIRC) | 12:26 |
| evrardjp | anyway I'll check if I could cleanup this haproxy setup | 12:26 |
| evrardjp | to be a better fit with keepalived | 12:26 |
| *** woodard has quit IRC | 12:27 | |
| *** gparaskevas has joined #openstack-ansible | 12:27 | |
| *** woodard has joined #openstack-ansible | 12:28 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Allow cinder-backup to use ceph https://review.openstack.org/209537 | 12:40 |
| *** cfarquhar has quit IRC | 12:42 | |
| svg | Would it be possible to add an osad specific topic to http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev? and/or -operators? | 12:53 |
| *** tlian has joined #openstack-ansible | 12:54 | |
| *** tlian has quit IRC | 12:54 | |
| evrardjp | you think of sending with [OSAD] in title? | 12:58 |
| odyssey4me | svg just use the tag [openstack-ansible] in your subject line | 12:58 |
| odyssey4me | you'll find similar tags in the history | 12:58 |
| svg | yes, but given those tags are not registered in mailman, you can't say "I only want those mails" | 12:59 |
| svg | Which one can at least on the -dev ML for other topics | 12:59 |
| odyssey4me | benwh4 the ssh issue appears to be related to using too many forks - in 11.1.1 we've included a patch to remove the forks setting in ansible.cfg so that it goes back to the default of 5 (instead of 15) | 12:59 |
| svg | sure, I could handle that with a rule, but doing it it mailman would be cleaner | 13:00 |
| *** sdake has joined #openstack-ansible | 13:00 | |
| svg | odyssey4me: any idea why 15 is too many? I sometimes krank up the forks (for small runs in other projects) and never had an issue with that | 13:00 |
| *** tlian has joined #openstack-ansible | 13:00 | |
| odyssey4me | svg the two tags mostly used are '[os-ansible-deployment]' and '[openstack-ansible]' | 13:01 |
| *** ashishb has quit IRC | 13:02 | |
| svg | I guess you just proved my point | 13:02 |
| odyssey4me | I use filters to mark emails on -dev and -operators that I'm not interested in as read. | 13:02 |
| odyssey4me | svg - thw two tags are because we're between names :p | 13:02 |
| svg | Thanks for answering, but I already refuted it :) | 13:03 |
| odyssey4me | svg not sure why 15 forks is too many. I've recently applied the # of forks as being the same as the number of cpu's and that seems to have worked well | 13:03 |
| odyssey4me | (sorry, catching up on backscroll)\ | 13:03 |
| svg | makes sense | 13:04 |
| evrardjp | I've reduced and I have the feeling it's worse than before | 13:04 |
| odyssey4me | https://github.com/stackforge/os-ansible-deployment/blob/master/scripts/scripts-library.sh#L27 | 13:04 |
| *** britthouser has joined #openstack-ansible | 13:05 | |
| svg | either way, there is imho no good reason why the number of forks should interfere with connecion timeouts | 13:05 |
| svg | (unless they would be really too big) | 13:05 |
| odyssey4me | benwh4 how do you mean you have an aio setup running? have you setup a multi-node environment, or an aio environment? | 13:06 |
| odyssey4me | svg agreed, but after spending several weeks trying to figure it out we eventually have up and implemented the 'successerator' which automatically re-runs the playbooks for the gate check | 13:06 |
| odyssey4me | more recently we reduced the forks, which seems to have reduce the ssh connection failure incidents | 13:07 |
| svg | sure, just saying that sounds like a side effect, not the root cause | 13:07 |
| odyssey4me | with ansible v2 a patch we PR'd to ansible will have an automated retry in the ssh pipeline which will further reduce the issues | 13:07 |
| odyssey4me | if you can figure out the root cause, that's be awesome :) | 13:07 |
| *** sdake_ has joined #openstack-ansible | 13:15 | |
| benwh4 | when I run the setup-hosts playbook I have ... | 13:18 |
| benwh4 | ok: [aio1_heat_engine_container-edeb3aaa -> aio1] | 13:18 |
| *** sdake has quit IRC | 13:19 | |
| benwh4 | plus ok: [infra3_heat_apis_container-e77f9892 -> infra3] | 13:19 |
| *** sdake_ has quit IRC | 13:19 | |
| benwh4 | is it normal to have aio1 running when I want a mutli node target ? | 13:19 |
| odyssey4me | benwh4 it sounds to me like you've left the aio hosts in your openstack_user_config | 13:20 |
| odyssey4me | what branch/tag are you using for your installation? | 13:21 |
| benwh4 | I use the master tag | 13:21 |
| *** leakypipes is now known as jaypipes | 13:25 | |
| *** javeriak has quit IRC | 13:31 | |
| *** Mudpuppy has joined #openstack-ansible | 13:34 | |
| *** markvoelker has joined #openstack-ansible | 13:35 | |
| *** britthou_ has joined #openstack-ansible | 13:36 | |
| *** Mudpuppy has quit IRC | 13:38 | |
| *** britthouser has quit IRC | 13:39 | |
| *** markvoelker has quit IRC | 13:39 | |
| mgariepy | odyssey4me, mattt , this patch seems really hard to get through. https://review.openstack.org/#/c/213170/ | 13:46 |
| mattt | mgariepy: it's not that review specifically | 13:48 |
| mattt | we're seeing a lot of gate failures in master at the minute | 13:48 |
| *** sdake has joined #openstack-ansible | 13:49 | |
| *** sdake has quit IRC | 13:49 | |
| *** javeriak has joined #openstack-ansible | 13:49 | |
| *** sdake has joined #openstack-ansible | 13:49 | |
| odyssey4me | benwh4 did you copy the aio openstack_user_config for your setup? | 13:50 |
| odyssey4me | benwh4 essentially https://github.com/stackforge/os-ansible-deployment/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio#L72 the name there is the name referred to for the host which has the IP address of https://github.com/stackforge/os-ansible-deployment/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio#L77 | 13:51 |
| odyssey4me | the name can be anything | 13:51 |
| odyssey4me | the playbooks implement /etc/hosts entries for them | 13:51 |
| benwh4 | no I used the openstack_user_config.yml not the aio | 13:52 |
| odyssey4me | benwh4 you should perhaps read through http://osad.readthedocs.org/en/latest/install-guide/configure.html | 13:52 |
| benwh4 | I will wait until the playbooks is finish then double check again | 13:53 |
| odyssey4me | benwh4 well, the aio name will only come from an inventory reference as defined in openstack_user_config.yml - so you must have referred to a host as aio1 | 13:53 |
| odyssey4me | either that or you have an inventory with entries from a previous test which has hosts called aio1 | 13:53 |
| benwh4 | ok I will look into to see if I found an aio entry and remove it | 13:54 |
| odyssey4me | benwh4 you'd have to then use the inventory-manage script to remove the host/containers from the inventory | 13:55 |
| *** KLevenstein has joined #openstack-ansible | 14:04 | |
| *** woodard has quit IRC | 14:04 | |
| *** woodard has joined #openstack-ansible | 14:05 | |
| *** sdake has quit IRC | 14:10 | |
| *** shoutm has quit IRC | 14:12 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:15 | |
| *** sdake has joined #openstack-ansible | 14:17 | |
| sigmavirus24 | mornings all | 14:19 |
| *** Mudpuppy has joined #openstack-ansible | 14:19 | |
| svg | Good afternoon sigmavirus24 :) | 14:24 |
| sigmavirus24 | How're you svg? | 14:25 |
| svg | chilling! and you? | 14:25 |
| *** phalmos has joined #openstack-ansible | 14:26 | |
| *** gparaskevas has quit IRC | 14:26 | |
| odyssey4me | andymccr mattt sigmavirus24 palendae d34dh0r53 please review this backport asap https://review.openstack.org/213442 | 14:28 |
| odyssey4me | then the SHA update to follow it: https://review.openstack.org/213439 | 14:29 |
| *** jwagner is now known as jwagner_away | 14:30 | |
| sigmavirus24 | odyssey4me: lolno =P | 14:30 |
| mattt | odyssey4me: your wife must love you | 14:31 |
| mattt | i'll have some tea, LIKE RIGHT THIS SECOND | 14:31 |
| svg | 8-| | 14:31 |
| evrardjp | :) | 14:31 |
| evrardjp | too much redbull | 14:31 |
| odyssey4me | hahaha | 14:31 |
| odyssey4me | oooh evrardjp good idea! | 14:32 |
| evrardjp | but he gets results! | 14:32 |
| mattt | svg: you back to work this week? :) | 14:32 |
| svg | nope | 14:32 |
| svg | the problem is, I have wifi here | 14:32 |
| evrardjp | bad idea! | 14:32 |
| evrardjp | but it's in france, it doesn't mean that you'll have decent uplink to the interwebz | 14:32 |
| mattt | svg: still raining this week? | 14:32 |
| svg | but, but, the pool water is too damn cold today! | 14:33 |
| svg | no, but cloudy | 14:33 |
| evrardjp | am I the only one to think that haproxy should be restarted only when it could bind to its correct ip address ? | 14:33 |
| evrardjp | (so I plan to remove all the handlers "restart haproxy" and create a task that checks if haproxy could bind, and then only restart | 14:34 |
| evrardjp | ) | 14:34 |
| *** javeriak has quit IRC | 14:37 | |
| mgariepy | is there documentation somewhere to add external network in osad ? | 14:37 |
| *** britthou_ has quit IRC | 14:39 | |
| *** britthouser has joined #openstack-ansible | 14:39 | |
| odyssey4me | mgariepy like http://osad.readthedocs.org/en/latest/install-guide/targethosts-network.html and http://osad.readthedocs.org/en/latest/install-guide/configure-networking.html perhaps? | 14:40 |
| mgariepy | odyssey4me, well, br-vlan/vxlan are there to host the tenant networks, i would like to know how to add the "external network" | 14:44 |
| mgariepy | i guess that i need to add a network on neutron_linuxbridge_agent, to be able to assign floating ips to routers and tenants vms. | 14:44 |
| odyssey4me | mgariepy Sam-I-Am can likely help | 14:44 |
| odyssey4me | I think it's typically done by adding a flat network, but it all depends on how you want to do it really | 14:45 |
| mgariepy | basicly : br-ex on hosts > ethX interface in neutron containers. | 14:46 |
| mgariepy | does linuxbridge allow to use DVR with neutron ? | 14:47 |
| *** phalmos has quit IRC | 14:49 | |
| odyssey4me | mgariepy as I recall DVR is still somewhat broken in Kilo when using linuxbridge - there are upstream efforts to make it work better | 14:49 |
| odyssey4me | don't quote me on that though, I'm not the expert on anything neutron | 14:50 |
| *** markvoelker has joined #openstack-ansible | 14:51 | |
| mgariepy | lol ok | 14:51 |
| sigmavirus24 | odyssey4me: I'm reviewing your backport now that I'm off that video call | 14:55 |
| *** markvoelker has quit IRC | 14:55 | |
| mgariepy | odyssey4me, is the comment: 'recheck test_volumes_backup.VolumesBackupsV1Test,test_minimum_basic_scenario' remove the "VolumesBackupsV1Test" tempest check from the gate-os-ansible-deployment-dsvm-commit ? | 14:58 |
| odyssey4me | mgariepy nope, it's just a note on why the recheck is being done - the words after 'recheck' are a comment so that we can see if there's a pattern | 14:58 |
| palendae | mattt: No, that won't remove it. 'recheck' triggers the whole test suite to recheck, that's just a note | 14:58 |
| odyssey4me | clearly there is, so we need to figure out why the volume backup test is failing | 14:58 |
| mgariepy | i saw that https://review.openstack.org/#/c/211543/ had the same issue, and when jenkins recheck it, this test wasn't included. | 15:00 |
| palendae | Interesting | 15:00 |
| *** phalmos has joined #openstack-ansible | 15:00 | |
| odyssey4me | mgariepy the volumev1test was included: http://logs.openstack.org/43/211543/3/check/gate-os-ansible-deployment-dsvm-commit/e908b51/console.html.gz#_2015-08-13_15_10_02_599 | 15:01 |
| *** KLevenstein has quit IRC | 15:02 | |
| *** KLevenstein has joined #openstack-ansible | 15:03 | |
| benwh4 | how do you use the inventory-manage.py ? | 15:05 |
| *** KLevenstein has quit IRC | 15:07 | |
| mattt | benwh4: you can run it without args, it will spit out some help | 15:07 |
| mgariepy | hmm looks like i misread the logs.. | 15:08 |
| mattt | benwh4: essentially, there's --list to list what's in your inventory, or --remove to remove entries | 15:08 |
| evrardjp | so odyssey4me: are you interested by a conditional restart of haproxy instead of a definitive restart? | 15:14 |
| *** KLevenstein has joined #openstack-ansible | 15:14 | |
| odyssey4me | evrardjp sure, although I'm more interested in haproxy's services being able to be configured/deconfigured for various deployments. | 15:15 |
| *** markvoelker has joined #openstack-ansible | 15:15 | |
| odyssey4me | ie If I'm doing a swift-only deployment, I'd like haproxy to only deploy configuration for the services that are in the environment. | 15:15 |
| evrardjp | damn that was what I dropped | 15:17 |
| *** spotz_zzz is now known as spotz | 15:17 | |
| *** javeriak has joined #openstack-ansible | 15:20 | |
| *** javeriak has quit IRC | 15:25 | |
| *** javeriak has joined #openstack-ansible | 15:25 | |
| odyssey4me | evrardjp as a separate patch I'd like to see the ability to deploy user-provided certificates for HAproxy - similar to how we do it for Horizon | 15:26 |
| odyssey4me | and yes, high availability would be nice - especially if haproxy only runs on the host being used at the time (ie it shuts down when not in use) so that it reduces the log noise from the service up checks | 15:27 |
| *** javeriak_ has joined #openstack-ansible | 15:29 | |
| *** javeriak has quit IRC | 15:30 | |
| evrardjp | odyssey4me: how would you see that a component is enabled or not? based on the presence of a valid file in env.d, or in a user_variable? | 15:41 |
| evrardjp | conf.d not env.d | 15:41 |
| palendae | evrardjp: Generally, we've been detecting whether or not the hosts/containers are defined for it | 15:41 |
| evrardjp | so based on inventory | 15:42 |
| palendae | Yes | 15:42 |
| palendae | Let me find an example | 15:42 |
| odyssey4me | evrardjp yeah, inventory would seem to be a good option - however that only caters for something that's already defined in vars/haproxy | 15:42 |
| odyssey4me | what about when we want to add a new service, for example elasticsearch | 15:43 |
| odyssey4me | we need to define the confgs for the LB for that service, but also only implement the config if there's a container in the inventory for it | 15:43 |
| evrardjp | I should create a blueprint first, this way we can think of how we do it, right? | 15:43 |
| palendae | odyssey4me: RPC implements elasticsearch as a new container/in inventory... | 15:43 |
| palendae | evrardjp: Yeah, that'd be ideal | 15:43 |
| odyssey4me | palendae yes, but it implements the LB for elasticsearch through a hack | 15:44 |
| palendae | Ah | 15:44 |
| odyssey4me | it doesn't use the current haproxy role | 15:44 |
| palendae | We should unhack that :) | 15:44 |
| palendae | Oh, no | 15:44 |
| palendae | Because RPC assumes F5s | 15:44 |
| palendae | Because there's some bias against haproxy | 15:44 |
| odyssey4me | sure, but that functionality is something we need | 15:44 |
| palendae | I agree | 15:44 |
| odyssey4me | for dev/test purposes and for gating | 15:44 |
| palendae | Fair enough | 15:44 |
| evrardjp | this could make it provider independant | 15:45 |
| evrardjp | ok | 15:45 |
| palendae | evrardjp: yeah, I'd say BP it | 15:45 |
| odyssey4me | evrardjp I would suggest trying to work up a prototype to figure out how much needs to change | 15:45 |
| palendae | odyssey4me: If you don't have any high priority reviews, I'm gonna look at BPs to review | 15:45 |
| odyssey4me | then write a blueprint based on your PoC and within the spec we can debate the merits of the approach and if anyone objects they can suggest an alternative | 15:45 |
| odyssey4me | palendae uh https://review.openstack.org/#/q/starredby:%22Jesse+Pretorius%22+project:stackforge/os-ansible-deployment,n,z | 15:46 |
| palendae | Thanks | 15:46 |
| evrardjp | a lot, compared to the current role, but this isn't something really difficult I'd say | 15:46 |
| odyssey4me | I would recommend ignoring the build success/failure at this point - most of them are unrelated to the patch | 15:46 |
| odyssey4me | palendae there are a lot of small reviews in flight, so once you've tackled the starred ones then feel free to just run through the normal list | 15:48 |
| odyssey4me | we're well behind :/ | 15:48 |
| palendae | I know | 15:48 |
| palendae | We're behind all round | 15:48 |
| *** benwh4 has quit IRC | 15:52 | |
| *** b3rnard0 is now known as b3rnard0_away4de | 15:54 | |
| *** woodard has quit IRC | 15:56 | |
| *** woodard has joined #openstack-ansible | 16:00 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Add iptables rule to neutron agents containers https://review.openstack.org/213442 | 16:00 |
| *** yaya has joined #openstack-ansible | 16:01 | |
| *** yaya has quit IRC | 16:02 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Enable debug logging for gate checks https://review.openstack.org/213438 | 16:03 |
| *** yaya has joined #openstack-ansible | 16:04 | |
| *** alop has joined #openstack-ansible | 16:08 | |
| *** daneyon has quit IRC | 16:30 | |
| *** daneyon has joined #openstack-ansible | 16:30 | |
| *** ashishb has joined #openstack-ansible | 16:39 | |
| *** woodard has quit IRC | 16:42 | |
| *** c0m0 has quit IRC | 16:45 | |
| *** b3rnard0_away4de is now known as b3rnard0 | 16:45 | |
| *** vdo has quit IRC | 16:46 | |
| *** woodard has joined #openstack-ansible | 16:46 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment-specs: Add spec for Independent Role Repositories https://review.openstack.org/213779 | 16:48 |
| palendae | odyssey4me: So, I feel weird voting on https://review.openstack.org/#/c/212919/ :p | 16:48 |
| odyssey4me | lol - well, I took it over from you - note that if you like? | 16:48 |
| palendae | Yep | 16:49 |
| odyssey4me | palendae ^ spec for your reading pleasure :) | 16:49 |
| *** woodard has quit IRC | 16:50 | |
| *** woodard has joined #openstack-ansible | 16:52 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Adds a pep8 target to tox.ini https://review.openstack.org/207662 | 16:53 |
| *** javeriak has joined #openstack-ansible | 16:56 | |
| *** javeriak_ has quit IRC | 16:57 | |
| *** sdake_ has joined #openstack-ansible | 17:20 | |
| *** sdake has quit IRC | 17:24 | |
| *** sdake has joined #openstack-ansible | 17:31 | |
| *** sdake_ has quit IRC | 17:35 | |
| *** wmlynch has joined #openstack-ansible | 17:35 | |
| *** woodard has quit IRC | 17:40 | |
| *** javeriak_ has joined #openstack-ansible | 17:40 | |
| *** javeriak has quit IRC | 17:41 | |
| *** woodard has joined #openstack-ansible | 17:46 | |
| *** jwagner_away is now known as jwagner | 17:46 | |
| *** yaya has quit IRC | 18:05 | |
| *** woodard has quit IRC | 18:11 | |
| *** ashishb has quit IRC | 18:21 | |
| *** prometheanfire has quit IRC | 18:32 | |
| *** prometheanfire has joined #openstack-ansible | 18:32 | |
| *** d34dh0r53 is now known as VD | 18:42 | |
| *** britthou_ has joined #openstack-ansible | 18:43 | |
| *** VD is now known as Guest69442 | 18:43 | |
| *** britthouser has quit IRC | 18:45 | |
| *** Guest69442 is now known as d34dh0r53 | 18:47 | |
| *** jmccrory has quit IRC | 19:12 | |
| *** yaya has joined #openstack-ansible | 19:34 | |
| *** woodard has joined #openstack-ansible | 19:42 | |
| *** yaya has quit IRC | 19:56 | |
| *** KLevenstein has quit IRC | 19:59 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Keystone SSL cert/key distribution and configuration https://review.openstack.org/194474 | 19:59 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Removes trailing whitespace for bashate https://review.openstack.org/207663 | 20:06 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Replace ADFS example DNS name with something appropriate https://review.openstack.org/211102 | 20:09 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Use slurp to get the content of the ceph.conf file https://review.openstack.org/213170 | 20:10 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Allow cinder-backup to use ceph https://review.openstack.org/209537 | 20:13 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Add nova_libvirt_live_migration_flag variable https://review.openstack.org/212452 | 20:13 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Implement /usr/bin/env as the shebang in all bash scripts https://review.openstack.org/211885 | 20:14 |
| *** daneyon has quit IRC | 20:14 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Add support for additional nova.conf options https://review.openstack.org/210492 | 20:15 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Update the documented ceph user variables https://review.openstack.org/209130 | 20:15 |
| *** spotz is now known as spotz_zzz | 20:28 | |
| *** yaya has joined #openstack-ansible | 20:29 | |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Update SHAs for kilo development https://review.openstack.org/213853 | 20:35 |
| Apsu | So many reviews omgherd | 20:48 |
| *** javeriak_ has quit IRC | 20:52 | |
| *** Mudpuppy has quit IRC | 20:55 | |
| sigmavirus24 | lol | 20:55 |
| *** Mudpuppy has joined #openstack-ansible | 21:01 | |
| *** Mudpuppy has quit IRC | 21:05 | |
| *** britthou_ has quit IRC | 21:14 | |
| *** woodard_ has joined #openstack-ansible | 21:14 | |
| *** woodard has quit IRC | 21:17 | |
| *** woodard_ has quit IRC | 21:18 | |
| *** Mudpuppy has joined #openstack-ansible | 21:29 | |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Update SHAs for kilo development https://review.openstack.org/213853 | 21:32 |
| *** shackit has quit IRC | 21:43 | |
| *** Mudpuppy has quit IRC | 21:45 | |
| *** tlian has quit IRC | 22:04 | |
| *** shoutm has joined #openstack-ansible | 22:06 | |
| *** sdake_ has joined #openstack-ansible | 22:10 | |
| *** sdake has quit IRC | 22:13 | |
| *** JRobinson__ has joined #openstack-ansible | 22:23 | |
| *** shoutm has quit IRC | 22:32 | |
| *** yaya has quit IRC | 22:32 | |
| *** shoutm has joined #openstack-ansible | 22:36 | |
| *** markvoelker has quit IRC | 22:51 | |
| *** sdake_ is now known as sdake | 23:01 | |
| *** sdake_ has joined #openstack-ansible | 23:22 | |
| *** sdake has quit IRC | 23:26 | |
| *** jwagner is now known as jwagner_away | 23:32 | |
| *** markvoelker has joined #openstack-ansible | 23:35 | |
| *** phalmos has quit IRC | 23:47 | |
| *** JRobinson__ has quit IRC | 23:52 | |
| *** britthouser has joined #openstack-ansible | 23:57 | |
| *** britthou_ has joined #openstack-ansible | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!