| *** mahito has joined #openstack-ansible | 00:00 | |
| *** Mudpuppy has joined #openstack-ansible | 00:21 | |
| *** sdake has joined #openstack-ansible | 00:23 | |
| *** sdake has quit IRC | 00:26 | |
| *** sdake has joined #openstack-ansible | 00:26 | |
| *** daneyon has joined #openstack-ansible | 00:47 | |
| *** daneyon_ has joined #openstack-ansible | 01:08 | |
| *** daneyon has quit IRC | 01:08 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Move SYMLINK_DIR declaration to bootstrap-aio.sh https://review.openstack.org/167316 | 01:20 |
|---|---|---|
| *** stevemar has quit IRC | 01:45 | |
| *** stevemar has joined #openstack-ansible | 01:46 | |
| *** sdake has quit IRC | 01:51 | |
| *** gregdek_ has quit IRC | 03:19 | |
| *** sandywalsh has quit IRC | 03:28 | |
| *** sandywalsh has joined #openstack-ansible | 03:37 | |
| *** sdake has joined #openstack-ansible | 03:49 | |
| *** sdake has quit IRC | 04:49 | |
| *** mrodden has quit IRC | 05:17 | |
| *** mrodden has joined #openstack-ansible | 05:31 | |
| *** vmtrooper has joined #openstack-ansible | 05:50 | |
| *** Mudpuppy has quit IRC | 06:05 | |
| *** vmtrooper has quit IRC | 06:10 | |
| *** vmtrooper has joined #openstack-ansible | 06:11 | |
| *** vmtrooper has quit IRC | 06:15 | |
| *** markvoelker has joined #openstack-ansible | 06:22 | |
| *** stevemar has quit IRC | 06:27 | |
| *** markvoelker has quit IRC | 06:27 | |
| *** stevemar has joined #openstack-ansible | 06:27 | |
| *** stevemar has quit IRC | 06:34 | |
| *** stevemar has joined #openstack-ansible | 06:34 | |
| *** Mudpuppy has joined #openstack-ansible | 06:36 | |
| *** Mudpuppy has quit IRC | 06:40 | |
| *** stevemar has quit IRC | 07:19 | |
| *** markvoelker has joined #openstack-ansible | 07:23 | |
| *** markvoelker has quit IRC | 07:27 | |
| svg | I have no direct internet access on my target hosts. I already mirrored apt repo's on out internal server, but the scripts still need to access a bunch of external servers. Is there a config option to use a proxy host for everything? | 07:37 |
| odyssey4me | svg not within os-ansible-deployment, no, but you should easily enough be able to use ansible to deploy proxy configuration for apt/wget/etc to all the hosts by deploying the required configs | 07:45 |
| svg | good point | 07:49 |
| *** vmtrooper has joined #openstack-ansible | 07:59 | |
| *** vmtrooper has quit IRC | 08:04 | |
| svg | hmz | 08:08 |
| svg | dpkg: regarding .../mariadb-client-core-5.5_5.5.41+maria-1~trusty_amd64.deb containing mariadb-client-core-5.5: | 08:08 |
| svg | mariadb-client-5.5 conflicts with mysql-client-core-5.5 | 08:08 |
| svg | mariadb-client-core-5.5 provides mysql-client-core-5.5 and is to be installed. | 08:08 |
| svg | that looks like sth is messed up in that mariadb repo | 08:08 |
| odyssey4me | svg it's more likely that the server installed the trusty package before the mariadb repo package - but it is possible that the mariadb repo has broken something again | 08:23 |
| openstackgerrit | git-harry proposed stackforge/os-ansible-deployment: Updated repository for minimum viable kilo install https://review.openstack.org/166986 | 08:24 |
| svg | odyssey4me: no, it had the mariadb version installed | 08:57 |
| svg | problem is obvious here: | 08:57 |
| svg | http://paste.ubuntu.com/10705761/ | 08:57 |
| svg | the mariadb version both provides and conflicts with mysql-client-5.5 ... | 08:58 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Remove support for ~/.openstack_deploy/ https://review.openstack.org/167010 | 09:22 |
| *** markvoelker has joined #openstack-ansible | 09:24 | |
| *** mahito_ has joined #openstack-ansible | 09:25 | |
| *** mahito has quit IRC | 09:28 | |
| *** markvoelker has quit IRC | 09:29 | |
| *** vmtrooper has joined #openstack-ansible | 09:48 | |
| *** vmtrooper has quit IRC | 09:53 | |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment-specs: Manage resolv.conf https://review.openstack.org/168074 | 09:54 |
| *** mahito_ has quit IRC | 10:04 | |
| *** markvoelker has joined #openstack-ansible | 10:25 | |
| *** markvoelker has quit IRC | 10:30 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Fix dedicated replication network logic https://review.openstack.org/167177 | 10:42 |
| *** jaypipes has joined #openstack-ansible | 11:17 | |
| *** markvoelker has joined #openstack-ansible | 11:26 | |
| *** markvoelker has quit IRC | 11:31 | |
| openstackgerrit | git-harry proposed stackforge/os-ansible-deployment: Update cinder services to kilo https://review.openstack.org/168066 | 11:33 |
| *** vmtrooper has joined #openstack-ansible | 11:37 | |
| *** nosleep77 has joined #openstack-ansible | 11:38 | |
| *** vmtrooper has quit IRC | 11:42 | |
| *** galstrom_zzz is now known as galstrom | 12:00 | |
| *** markvoelker has joined #openstack-ansible | 12:10 | |
| svg | Is there a mailing list for the rpc project? | 12:11 |
| odyssey4me | svg openstack-dev/openstack-operators would be the appropriate mailing lists to use | 12:15 |
| svg | ok thx | 12:15 |
| svg | I'm in the proces of playbooks/infrastructure/infrastructure-setup.yml (v10) - I notice that it fails each time it waits on the container's networking to come up, it then goes on after a playbook re-run - sounds like some timeout being too short? | 12:19 |
| odyssey4me | svg it could be, although we run the same process in the single node and multi node gate checks... so there might be something peculiar to your environment/config that's causing it to take longer to come up | 12:21 |
| svg | no idea, I'm testing on very recent machines (blades with 256GB Ram, 40 cores, ..) | 12:22 |
| odyssey4me | it'll possibly be something related to how the host networking is setup - maybe something to do with bonding or something else that would cause physical/virtual interface startup delays | 12:24 |
| odyssey4me | anyway, considering that it's easily worked around, I expect that it's not a crucial issue which is holding you back at all | 12:25 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment-specs: Manage resolv.conf https://review.openstack.org/168074 | 12:33 |
| svg | no, certainly not, just mentioning | 12:34 |
| mattt | svg: maybe you need to turn the turbo button off | 12:36 |
| mattt | :) | 12:37 |
| * svg looks | 12:37 | |
| mattt | svg: that sounds like a pretty sweet rig :) | 12:37 |
| *** sandywalsh has quit IRC | 12:37 | |
| svg | those are thoe hosts meant to be used for kvm/compute | 12:37 |
| svg | lxc-attach keeps yelling "$: command not found", for whatever container I try | 12:38 |
| *** sandywalsh has joined #openstack-ansible | 12:40 | |
| mattt | svg: i don't have an install in front of me right now but it sounds like you're using a non-root user | 12:41 |
| svg | a non-root user when issuing that command? | 12:42 |
| mattt | svg: yeah, did you deploy this all using root? | 12:43 |
| svg | localy a non privileged user on the ansible side, but logging in with root on the target hosts, yes | 12:44 |
| svg | trying the lxc-attach as root | 12:44 |
| mattt | svg: lxc-attach is in /usr/bin, so your user should see it irrespective of which user you are | 12:48 |
| svg | ouch, nm, that was a PEBCAK, sorry guy | 12:50 |
| svg | guys | 12:50 |
| mattt | svg: don't worry, my day is usually filled w/ those | 12:51 |
| svg | damn, rabbitmq site is slow as hell now | 13:01 |
| mattt | svg: https://review.openstack.org/#/c/167723/ | 13:02 |
| mattt | we ran into that issue last week | 13:02 |
| svg | aha | 13:05 |
| *** galstrom is now known as galstrom_zzz | 13:12 | |
| *** sdake has joined #openstack-ansible | 13:25 | |
| *** vmtrooper has joined #openstack-ansible | 13:26 | |
| *** sdake_ has joined #openstack-ansible | 13:28 | |
| *** vmtrooper has quit IRC | 13:31 | |
| *** sdake has quit IRC | 13:31 | |
| *** britthouser has joined #openstack-ansible | 13:39 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Allow swift stand alone to be installed for Juno https://review.openstack.org/167170 | 13:40 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: [WIP] Implement tunable keystone configuration https://review.openstack.org/165683 | 13:41 |
| *** britthouser has quit IRC | 13:46 | |
| *** britthouser has joined #openstack-ansible | 13:47 | |
| *** britthouser has quit IRC | 13:49 | |
| *** sdake has joined #openstack-ansible | 13:49 | |
| *** britthouser has joined #openstack-ansible | 13:50 | |
| *** stevemar has joined #openstack-ansible | 13:52 | |
| *** sdake_ has quit IRC | 13:53 | |
| *** KLevenstein has joined #openstack-ansible | 13:55 | |
| *** ccrouch has left #openstack-ansible | 13:55 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:09 | |
| sigmavirus24 | oh | 14:17 |
| * sigmavirus24 wonders if the GitHub DDoS is affecting RabbitMQ | 14:17 | |
| * sigmavirus24 checks who Rabbit uses for a CDN | 14:17 | |
| *** hj-hp has joined #openstack-ansible | 14:30 | |
| *** galstrom_zzz is now known as galstrom | 14:33 | |
| *** hj-hp has quit IRC | 14:33 | |
| *** Mudpuppy has joined #openstack-ansible | 14:34 | |
| *** hj-hp has joined #openstack-ansible | 14:38 | |
| *** ccrouch has joined #openstack-ansible | 14:49 | |
| svg | hm, so it seems my containers restart very quickyly (as in immediately), I can ssh into them practiclly as soon as the restart container task finishes, but the ansible wait_for task doesn't seem to work here, until I restart the playbook | 14:52 |
| *** jwagner_away is now known as jwagner | 14:55 | |
| *** alextricity has quit IRC | 15:09 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: [WIP] Implement tunable keystone configuration https://review.openstack.org/165683 | 15:10 |
| *** britthouser has quit IRC | 15:11 | |
| *** britthouser has joined #openstack-ansible | 15:12 | |
| *** vmtrooper has joined #openstack-ansible | 15:15 | |
| *** vmtrooper has quit IRC | 15:20 | |
| *** sdake_ has joined #openstack-ansible | 15:26 | |
| *** sdake has quit IRC | 15:30 | |
| *** alextricity has joined #openstack-ansible | 15:36 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Remove logstash multiline filter and change default number of workers https://review.openstack.org/165546 | 15:46 |
| *** galstrom is now known as galstrom_zzz | 15:48 | |
| cloudnull | palendae you're now core. congrats and sorry :) | 15:50 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment-specs: Tunable OpenStack Configuration Specification https://review.openstack.org/168976 | 15:51 |
| * palendae failed at not being seen | 15:51 | |
| d34dh0r53 | congrats palendae | 15:53 |
| mattt | palendae: congrats! | 15:56 |
| cloudnull | PSA: if you had a change that was dependent on https://review.openstack.org/#/c/166986/ (Minimal Kilo Patch) post git-harry 's changes you will need to rebase your dependent patch. | 15:57 |
| * mattt needs to read openstack-dev more often :( | 15:57 | |
| palendae | Thanks | 15:57 |
| mattt | cloudnull: what are we doing regarding osprofiler? | 15:57 |
| mattt | are we leaving it in api-paste.ini and adding toggles to allow it to be turned on/off? | 15:58 |
| cloudnull | nothing at this point. the paste configs from upstream we're brought in, wholesale, from upstream . | 15:58 |
| cloudnull | i'd say that we make it configurable. | 15:58 |
| cloudnull | and leave the default paste api untouched wherever possible. | 15:59 |
| Sam-I-Am | does osprofiler do anything without additional config in the service config file? | 16:00 |
| *** daneyon has joined #openstack-ansible | 16:07 | |
| stevelle | Sam-I-Am: are you asking for documented behavior of osprofiler or actual behavior? | 16:07 |
| stevelle | documented behavior is set up to be lazy-on | 16:08 |
| stevelle | so it should be mostly harmless | 16:08 |
| sigmavirus24 | why are we still talking about osprofiler? | 16:08 |
| Sam-I-Am | stevelle: i'm trying to make sure glance in the upstream install guide doesnt blast useless crap into rabbitmq | 16:09 |
| *** daneyon_ has quit IRC | 16:09 | |
| *** rrrobbb has joined #openstack-ansible | 16:10 | |
| *** rromans has quit IRC | 16:10 | |
| stevelle | Sam-I-Am: documented behavior says it wont | 16:10 |
| *** rromans has joined #openstack-ansible | 16:11 | |
| *** rromans has quit IRC | 16:19 | |
| *** rromans has joined #openstack-ansible | 16:19 | |
| sigmavirus24 | Sam-I-Am: osprofiler should be off by default in Glance | 16:26 |
| sigmavirus24 | Sam-I-Am: if it isn't, then you should file a bug so I can fix it | 16:26 |
| mattt | cloudnull: cool i'll do that then | 16:26 |
| mattt | sigmavirus24: did you write osprofiler | 16:27 |
| sigmavirus24 | ಠ_ಠ | 16:27 |
| Sam-I-Am | sigmavirus24: so, even if its in -paste.ini... | 16:34 |
| sigmavirus24 | Sam-I-Am: we only load one profile from paste.ini (flavor = ) | 16:34 |
| sigmavirus24 | Oh I see why you're asking: https://github.com/openstack/glance/blob/master/etc/glance-api-paste.ini#L15 | 16:35 |
| Sam-I-Am | sigmavirus24: yes | 16:35 |
| Sam-I-Am | ubuntu picked up that file | 16:35 |
| sigmavirus24 | https://github.com/openstack/glance/blob/master/etc/glance-api.conf#L459 | 16:35 |
| sigmavirus24 | False is the default iirc | 16:35 |
| sigmavirus24 | so while the middleware is there, it'll be non-operative iirc | 16:35 |
| Sam-I-Am | sigmavirus24: ok, that works | 16:35 |
| * sigmavirus24 double checks to be paranoid | 16:36 | |
| Sam-I-Am | looks to be disabled in ubuntu too | 16:36 |
| sigmavirus24 | Sam-I-Am: https://github.com/openstack/glance/blob/92180691248db1797591357e773166c0518bc0e0/glance/common/wsgi.py#L104..L109 | 16:36 |
| sigmavirus24 | False is the default | 16:37 |
| Sam-I-Am | woo | 16:38 |
| Sam-I-Am | aight, back to concentrating on the install guide | 16:38 |
| *** jlvillal_ has joined #openstack-ansible | 16:40 | |
| *** jlvillal_ has left #openstack-ansible | 16:41 | |
| *** KLevenstein has quit IRC | 16:44 | |
| *** galstrom_zzz is now known as galstrom | 16:48 | |
| *** jaypipes has quit IRC | 16:49 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Change heat_metadata_server_url to external API. https://review.openstack.org/164785 | 16:54 |
| *** rrrobbb has quit IRC | 16:55 | |
| *** rrrobbb has joined #openstack-ansible | 16:56 | |
| *** vmtrooper has joined #openstack-ansible | 17:03 | |
| *** KLevenstein has joined #openstack-ansible | 17:07 | |
| *** vmtrooper has quit IRC | 17:08 | |
| *** daneyon_ has joined #openstack-ansible | 17:10 | |
| *** daneyon has quit IRC | 17:10 | |
| alextricity | Not too familiar with the recent openstack_environement changes. I'm looking through the openstack_user_config.yml and I see repo-infra_hosts. What is this exactly? | 17:15 |
| alextricity | Does the frozen repo get copied over to these hosts, then maybe the playbooks use these repos to install pkgs and such? | 17:16 |
| *** daneyon has joined #openstack-ansible | 17:20 | |
| *** daneyon_ has quit IRC | 17:23 | |
| *** jmccrory has joined #openstack-ansible | 17:40 | |
| *** jaypipes has joined #openstack-ansible | 17:42 | |
| cloudnull | alextricity the repo bits are for the python source code. This allows you to maintain the python source code, built as python wheels, within your environment. | 17:56 |
| alextricity | cool! thanks cloudnull | 17:57 |
| palendae | alextricity: The repo is powered by https://github.com/cloudnull/yaprt. It looks though python requirements and builds a graph of dependency versions, then builds them and puts them in the repo | 18:02 |
| *** Mudpuppy_ has joined #openstack-ansible | 18:15 | |
| *** Mudpuppy has quit IRC | 18:15 | |
| alextricity | Oh cool. Also, what is the best way to destroy my containers and rebuild my inventory now? | 18:16 |
| cloudnull | you want to build all the things again ? | 18:17 |
| alextricity | yeah | 18:18 |
| cloudnull | ./scripts/teardown.sh | 18:18 |
| alextricity | I forget to put is_metal under "properties" in the environment.yml file :{ | 18:18 |
| *** Mudpuppy_ is now known as Mudpuppy | 18:20 | |
| *** Mudpuppy has quit IRC | 18:27 | |
| *** daneyon_ has joined #openstack-ansible | 18:27 | |
| *** Mudpuppy has joined #openstack-ansible | 18:28 | |
| *** daneyon has quit IRC | 18:28 | |
| alextricity | Dang...the teardown script ran ansible, then ansible failed parsing the JSON, proceeding to delete my inventory file. Doesn't look like it creates a backup either. | 18:29 |
| alextricity | So none of the containers got deleted. Maybe it would be a good idea to have the script fail if that happens? Does this look familiar to anyone: http://pastebin.com/RdeT0mw0 | 18:30 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Updated nova logic to correct issues with addon devices https://review.openstack.org/166578 | 18:30 |
| palendae | alextricity: Teardown definitely nukes everything | 18:31 |
| openstackgerrit | David Wilde proposed stackforge/os-ansible-deployment-specs: Initial commit of nova-kilofication https://review.openstack.org/169041 | 18:34 |
| *** KLevenstein_ has joined #openstack-ansible | 18:38 | |
| *** KLevenstein has quit IRC | 18:41 | |
| *** KLevenstein_ is now known as KLevenstein | 18:41 | |
| *** vmtrooper has joined #openstack-ansible | 18:52 | |
| *** vmtrooper has quit IRC | 18:58 | |
| *** jwagner is now known as jwagner_away | 19:00 | |
| *** jwagner_away is now known as jwagner | 19:01 | |
| *** jwagner is now known as jwagner_lunch | 19:02 | |
| openstackgerrit | Miguel Grinberg proposed stackforge/os-ansible-deployment: WIP Horizon config changes for kilo https://review.openstack.org/167460 | 19:09 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Improve Galera running, stop, bootstrap behavior https://review.openstack.org/169049 | 19:12 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Improve Galera running, stop, bootstrap behavior https://review.openstack.org/169049 | 19:16 |
| *** jwagner_lunch is now known as jwagner | 19:45 | |
| *** Mudpuppy has quit IRC | 19:56 | |
| *** rrrobbb has quit IRC | 20:05 | |
| *** Mudpuppy has joined #openstack-ansible | 20:06 | |
| *** Mudpuppy has quit IRC | 20:06 | |
| *** Mudpuppy has joined #openstack-ansible | 20:06 | |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Ensure LXC bridge exists, is up, and has rules https://review.openstack.org/169069 | 20:14 |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Keystone config changes for Kilo https://review.openstack.org/167475 | 20:16 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Improve Galera running, stop, bootstrap behavior https://review.openstack.org/169049 | 20:17 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Ensure LXC bridge exists, is up, and has rules https://review.openstack.org/169069 | 20:17 |
| sigmavirus24 | stevelle: cloudnull so we're splitting out the minimal kilo patch? | 20:35 |
| stevelle | I'm attempting to smooth the ramp with the bits that can be pulled out | 20:36 |
| cloudnull | sigmavirus24 yup. | 20:37 |
| cloudnull | stevelle for prez! | 20:37 |
| sigmavirus24 | Send pez to stevelle ? Okay! | 20:38 |
| cloudnull | the yak needed more shaving. | 20:38 |
| *** vmtrooper has joined #openstack-ansible | 20:41 | |
| *** vmtrooper has quit IRC | 20:47 | |
| *** KLevenstein has quit IRC | 20:50 | |
| *** jmccrory has quit IRC | 20:56 | |
| *** jmccrory has joined #openstack-ansible | 21:00 | |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Improve Galera running, stop, bootstrap behavior https://review.openstack.org/169049 | 21:04 |
| *** jmccrory1 has joined #openstack-ansible | 21:04 | |
| *** jmccrory1 has quit IRC | 21:04 | |
| *** jmccrory has quit IRC | 21:07 | |
| *** sdake has joined #openstack-ansible | 21:18 | |
| *** nosleep77 has quit IRC | 21:20 | |
| *** sdake_ has quit IRC | 21:21 | |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Improve Galera running, stop, bootstrap behavior https://review.openstack.org/169049 | 21:26 |
| palendae | stevelle: Did you figure out why it was failing the tests? | 21:31 |
| stevelle | palendae: no | 21:31 |
| stevelle | palendae: have you been able to do any more for the scenario tests? | 21:35 |
| palendae | stevelle: I haven't been able to get a case where it's not crashing on network stuff | 21:36 |
| palendae | Apsu: Have you been able to narrow down the kernel issue? Is it a linux bridge thing? | 21:36 |
| stevelle | cloudnull: was mentioning some of the API calls returning 200 instead of expected 202, would be cool to isolate that from the "intermittent" issues | 21:37 |
| palendae | I'm seeing the time outs, not 200/202 | 21:37 |
| Apsu | palendae: Actually I believe the kernel is returning from the netlink/ioctl/syscalls involved in link creation/bridge enslaving before the process is complete. | 21:37 |
| palendae | Which is weird, cause I'm turning off network scenarios | 21:37 |
| Apsu | I wrote a fuzzing script to slam it with bridge/port operations and test the state after every single step, with adjustable delays | 21:38 |
| Apsu | And I'm finding that all kinds of points in time, ip link will return 0 and yet the thing you just did is *not* in effect immediately afterwards. | 21:38 |
| Apsu | There's delays that vary in timing. | 21:38 |
| palendae | Does 'the thing you just did' finish? | 21:39 |
| palendae | Like, does it become true, just outside of that window between when you called it and you checked? | 21:39 |
| Apsu | Yes. | 21:40 |
| Apsu | It does. | 21:40 |
| Apsu | Which is slightly different than not ending up in a bridge. But | 21:40 |
| Apsu | I just finished the script and I'm testing A) dummy port types instead of veth pairs and B) without network namespaces (but still in the agents container) | 21:40 |
| Apsu | So my next step is to test veth pairs hooking into containerized netns's | 21:41 |
| *** jwagner is now known as jwagner_away | 21:41 | |
| Apsu | Interestingly enough... when you don't create a lot of interfaces to add at once, it doesn't seem to fail, even running in parallel at fastest speed | 21:42 |
| Apsu | But if you make a lot of interfaces to add at once, invariably there's some kind of delay in there | 21:42 |
| Apsu | Adding port 'br-5.port-3' to bridge 'br-5' | 21:43 |
| Apsu | -- Checking port 'br-5.port-3' master is bridge 'br-5' | 21:43 |
| Apsu | -- Port 'br-5.port-3' not in bridge 'br-5'! Exiting | 21:43 |
| Apsu | ^ one of three or four "doesn't exist right after successfully doing the thing" errors occurring without delays inserted. | 21:44 |
| Apsu | So... I almost think we might need some non-busy-wait loops in neutron to verify certain actions. | 21:44 |
| palendae | So bringing that up to the tempest level, the workers (and probably even just the tests when run in serial) are creating stuff faster than the kernel is registering | 21:44 |
| Apsu | Essentially yeah, because neutron's LBA is entirely convinced that if ip/brctl return 0, what you told them to do is FULLY done. | 21:45 |
| Apsu | And it's not even close. | 21:45 |
| *** KLevenstein has joined #openstack-ansible | 21:52 | |
| stevelle | so neutron is a bit naive, expecting synchronous execution. | 21:53 |
| *** daneyon has joined #openstack-ansible | 21:58 | |
| *** daneyon_ has quit IRC | 21:59 | |
| stevelle | palendae: made it over 30m w/o failure on the patch this time, so it will probably pass. the following patch just needs rebase clicked after that, afaik. | 22:00 |
| *** Mudpuppy has quit IRC | 22:04 | |
| *** KLevenstein has quit IRC | 22:25 | |
| *** vmtrooper has joined #openstack-ansible | 22:30 | |
| *** vmtrooper has quit IRC | 22:35 | |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Ensure LXC bridge exists, is up, and has rules https://review.openstack.org/169069 | 22:36 |
| *** galstrom is now known as galstrom_zzz | 22:54 | |
| *** markvoelker has quit IRC | 22:57 | |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Updated repository for minimum viable kilo install https://review.openstack.org/166986 | 23:09 |
| cloudnull | stevelle this is the commit that deps on your two changes thus far. ^ | 23:10 |
| cloudnull | thanks for doing all of that btw. :) | 23:10 |
| palendae | stevelle, cloudnull: Left a comment on https://review.openstack.org/#/c/169069/ just so I understand what the new lxc bridge stuff is doing | 23:15 |
| *** stevemar has quit IRC | 23:18 | |
| *** logan2 has quit IRC | 23:25 | |
| *** mattoliverau has quit IRC | 23:29 | |
| stevelle | palendae: will bang on it again | 23:29 |
| *** mattoliverau has joined #openstack-ansible | 23:29 | |
| *** logan2 has joined #openstack-ansible | 23:33 | |
| *** bluebox has joined #openstack-ansible | 23:37 | |
| bluebox | hey guys, there seems to be only a /kilo under the specs repo tree. Where would it be appropiate to push specs generally for master | 23:38 |
| *** hj-hp has quit IRC | 23:39 | |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Ensure LXC bridge exists, is up, and has rules https://review.openstack.org/169069 | 23:41 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Ensure LXC bridge exists, is up, and has rules https://review.openstack.org/169069 | 23:43 |
| *** bluebox is now known as javeriak | 23:44 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Improve Galera running, stop, bootstrap behavior https://review.openstack.org/169049 | 23:45 |
| javeriak | /msg NickServ VERIFY REGISTER javeriak kcemobngxnln | 23:45 |
| *** markvoelker has joined #openstack-ansible | 23:53 | |
| *** mahito has joined #openstack-ansible | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!