| *** sdake_ has joined #openstack-ansible | 00:00 | |
| *** sdake__ has joined #openstack-ansible | 00:10 | |
| *** jmccrory has quit IRC | 00:11 | |
| *** sdake_ has quit IRC | 00:14 | |
| *** vmtrooper has joined #openstack-ansible | 01:30 | |
| *** vmtrooper has quit IRC | 01:35 | |
| *** markvoelker has joined #openstack-ansible | 02:51 | |
| *** daneyon has joined #openstack-ansible | 03:03 | |
| *** vmtrooper has joined #openstack-ansible | 03:19 | |
| *** sdake__ has quit IRC | 03:23 | |
| *** vmtrooper has quit IRC | 03:24 | |
| *** sdake_ has joined #openstack-ansible | 03:47 | |
| *** markvoelker has quit IRC | 03:53 | |
| *** markvoelker has joined #openstack-ansible | 03:53 | |
| *** markvoelker has quit IRC | 03:57 | |
| *** markvoelker has joined #openstack-ansible | 04:24 | |
| *** markvoelker has quit IRC | 04:41 | |
| sdake_ | quick development question | 04:55 |
|---|---|---|
| sdake_ | in playbooks/roles/os_heat/defaults/main.yml | 04:55 |
| sdake_ | is this the file that creates the heat container? | 04:56 |
| *** vmtrooper has joined #openstack-ansible | 05:08 | |
| *** vmtrooper has quit IRC | 05:12 | |
| *** stevemar has quit IRC | 05:31 | |
| *** markvoelker has joined #openstack-ansible | 05:38 | |
| *** vmtrooper has joined #openstack-ansible | 06:56 | |
| *** vmtrooper has quit IRC | 07:01 | |
| *** openstackgerrit has quit IRC | 07:35 | |
| *** openstackgerrit has joined #openstack-ansible | 07:36 | |
| *** vmtrooper has joined #openstack-ansible | 08:45 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Set backend_alg for swift_proxy to source https://review.openstack.org/162914 | 08:46 |
| *** vmtrooper has quit IRC | 08:51 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Backport 94ba3bc to juno https://review.openstack.org/162917 | 09:07 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Backport 51abd0b to juno branch https://review.openstack.org/162922 | 09:17 |
| *** markvoelker has quit IRC | 09:22 | |
| *** sdake_ has quit IRC | 10:09 | |
| *** sdake_ has joined #openstack-ansible | 10:14 | |
| *** vmtrooper has joined #openstack-ansible | 10:34 | |
| *** vmtrooper has quit IRC | 10:39 | |
| *** sdake_ has quit IRC | 10:51 | |
| *** markvoelker has joined #openstack-ansible | 10:53 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Set backend_alg for swift_proxy to source https://review.openstack.org/162967 | 11:23 |
| mattt | 3rd time lucky | 11:23 |
| cloudnull | Morning. | 11:48 |
| cloudnull | Sdake_ the containers are created in the lxc_container_create role. | 11:49 |
| odyssey4me | cloudnull: in preparation for an addition to the contributing guidelines, we've got an etherpad to put down some points of view: https://etherpad.openstack.org/p/os-ansible-deployment-backports | 11:54 |
| odyssey4me | the idea is to be a bit more explicit about how backports should be done to settle confusion around cherry picks, mostly | 11:55 |
| cloudnull | Nice. I'll give that a look. | 11:55 |
| *** vmtrooper has joined #openstack-ansible | 12:23 | |
| *** vmtrooper has quit IRC | 12:28 | |
| *** markvoelker has quit IRC | 12:38 | |
| *** markvoelker has joined #openstack-ansible | 12:39 | |
| *** markvoelker has quit IRC | 12:43 | |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Revert "Trim ansible's output to console.log" https://review.openstack.org/162984 | 12:47 |
| *** stevemar has joined #openstack-ansible | 13:06 | |
| odyssey4me | haha - I've solved the dict merge issue for the tunable blueprint | 13:09 |
| odyssey4me | check it out: https://gist.github.com/odyssey4me/bb1ae8d6f8b6c00a730a | 13:09 |
| *** markvoelker has joined #openstack-ansible | 13:13 | |
| openstackgerrit | git-harry proposed stackforge/os-ansible-deployment: WIP: Testing hash_behaviour https://review.openstack.org/162989 | 13:17 |
| *** stevemar has quit IRC | 13:18 | |
| *** stevemar has joined #openstack-ansible | 13:19 | |
| odyssey4me | git-harry ^ | 13:20 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 13:25 | |
| *** sdake_ has joined #openstack-ansible | 13:49 | |
| sdake_ | morning | 13:51 |
| sdake_ | if i were to add a conditional variable for installing from source vs installing from packages, do folks have a variable name preference | 13:51 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Restore tempest install branch to pre-refactor SHA https://review.openstack.org/163006 | 13:52 |
| odyssey4me | sdake_ not particularly - I would guess that it'd be discussed properly within a review submitted | 13:53 |
| sdake_ | odyssey4me thanks i'll make something up then | 13:53 |
| odyssey4me | something like openstack_install_method | 13:53 |
| odyssey4me | or considering that each role is separate, perhaps glance_install_method and others for each role | 13:54 |
| odyssey4me | ie namespaced to the role | 13:54 |
| sdake_ | the problem with namespaced to the role is it would have to be configured in a bunch of palces right? | 13:56 |
| mattt | sdake_: would there be a case where you'd want one service installed via packages and the rest via source, or vice-versa? | 13:58 |
| sdake_ | matt I cannot think of any | 13:58 |
| sdake_ | either you will want all source or all packages | 13:58 |
| sdake_ | the all packages use case comes from "I want support from RHT, so I must use their packages" | 13:58 |
| sdake_ | or ubuntu in same scenario | 13:59 |
| *** Mudpuppy has joined #openstack-ansible | 14:00 | |
| odyssey4me | sdake_ sure, it does mean managing it per role - but the user can then override each role individually | 14:00 |
| sdake_ | odyssey4me so this is probably stuff that hsould be hashed in a review so folks dont lose cotext :) | 14:01 |
| odyssey4me | the alternative is to use one variable - something like openstack_deploy_method - and use that same variable across all roles, but provide a default per role | 14:01 |
| sdake_ | but thanks for the initial suggestion | 14:01 |
| odyssey4me | yup, agreed :) | 14:01 |
| *** alextricity has quit IRC | 14:03 | |
| odyssey4me | note that for 'debug' or 'verbose' we kind-of do that | 14:03 |
| odyssey4me | for each role, there are defaults | 14:03 |
| odyssey4me | but we also set a value in https://github.com/stackforge/os-ansible-deployment/blob/master/playbooks/inventory/group_vars/all.yml which the inventory overrides | 14:04 |
| odyssey4me | (I think) | 14:04 |
| *** vmtrooper has joined #openstack-ansible | 14:12 | |
| sdake_ | odyssey4me is there an easy way to do an all in one install with only 3 NICs? | 14:14 |
| *** alextricity has joined #openstack-ansible | 14:14 | |
| odyssey4me | the AOI config we have is designed for only 1 NIC | 14:14 |
| sdake_ | nice | 14:14 |
| odyssey4me | but you can adjust it by simply adjusting your interfaces configuration file | 14:15 |
| sdake_ | i've read most of hte code base and I didn't come across a doc that describes launching aio | 14:15 |
| sdake_ | but admittidly i read quickly | 14:15 |
| odyssey4me | scripts/bootstrap-aio.sh does the AIO host setup | 14:15 |
| sdake_ | just to get a feel for structure | 14:15 |
| sdake_ | cool thanks! | 14:15 |
| odyssey4me | https://github.com/stackforge/os-ansible-deployment/blob/master/development-stack.rst | 14:16 |
| odyssey4me | it's entirely plausible that we may have some inaccurate bits there - so please do submit patches for any fixes you find are necessary | 14:17 |
| *** vmtrooper has quit IRC | 14:17 | |
| *** markvoelker has quit IRC | 14:37 | |
| *** markvoelker has joined #openstack-ansible | 14:37 | |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Add oslo.context to tempest packages https://review.openstack.org/163025 | 14:40 |
| *** markvoelker has quit IRC | 14:42 | |
| palendae | andymccr: On https://review.openstack.org/#/c/163006/, what are some of the ramifications of pinning the SHA that are unexpected? | 14:42 |
| andymccr | palendae: the reason for the SHA change was to go back to requirements that were added (oslo.context) but i think its not a direct requirement and is required by another package that is already required in the SHA | 14:43 |
| andymccr | so it wont actually change anything there. | 14:43 |
| palendae | I've been changing the SHA when doing neutron tempest tests to make sure I actually get tests that exercise the code | 14:43 |
| palendae | fwiw, when I changed it in my environments, I set tempest_git_install_branch in user_variables.yml and got that SHA | 14:44 |
| andymccr | we should probably set it to something. so it might be worth still making but im not sure it will solve the problem we have now | 14:45 |
| palendae | Gotcha | 14:46 |
| palendae | I do know on master, it did not have the same scenario tests as the SHA | 14:46 |
| *** markvoelker has joined #openstack-ansible | 14:49 | |
| *** jaypipes has quit IRC | 14:53 | |
| *** BjoernT has joined #openstack-ansible | 14:58 | |
| BjoernT | Morning | 14:58 |
| palendae | Morning BjoernT | 14:59 |
| b3rnard0 | morning | 14:59 |
| *** stevemar2 has joined #openstack-ansible | 15:01 | |
| *** stevemar has quit IRC | 15:01 | |
| BjoernT | lets do this | 15:01 |
| b3rnard0 | welcome to another edition of the bug triage | 15:01 |
| b3rnard0 | notes will be here: https://etherpad.openstack.org/p/openstack_ansible_bug_triage.2015-03-10-16.00 | 15:02 |
| d34dh0r53 | mornings | 15:02 |
| b3rnard0 | d34dh0r53 and odyssey4me will help with this meeting as cloudnull is at the philly conference | 15:02 |
| b3rnard0 | okay, who has a bug they would like to discuss | 15:03 |
| odyssey4me | oh dear, time zone confusion... I was expecting it to be in another hour :) | 15:04 |
| d34dh0r53 | my phone says it should be at 11 | 15:04 |
| d34dh0r53 | which is in an hour | 15:04 |
| odyssey4me | anyway - morning everyone | 15:04 |
| b3rnard0 | oh damn you are right. it's set to utc | 15:05 |
| d34dh0r53 | yeah | 15:05 |
| b3rnard0 | BjoernT, can you attend at 11? | 15:05 |
| BjoernT | unfortunately no, I mail you if I have urgent bugs. From my side it's worth talking about spice and windows/linux issues | 15:06 |
| d34dh0r53 | we can hold the main meeting now and I can run another at 11 for UTC | 15:07 |
| b3rnard0 | BjoernT: links please | 15:08 |
| odyssey4me | this is the triage list: https://bugs.launchpad.net/openstack-ansible/+bugs?search=Search&field.status=New&orderby=importance&start=0 | 15:09 |
| odyssey4me | starting with https://bugs.launchpad.net/openstack-ansible/+bug/1430071 | 15:10 |
| openstack | Launchpad bug 1430071 in openstack-ansible "Swift MaaS Replication Checks Failing - Traceback - KeyError: 'replication_type'" [Undecided,New] | 15:10 |
| odyssey4me | andymccr any comment here? | 15:10 |
| odyssey4me | looks like a maas bug to me - needs to be moved to the maas/rpc repo? | 15:11 |
| andymccr | yeh it would have to be moved out of os-ansible-deployment | 15:11 |
| d34dh0r53 | +1 | 15:11 |
| andymccr | on the legitimacy or not of the bug, im not sure. | 15:11 |
| odyssey4me | right: https://bugs.launchpad.net/openstack-ansible/+bug/1429947 | 15:12 |
| openstack | Launchpad bug 1429947 in openstack-ansible "openstack_interface.cfg.example has incorrect configuration for br-vlan" [Undecided,New] | 15:12 |
| palendae | odyssey4me: I believe that's valid. Apsu, thoughts? ^ | 15:13 |
| odyssey4me | yup, this is valid - although low priority I would argue as the .example is effectively documentation | 15:13 |
| odyssey4me | any volunteers to get it fixed quickly? this is low hanging fruit | 15:13 |
| BjoernT | b3rnard0: https://bugs.launchpad.net/openstack-ansible/+bug/1428833 | 15:13 |
| openstack | Launchpad bug 1428833 in openstack-ansible "Add novnc console support in favor of spice" [Undecided,New] | 15:13 |
| d34dh0r53 | right, that could also be tagged low-hanging-fruit | 15:13 |
| palendae | odyssey4me: I can grab it | 15:13 |
| odyssey4me | great, thanks palendae | 15:13 |
| palendae | Assigned | 15:14 |
| odyssey4me | BjoernT let's finish the new bug triage before we go into anything else (if we have time) | 15:14 |
| BjoernT | sure | 15:14 |
| BjoernT | it was you fyi | 15:14 |
| odyssey4me | next: https://bugs.launchpad.net/openstack-ansible/+bug/1429201 | 15:14 |
| openstack | Launchpad bug 1429201 in openstack-ansible "Cinder.conf Default Availability Zone Config Name Incorrect" [Undecided,New] | 15:14 |
| BjoernT | thats a bug yes | 15:15 |
| odyssey4me | for icehouse, juno and master? | 15:16 |
| BjoernT | juno and higher | 15:16 |
| stevelle | low hanging fruit, also | 15:16 |
| odyssey4me | master is fine: https://github.com/stackforge/os-ansible-deployment/blob/master/playbooks/roles/os_cinder/templates/cinder.conf.j2#L59 | 15:16 |
| odyssey4me | any volunteers to fix this for juno then? | 15:17 |
| b3rnard0 | we are no longer updating icehouse, correct (except for super critical issues)? so any bugs will have to go into juno and higher | 15:17 |
| odyssey4me | any thoughts on severity here? | 15:18 |
| bgmccollum | basically if you dont specify an AZ, the volume creation fails | 15:18 |
| bgmccollum | annoying in dashboard since "any" is the default selection | 15:18 |
| BjoernT | icehouse does not have az cinder capabilities in ansbile so Juno is ok | 15:18 |
| odyssey4me | so this is actually critical then, or high at least | 15:18 |
| BjoernT | +1 | 15:19 |
| palendae | Yeah, I agree | 15:19 |
| odyssey4me | vounteers? | 15:19 |
| stevelle | x | 15:19 |
| odyssey4me | alright stevelle, allocated | 15:20 |
| odyssey4me | next one: https://bugs.launchpad.net/openstack-ansible/+bug/1428833 | 15:20 |
| openstack | Launchpad bug 1428833 in openstack-ansible "Add novnc console support in favor of spice" [Undecided,New] | 15:20 |
| BjoernT | Yeah that's a tough one, I guess we need some devs support. I don't think the zombie issues in websockify will fix the horrendous windows experience | 15:21 |
| odyssey4me | well, this is essentially a wishlist item - but we can prioritise | 15:22 |
| BjoernT | I will test windows 2k12 instances with the latest qxl driver soon, redhat updated the driver if I saw correctly | 15:22 |
| openstackgerrit | Nolan Brubaker proposed stackforge/os-ansible-deployment: Fix the br-vlan mode in the example interfaces https://review.openstack.org/163044 | 15:22 |
| BjoernT | This issue is actually pretty serious for customers | 15:22 |
| odyssey4me | all the issues mentioned really need upstream fixes, and andymccr has been working on that | 15:22 |
| BjoernT | Upstream has most of the issues NOT fixed | 15:23 |
| BjoernT | just websockify | 15:23 |
| odyssey4me | a suitable workaround for the project is to implement novnc - to at least provide options to workaround | 15:23 |
| odyssey4me | this would be a new role and new features | 15:23 |
| d34dh0r53 | yeah, I think we probably need to allow the option of novnc | 15:23 |
| BjoernT | right, I still advocate to spend few days looking into the issues before giving up spice | 15:23 |
| odyssey4me | BjoernT yes, the fixes are needed upstream... that's the point | 15:24 |
| odyssey4me | any volunteers to do this for master first? to backport this to juno will need to be a subsequent decision | 15:24 |
| openstackgerrit | Nolan Brubaker proposed stackforge/os-ansible-deployment: Fix the br-vlan mode in the example interfaces https://review.openstack.org/163048 | 15:24 |
| bgmccollum | can spice and novnc be run at the same time? | 15:26 |
| odyssey4me | bgmccollum perhaps - they'd have to be deployed in seperate containers | 15:26 |
| BjoernT | no they cant | 15:26 |
| BjoernT | they are mutually exclusive configure inside the nova.conf, we can run the containers still at the same time | 15:27 |
| alextricity | pulled from the 10.1.2 tag yesterday and just noticed the /etc/rpc_deploy/rpc_user_config.yml file is built for an AIO. whoops | 15:27 |
| odyssey4me | andymccr you volunteering to setup a novnc role? | 15:27 |
| andymccr | odyssey4me: sure | 15:27 |
| odyssey4me | alextricity yes, that's by design - the .example file is the documentation | 15:28 |
| odyssey4me | right, next: https://bugs.launchpad.net/openstack-ansible/+bug/1428639 | 15:29 |
| openstack | Launchpad bug 1428639 in openstack-ansible "Nova requires python-libguestfs in Kilo" [Undecided,New] | 15:29 |
| palendae | I think any 'in Kilo' bugs should be associated with their appropriate blueprint | 15:29 |
| andymccr | +1 | 15:29 |
| odyssey4me | sure, when the blueprint has been registered | 15:29 |
| odyssey4me | importance? | 15:29 |
| odyssey4me | medium for now? | 15:30 |
| palendae | Yeah | 15:30 |
| d34dh0r53 | medium or high, not sure if this blows up every time | 15:30 |
| odyssey4me | it's still a month until the first kilo rc upstream | 15:30 |
| BjoernT | without that I would guess you can;t spin up instances | 15:30 |
| palendae | Right, but it's in Kilo | 15:30 |
| *** sdake__ has joined #openstack-ansible | 15:30 | |
| BjoernT | yes | 15:30 |
| odyssey4me | so for the project right now, I'd say medium - it can be reprioritised later | 15:30 |
| palendae | So it's not affecting anyone right now | 15:30 |
| d34dh0r53 | +1 | 15:30 |
| odyssey4me | https://bugs.launchpad.net/openstack-ansible/+bug/1428469 | 15:31 |
| openstack | Launchpad bug 1428469 in openstack-ansible "Neutron rootwarp(s) need to be updated for Kilo" [Undecided,New] | 15:31 |
| palendae | (hopefullly) | 15:31 |
| odyssey4me | again, same - medium | 15:31 |
| odyssey4me | happy? | 15:31 |
| BjoernT | yes | 15:31 |
| d34dh0r53 | yep | 15:32 |
| odyssey4me | right: https://bugs.launchpad.net/openstack-ansible/+bug/1428466 | 15:32 |
| openstack | Launchpad bug 1428466 in openstack-ansible "Upgrade from 9.0.6 to 10.1.2 - nova-manage.log permissions" [Undecided,New] | 15:32 |
| odyssey4me | interesting - has anyone else seen this? | 15:32 |
| palendae | I have not | 15:33 |
| palendae | But I haven't done many upgrades, either | 15:33 |
| odyssey4me | can we have a volunteer to triage this? | 15:33 |
| d34dh0r53 | I'll take a look at it | 15:33 |
| odyssey4me | any suggestions for importance? | 15:33 |
| d34dh0r53 | med for now | 15:33 |
| *** sdake_ has quit IRC | 15:34 | |
| bgmccollum | ive seen it twice now | 15:34 |
| odyssey4me | thanks d34dh0r53 | 15:34 |
| bgmccollum | out of two upgrades | 15:34 |
| palendae | 100% for bgmccollum >.< | 15:34 |
| BjoernT | I have not seen this but if we alter playbooks in that way we expect permission than we should be checking those prior | 15:34 |
| odyssey4me | ok, marking as confirmed for now until d34dh0r53 can triage | 15:34 |
| odyssey4me | next: https://bugs.launchpad.net/openstack-ansible/+bug/1428445 | 15:34 |
| openstack | Launchpad bug 1428445 in openstack-ansible "Neutron needs plugin references removed" [Undecided,New] | 15:34 |
| odyssey4me | ah, kilo - moving on | 15:35 |
| b3rnard0 | +1 | 15:35 |
| odyssey4me | right, next non-kilo one: https://bugs.launchpad.net/openstack-ansible/+bug/1428106 | 15:35 |
| openstack | Launchpad bug 1428106 in openstack-ansible "Old version of requests being installed post commit 8e6dbd01c9924ebecffdc04f74e4f6d042610ed8" [Undecided,New] | 15:35 |
| *** jwagner_away is now known as jwagner | 15:37 | |
| odyssey4me | mattt any comments on priority? | 15:38 |
| d34dh0r53 | can this even be fixed pre-kilo? | 15:39 |
| odyssey4me | it looks pretty much like a decision was reached - how should we set the status? triaged? | 15:39 |
| mattt | d34dh0r53: not according to cloudnull | 15:39 |
| d34dh0r53 | yeah, I would say triaged | 15:39 |
| odyssey4me | medium importance? | 15:39 |
| mattt | odyssey4me: that sounds ok to me | 15:39 |
| d34dh0r53 | yeah, me too | 15:39 |
| mattt | because juno/icehouse are unaffected | 15:39 |
| odyssey4me | mattt shall I allocate to you? | 15:39 |
| mattt | odyssey4me: yeah sure | 15:39 |
| odyssey4me | https://bugs.launchpad.net/openstack-ansible/+bug/1428435 | 15:40 |
| openstack | Launchpad bug 1428435 in openstack-ansible "Set the default ansible logging to log to a file" [Undecided,New] | 15:40 |
| odyssey4me | isn't this already done? | 15:40 |
| BjoernT | yeah wishlist. Don't even know if this is possible | 15:40 |
| BjoernT | there was actually a 2nd gub from myself | 15:41 |
| odyssey4me | ah - this is done for the AIO only right now | 15:41 |
| odyssey4me | currently the AIO has it implemented for the gate checks: https://github.com/stackforge/os-ansible-deployment/blob/master/scripts/bootstrap-aio.sh#L45-L49 | 15:42 |
| BjoernT | I would be good to have a standard log over all ansible playboook executions | 15:42 |
| odyssey4me | importance? | 15:42 |
| BjoernT | low ? | 15:43 |
| odyssey4me | yeah, I would agree either low or wishlist | 15:43 |
| d34dh0r53 | let's go low | 15:44 |
| odyssey4me | next: https://bugs.launchpad.net/openstack-ansible/+bug/1427846 | 15:45 |
| openstack | Launchpad bug 1427846 in openstack-ansible "Upgrade 9.0.6 to 10.1.2 - Horizon Static Content Permission Denied" [Undecided,New] | 15:45 |
| palendae | Sounds similar to the other 9.06 -> 10.1.2 bug | 15:45 |
| BjoernT | same issue as the nova manage bug | 15:45 |
| BjoernT | yes | 15:45 |
| d34dh0r53 | yeah, you can give this one to me as well | 15:45 |
| odyssey4me | priority? | 15:46 |
| d34dh0r53 | triaged, medium | 15:46 |
| odyssey4me | done, next: https://bugs.launchpad.net/openstack-ansible/+bug/1427893 | 15:46 |
| openstack | Launchpad bug 1427893 in openstack-ansible trunk "Please increate the network neighbour L2 tables " [Undecided,New] | 15:46 |
| *** nosleep77 has joined #openstack-ansible | 15:47 | |
| odyssey4me | I'm guessing that should be 'increase' ? | 15:47 |
| odyssey4me | looks sensible to me for larger deployments | 15:47 |
| Apsu | odyssey4me: Looking. | 15:47 |
| odyssey4me | BjoernT thoughts on priority? | 15:47 |
| BjoernT | low | 15:49 |
| BjoernT | sorry, my display stalled | 15:49 |
| d34dh0r53 | BjoernT: was this in response to a specific issue with a deployment? | 15:49 |
| BjoernT | or wishlist | 15:50 |
| Apsu | d34dh0r53: That's my question | 15:50 |
| BjoernT | no not yet | 15:50 |
| odyssey4me | wishlist then? | 15:50 |
| Apsu | I'm reading Mirantis' blog post on the issue that's linked in the bug and I see several things wrong. | 15:50 |
| BjoernT | yes | 15:50 |
| Apsu | odyssey4me: No, I'm not even sure it's reasonable yet or applies to us. | 15:50 |
| odyssey4me | ok, Apsu - can you dig into it a bit more | 15:50 |
| Apsu | I'm also not sure it's the solution to their problem. | 15:50 |
| Apsu | Yep. | 15:50 |
| odyssey4me | I'll set it as a wishlist item, but leave it on new | 15:50 |
| Apsu | Alrighty. | 15:51 |
| BjoernT | setting higher net.ipv4.neigh.default.gc_thresh1 reduces arp traffic imho | 15:51 |
| BjoernT | apart from dhcp | 15:51 |
| odyssey4me | next: https://bugs.launchpad.net/openstack-ansible/+bug/1427011 | 15:51 |
| openstack | Launchpad bug 1427011 in openstack-ansible "Upgrade from 9.0.6 to 10.1.2 - rabbitmq issue" [Undecided,New] | 15:51 |
| odyssey4me | this problem has a suitable workaround, and the code is actually in icehouse | 15:51 |
| Apsu | BjoernT: Yes it does, but you would need to first identify ARP traffic as an issue of significance. We can talk about it more after meeting. | 15:52 |
| BjoernT | I had only one customer dealing with this issue. | 15:52 |
| BjoernT | apsu: sure | 15:52 |
| BjoernT | I did downgrade manually rabbitmq to 3.4.3 on this customer and pinned it so we won't update and then downgrade again in a newer release | 15:52 |
| odyssey4me | this is essentially something that needs fixing in docs | 15:53 |
| BjoernT | yes | 15:53 |
| odyssey4me | it should be a known issue for anything < 9.0.7 | 15:53 |
| BjoernT | and 10.1.1 | 15:53 |
| BjoernT | everything prior to pinning | 15:53 |
| odyssey4me | high importance? | 15:54 |
| odyssey4me | yup - I've added a comment to that effect | 15:54 |
| BjoernT | I'm just falling into this issue with a new deployment. funny | 15:54 |
| BjoernT | We should add this soon | 15:54 |
| *** sdake__ has quit IRC | 15:54 | |
| BjoernT | to the docs | 15:54 |
| *** sdake_ has joined #openstack-ansible | 15:55 | |
| odyssey4me | well, os-ansible-deployment has no docs :p | 15:55 |
| BjoernT | yepp I can move it | 15:56 |
| odyssey4me | ta | 15:57 |
| odyssey4me | https://bugs.launchpad.net/openstack-ansible/+bug/1426260 | 15:57 |
| openstack | Launchpad bug 1426260 in openstack-ansible "tempest.scenario.test_minimum_basic.TestMinimumBasicScenario setups fails" [Undecided,New] | 15:57 |
| mattt | don't we run that in the aio gate ? | 15:58 |
| odyssey4me | sigmavirus24 cloudnull git-harry any comment? | 15:59 |
| d34dh0r53 | I just commented on that bug | 15:59 |
| odyssey4me | mattt yeah, this looks like a gate aio tempest fail | 15:59 |
| odyssey4me | transient though - definitely not consistant | 15:59 |
| *** markvoelker has quit IRC | 15:59 | |
| *** claco has joined #openstack-ansible | 15:59 | |
| d34dh0r53 | I think it's a parallel issue | 15:59 |
| mattt | k | 15:59 |
| odyssey4me | sigmavirus24 may be right that it relates to the edge case where the port is somehow used | 16:00 |
| *** markvoelker has joined #openstack-ansible | 16:00 | |
| sigmavirus24 | I forget what made me think that though | 16:00 |
| git-harry | I didn't look at it in any detail when I logged it but I noticed it had happened several times. I'd say it was a low priority | 16:00 |
| sigmavirus24 | Perhaps it was a moment of clarity in the haze of cloud | 16:00 |
| cloudnull | That's the issue I was trying to resolve by setting aio tempest gating to serial. | 16:00 |
| sigmavirus24 | I haven't seen it in a whiel though | 16:00 |
| odyssey4me | we have a low priority vote - any takers? | 16:00 |
| d34dh0r53 | cloudnull: did that help? | 16:01 |
| sigmavirus24 | odyssey4me: seconded | 16:01 |
| cloudnull | It seemed to. | 16:01 |
| *** vmtrooper has joined #openstack-ansible | 16:01 | |
| odyssey4me | any volunteers to try and recreate it after the serial gating tempest patch has landed? | 16:01 |
| cloudnull | I noticed 503 when all ram was consumed and instance was swapping heavily. | 16:01 |
| *** stevemar2 is now known as stevemar | 16:01 | |
| odyssey4me | need more info on the bug perhaps? | 16:02 |
| d34dh0r53 | odyssey4me: we should link the serial patch to this bug | 16:02 |
| palendae | I've seen that happen, too | 16:02 |
| palendae | The 503s | 16:02 |
| palendae | and the swapping | 16:02 |
| odyssey4me | seen in the gate, or in repeated run testing - it is different | 16:02 |
| palendae | repeated run testing | 16:02 |
| odyssey4me | this was specifically in the gate | 16:02 |
| palendae | Ah | 16:02 |
| palendae | Then I have no | 16:02 |
| palendae | not* | 16:02 |
| odyssey4me | repeated run testing is different because tempest doesn't clean up | 16:03 |
| * palendae shakes fist | 16:03 | |
| odyssey4me | so eventually, especially on an AIO, the system runs out of resources and dies | 16:03 |
| odyssey4me | alright - leaving it as confirmed... someone can volunteer to pick it up at some point | 16:03 |
| cloudnull | The issue is resolved in master we should bp to branches for gating. If its not already. | 16:04 |
| Apsu | odyssey4me: What resources? And what doesn't it clean up? I see it clean networks and instances, for sure. Which has been an issue in something palendae and I are working on, because it cleans everything we'd like to troubleshoot on error | 16:04 |
| cloudnull | As for aio dying because tempest idk think we care from a deployment standpoint. | 16:04 |
| * cloudnull going back to not paying attention. | 16:05 | |
| *** markvoelker has quit IRC | 16:05 | |
| odyssey4me | next: https://bugs.launchpad.net/openstack-ansible/+bug/1426254 | 16:05 |
| openstack | Launchpad bug 1426254 in openstack-ansible "Race creating /openstack/log" [Undecided,New] | 16:05 |
| BjoernT | that's an old issue, I only came across onece | 16:05 |
| palendae | Apsu: I think it's the testr repo, but not sure why that'd chew up RAM | 16:05 |
| BjoernT | once | 16:05 |
| *** erikmwilson has joined #openstack-ansible | 16:05 | |
| *** vmtrooper has quit IRC | 16:06 | |
| odyssey4me | yeah BjoernT I thought this was resolved ages back | 16:06 |
| odyssey4me | git-harry did you see this recently? | 16:06 |
| b3rnard0 | hey odyssey4me, are we going to continue for another hour? | 16:06 |
| BjoernT | it might be. Didn't see it for few month now | 16:06 |
| odyssey4me | whoops - out of time, I see now | 16:06 |
| git-harry | I think there was a related bug | 16:06 |
| odyssey4me | 4-5 bugs left to quickly go through I guess | 16:07 |
| git-harry | I think hughsaunders fixed the other one but that was prior to the megapatch | 16:07 |
| d34dh0r53 | I've seen some other people join, after the remaining bugs we can open it up to see if any of the new folks have any specific bug related issues | 16:08 |
| odyssey4me | related: https://bugs.launchpad.net/openstack-ansible/+bug/1407121 | 16:08 |
| openstack | Launchpad bug 1407121 in openstack-ansible trunk "/openstack directory creation race" [Medium,Fix committed] - Assigned to Hugh Saunders (hughsaunders) | 16:08 |
| hughsaunders | haha | 16:08 |
| hughsaunders | voluntold | 16:08 |
| hughsaunders | oh, thats the old one.. | 16:09 |
| odyssey4me | ok, I'll add a link to the old one - but this one needs some sort of triage | 16:09 |
| odyssey4me | any volunteers? | 16:09 |
| Apsu | hughsaunders: Your analysis is likely still correct, though. https://github.com/stackforge/os-ansible-deployment/blob/master/playbooks/roles/lxc_container_create/tasks/container_create.yml is targeted at containers but the directory creation is delegated to the host | 16:10 |
| Apsu | So presumably the parallel delegates don't serialize just because they're delegated to the same host | 16:10 |
| Apsu | Hence, race | 16:10 |
| Apsu | Perhaps a better way to do it is replace the file: module usage with mkdir -p, which won't fail on existing path. | 16:10 |
| Apsu | I don't like it but it looks like the file module assumes some things about state that are time sensitive. Most likely it checks for directory existence and if it's not there tries to make it. If it gets created inbetween those steps, boom | 16:11 |
| git-harry | odyssey4me: sounds like Apsu wants this one | 16:11 |
| hughsaunders | or fail_when: never | 16:12 |
| Apsu | hughsaunders: There's that too. | 16:12 |
| Apsu | git-harry: hughsaunders made a suggestion last, so it's all his | 16:12 |
| Apsu | :) | 16:12 |
| odyssey4me | yeah, hughsaunders has suggested the more ansible-like way | 16:12 |
| odyssey4me | so - any volunteers to work on this? | 16:12 |
| odyssey4me | comments on importance? | 16:12 |
| hughsaunders | can assign me | 16:12 |
| odyssey4me | next: https://bugs.launchpad.net/openstack-ansible/+bug/1426191 | 16:13 |
| openstack | Launchpad bug 1426191 in openstack-ansible trunk "make keystone ensure endpoints command stateful" [Undecided,New] | 16:13 |
| odyssey4me | looks like a wishlist item to me for now | 16:14 |
| d34dh0r53 | +1 | 16:15 |
| odyssey4me | next: https://bugs.launchpad.net/openstack-ansible/+bug/1425953 | 16:15 |
| openstack | Launchpad bug 1425953 in openstack-ansible "Unknown var external_vip_address referenced in os-horizon" [Undecided,New] | 16:15 |
| *** jmccrory has joined #openstack-ansible | 16:15 | |
| b3rnard0 | i have to head out for my next meeting but bug notes are here: https://etherpad.openstack.org/p/openstack_ansible_bug_triage.2015-03-10-16.00 | 16:16 |
| odyssey4me | is this master only git-harry ? | 16:17 |
| git-harry | odyssey4me: it should be | 16:17 |
| odyssey4me | alright, importance suggestions? | 16:18 |
| BjoernT | https://bugs.launchpad.net/openstack-ansible/+bug/1427011 can be closed | 16:20 |
| openstack | Launchpad bug 1427011 in openstack-ansible "Upgrade from 9.0.6 to 10.1.2 - rabbitmq issue" [High,Triaged] | 16:20 |
| BjoernT | moved to docs-issues | 16:20 |
| Apsu | odyssey4me: Thinking somewhat high, as a broken variable reference will halt a deploy | 16:20 |
| odyssey4me | alright, done - needs confirmation and a volunteer, but that doesn't have to happen now | 16:21 |
| stevelle | x | 16:21 |
| git-harry | :q | 16:21 |
| odyssey4me | stevelle thanks | 16:21 |
| odyssey4me | last one: https://bugs.launchpad.net/openstack-ansible/+bug/1423783 | 16:22 |
| openstack | Launchpad bug 1423783 in openstack-ansible "Horizon is not generating CSS correctly" [Undecided,New] | 16:22 |
| odyssey4me | extensive digging has been done into this, without finding a root cause | 16:22 |
| BjoernT | interestingly the statements I put in always fixes that | 16:23 |
| bgmccollum | Initial deployments will be OK, but over time, as resources are recompressed, their names diverge | 16:23 |
| bgmccollum | specifying the HASHING_METHOD to content ensures consistent resource names | 16:24 |
| odyssey4me | bgmccollum oh? are you able to dig a little more into it? | 16:24 |
| odyssey4me | bgmccollum the hashing method is already set | 16:24 |
| bgmccollum | leave a multi-controller install running for a while..it just falls over | 16:24 |
| bgmccollum | odyssey4me: https://github.com/stackforge/os-ansible-deployment/blob/juno/rpc_deployment/roles/horizon_common/templates/local_settings.py | 16:25 |
| bgmccollum | i dont see hasing method for juno | 16:25 |
| odyssey4me | bgmccollum the hashing method is a default in horizon | 16:26 |
| odyssey4me | https://github.com/openstack/horizon/blob/stable/juno/openstack_dashboard/settings.py#L212 | 16:26 |
| odyssey4me | https://github.com/openstack/horizon/blob/stable/icehouse/openstack_dashboard/settings.py#L145 | 16:26 |
| odyssey4me | please see the previous investigation: https://bugs.launchpad.net/openstack-ansible/+bug/1402890 | 16:26 |
| openstack | Launchpad bug 1423783 in openstack-ansible "duplicate for #1402890 Horizon is not generating CSS correctly" [Undecided,New] | 16:27 |
| odyssey4me | any my last comment here: https://bugs.launchpad.net/openstack-ansible/+bug/1423783 | 16:27 |
| openstack | Launchpad bug 1423783 in openstack-ansible "Horizon is not generating CSS correctly" [Undecided,New] | 16:27 |
| odyssey4me | right now, unless we have a dev to volunteer to try and validate this we're going to have to mark it as needing more info | 16:27 |
| odyssey4me | every investigation thus far has led us back to the same point - everything is set as it should be and it should just work | 16:28 |
| Apsu | I can try to validate it. | 16:28 |
| odyssey4me | and we have nothing else to go on without more detailed triage | 16:28 |
| Apsu | Already looking at the code | 16:28 |
| *** daneyon_ has joined #openstack-ansible | 16:29 | |
| odyssey4me | Apsu ok, I'll allocate to you - just look carefully through all the work that's already been done on it please. | 16:29 |
| Apsu | odyssey4me: Yep, read both bugs fully | 16:29 |
| odyssey4me | I'll mark is as low for now, so we don't go around this bush again - it can be re-classified at a later date. | 16:30 |
| odyssey4me | alright, that's all bugs classified | 16:30 |
| mattt | Apsu: i can look at that in a future sprint, but i'm tied up w/ work in this sprint | 16:31 |
| odyssey4me | the floor is open to any questions/comments | 16:31 |
| Apsu | mattt: np | 16:31 |
| odyssey4me | sorry about rushing through all those - there were rather a lot to go through :) | 16:31 |
| *** daneyon has quit IRC | 16:31 | |
| odyssey4me | alright, if there's nothing more then I think we can say we're done | 16:33 |
| Apsu | odyssey4me: Say it. Say it! | 16:34 |
| odyssey4me | lol | 16:35 |
| odyssey4me | we're done? | 16:35 |
| Apsu | There it is | 16:35 |
| Apsu | More conviction next time. Less question marks ;P | 16:35 |
| odyssey4me | #jobdone #meeting_over #hastagssuck | 16:35 |
| Apsu | lulz | 16:36 |
| d34dh0r53 | lol, thanks everyone | 16:36 |
| odyssey4me | thanks for keeping the notes after b3rnard0 left d34dh0r53 | 16:37 |
| odyssey4me | now, where was I | 16:38 |
| d34dh0r53 | sure thing, final notes are here: https://etherpad.openstack.org/p/openstack_ansible_bug_triage.2015-03-10-16.00 | 16:38 |
| odyssey4me | I see you're still tweaking them :p | 16:38 |
| d34dh0r53 | :) done now | 16:39 |
| odyssey4me | mattt the nightly external CI heat multinode tests - does it just run the smoke tests? | 16:39 |
| odyssey4me | or more than that | 16:39 |
| odyssey4me | and do the smoke tests include neutron tests? | 16:40 |
| *** jaypipes has joined #openstack-ansible | 16:40 | |
| mattt | odyssey4me: yeah smoke tests do create networks, subnets, routers, etc. | 16:43 |
| mattt | odyssey4me: the heat multi node stuff does a smoke test, the original multi node job ran a small subset of api tests only | 16:43 |
| odyssey4me | awesome, thanks | 16:43 |
| mattt | np | 16:47 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Dont look for used_ips on containerless hosts https://review.openstack.org/163083 | 16:47 |
| *** jaypipes has quit IRC | 17:07 | |
| *** markvoelker has joined #openstack-ansible | 17:09 | |
| *** markvoelker has quit IRC | 17:19 | |
| *** markvoelker has joined #openstack-ansible | 17:19 | |
| *** sdake_ has quit IRC | 17:21 | |
| *** markvoelker_ has joined #openstack-ansible | 17:21 | |
| *** markvoelker has quit IRC | 17:24 | |
| hughsaunders | cloudnull: if I add similarly formatted vars for oslo.context to https://github.com/stackforge/os-ansible-deployment/blob/master/playbooks/vars/repo_packages/openstack_other.yml will the package get built? | 17:26 |
| *** stevemar has quit IRC | 17:32 | |
| palendae | odyssey4me: Are you around? | 17:42 |
| *** claco has left #openstack-ansible | 17:43 | |
| andymccr | palendae: he headed home | 17:44 |
| palendae | andymccr: Thanks | 17:46 |
| *** vmtrooper has joined #openstack-ansible | 17:50 | |
| *** vmtrooper has quit IRC | 17:54 | |
| alextricity | Ansible question: When calling ansible-playbooks, how does ansible load the dynamic inventory? | 18:01 |
| alextricity | In other words, when and how is dynamic_inventory.py invoked? | 18:02 |
| *** jaypipes has joined #openstack-ansible | 18:04 | |
| palendae | alextricity: Ansible will look for executables in the inventory/ dir and run them, iirc | 18:06 |
| palendae | Also I think you can specify in the ansible.cfg file | 18:07 |
| alextricity | palendae. Yes it's in there. I wasn't sure why it was configured to the inventory/ directory and not the dynamic_inventory.py script. I didn't know that ansible looks for executables in said directory | 18:08 |
| alextricity | Thank you | 18:08 |
| palendae | alextricity: Yeah, it's not well documented in ansible itself, sadly | 18:08 |
| palendae | I think if there's multiple in there, it'll do it in directory listing order (so whatever ls would produce) | 18:09 |
| alextricity | palendae: haha. Yeah after spending like 30 min trying to find that info in the docs, I just decided to ask here | 18:09 |
| Apsu | alextricity: http://docs.ansible.com/intro_dynamic_inventory.html#using-multiple-inventory-sources https://github.com/ansible/ansible/blob/devel/lib/ansible/inventory/dir.py#L53 | 18:19 |
| alextricity | Apsu, thank you | 18:20 |
| Apsu | Essentially it loops over the directory entries, and recursively through subdirectories that aren't host_vars/group_vars/vars_plugins | 18:20 |
| Apsu | If it's an executable, it runs it as a dynamic inventory script | 18:20 |
| Apsu | If it's not a directory and not an executable, it parses it as a static inventory file | 18:20 |
| Apsu | It merges the results, same as using multiple static inventory files. | 18:21 |
| bgmccollum | Apsu, odyssey4me: I have an environment with diverged static resources | 18:24 |
| Apsu | bgmccollum: Oo | 18:24 |
| bgmccollum | http://paste.openstack.org/show/191369/ | 18:24 |
| Apsu | bgmccollum: Now, those resources get created on-demand, from what I understand so far | 18:27 |
| Apsu | The 6c6 files are the same. 8ed appears unique on infra3, and obviously there's lots of uniques on infra1 | 18:27 |
| Apsu | 8ed doesn't appear to have the same size as any of the others | 18:27 |
| Apsu | 18c and 8f4 on infra1 are very close in size. Possible they're duplicates | 18:28 |
| Apsu | Can you md5sum those two? | 18:28 |
| *** sdake_ has joined #openstack-ansible | 18:28 | |
| bgmccollum | yeah | 18:28 |
| bgmccollum | http://paste.openstack.org/show/191370/ | 18:29 |
| Apsu | Oh, hah | 18:29 |
| Apsu | The names Are the md5sums. The prefixes at least. | 18:29 |
| Apsu | So that tells us one useful thing. | 18:30 |
| Apsu | With the hash/content method (which appear identical in the django compressor source), we get md5 = filename | 18:30 |
| Apsu | Which means that if something is missing, it's not because of name divergence. | 18:30 |
| Apsu | It must be due to a failure to generate or process the resource. | 18:30 |
| Apsu | bgmccollum: Can you identify where the unique files on infra1 are referenced in the html, so you can curl/browse those pages on infra2/3 and see if they get generated? | 18:32 |
| Apsu | bgmccollum: Or did you already encounter an error | 18:32 |
| bgmccollum | Yeah, I'm seeing 404 is browser console | 18:33 |
| bgmccollum | admin/hypervisors is trying to load 18c03018c8f6.css | 18:33 |
| bgmccollum | same with project/volumes - completely unstyled | 18:34 |
| Apsu | Interesting. | 18:35 |
| bgmccollum | do you want access to this env? | 18:35 |
| Apsu | I'd take it, sure. My next step would to be to find out exactly what creates the files and when/how it gets run | 18:36 |
| bgmccollum | Apsu: emailed | 18:38 |
| cloudnull | Hughsaunders yes. If we add it to the vars it'll get built. | 18:39 |
| *** stevemar has joined #openstack-ansible | 18:41 | |
| cloudnull | In master. For the other branches we'll need to go build it. | 18:44 |
| *** markvoelker_ has quit IRC | 19:01 | |
| *** prometheanfire is now known as blueness-evil-tw | 19:01 | |
| *** blueness-evil-tw is now known as prometheanfire | 19:02 | |
| *** jmccrory has quit IRC | 19:03 | |
| *** jaypipes is now known as jay-afk | 19:09 | |
| *** jay-afk is now known as jaypipes-afk | 19:10 | |
| palendae | andymccr, hughsaunders: If either of you are here - is tempest broken across the board now due to the oslo.context issue? Seeing it even if I pin to master now | 19:10 |
| hughsaunders | palendae: yep | 19:11 |
| palendae | Son of a | 19:11 |
| palendae | Ok | 19:11 |
| hughsaunders | needs oslo.context 0.2.0 | 19:12 |
| palendae | Even that older SHA does | 19:12 |
| hughsaunders | yeah :( | 19:13 |
| hughsaunders | the dep is through tempest-lib | 19:13 |
| palendae | That certainly puts a damper on getting the neutron tests sorted | 19:13 |
| hughsaunders | which is not capped in the old sha | 19:13 |
| d34dh0r53 | can I get looks at: https://review.openstack.org/#/c/162967/ | 19:13 |
| palendae | And it doesn't look like we pin oslo.contect anywhere | 19:13 |
| d34dh0r53 | and https://review.openstack.org/#/c/163048/ | 19:13 |
| hughsaunders | i ran out of time looking at how to get 0.2.0 built | 19:13 |
| palendae | hughsaunders: Is there a bug for this yet? | 19:18 |
| hughsaunders | yeah should be linked to my reviews | 19:19 |
| *** jmccrory has joined #openstack-ansible | 19:27 | |
| *** vmtrooper has joined #openstack-ansible | 19:39 | |
| *** vmtrooper has quit IRC | 19:44 | |
| Apsu | bgmccollum: Got an example of an unstyled area for easy testing, btw? | 19:48 |
| *** daneyon has joined #openstack-ansible | 19:50 | |
| *** daneyon_ has quit IRC | 19:53 | |
| *** britthouser has joined #openstack-ansible | 20:15 | |
| *** ccrouch has joined #openstack-ansible | 20:27 | |
| *** britthouser has quit IRC | 20:40 | |
| *** jwagner is now known as jwagner_away | 20:42 | |
| *** jwagner_away is now known as jwagner | 20:44 | |
| openstackgerrit | Evan Callicoat proposed stackforge/os-ansible-deployment: Enable offline compression for resource files https://review.openstack.org/163183 | 21:08 |
| BjoernT | apsu: I just have a new 10.1.2 with failed css , who wants to debug it before I add my config again? | 21:11 |
| Apsu | BjoernT: I don't, the issue is definitely missing OFFLINE_COMPRESS :) | 21:12 |
| Apsu | The default is False, and I can see how it's broken without it enabled. | 21:12 |
| Apsu | BjoernT: That's what ^ that commit is for. Just amending it to target it at the right bug since we have dupes | 21:12 |
| openstackgerrit | Evan Callicoat proposed stackforge/os-ansible-deployment: Enable offline compression for resource files https://review.openstack.org/163183 | 21:14 |
| BjoernT | nice, I try that | 21:14 |
| BjoernT | lol I get the permission issue now: | 21:15 |
| BjoernT | TASK: [horizon_common | Collect and compress static files] ******************** | 21:15 |
| BjoernT | failed: [667164-infra01_horizon_container-8e9851a1] => (item=horizon-manage.py collectstatic --noinput) => {"changed": true, "cmd": ["horizon-manage.py", "collectstatic", "--noinput"], "delta": "0:00:00.338413", "end": "2015-03-10 21:15:06.758614", "item": "horizon-manage.py collectstatic --noinput", "rc": 1, "start": "2015-03-10 21:15:06.420201"} | 21:15 |
| BjoernT | stderr: Traceback (most recent call last): | 21:15 |
| BjoernT | that has not yet been reported? | 21:16 |
| Apsu | No idea, I just got voluntold to help figure this one out today :) | 21:16 |
| BjoernT | ok, it is reported https://bugs.launchpad.net/openstack-ansible/+bug/1427846 | 21:18 |
| openstack | Launchpad bug 1427846 in openstack-ansible "Upgrade 9.0.6 to 10.1.2 - Horizon Static Content Permission Denied" [Medium,Triaged] - Assigned to David Wilde (david-wilde-rackspace) | 21:18 |
| openstackgerrit | Nolan Brubaker proposed stackforge/os-ansible-deployment: Fix tempest dependency versions https://review.openstack.org/163194 | 21:25 |
| *** stevemar has quit IRC | 21:27 | |
| *** stevemar has joined #openstack-ansible | 21:27 | |
| *** vmtrooper has joined #openstack-ansible | 21:28 | |
| sigmavirus24 | BjoernT: I suspect that role isn't running as the right user | 21:29 |
| sigmavirus24 | I know mattt had sent a PR updating a bunch of roles to run as specific users but I don't remember if that was pre-stackforge or post | 21:30 |
| BjoernT | I think the role was changed to run as www-data ? Saw a old commit somewhere. The old role ran as root | 21:30 |
| BjoernT | that's why it bails out | 21:30 |
| BjoernT | we should have a removal of the old files prior to the compress | 21:30 |
| *** vmtrooper has quit IRC | 21:33 | |
| mattt | sigmavirus24: that was in stackforge i believe | 21:36 |
| mattt | sadly we never tested upgrades thoroughly :( | 21:36 |
| *** sdake_ has quit IRC | 21:37 | |
| *** jwagner is now known as jwagner_away | 21:39 | |
| *** sdake_ has joined #openstack-ansible | 21:40 | |
| *** Mudpuppy_ has joined #openstack-ansible | 21:42 | |
| *** Mudpuppy has quit IRC | 21:45 | |
| *** Mudpuppy_ has quit IRC | 21:47 | |
| *** jaypipes-afk is now known as JAYPIPES | 21:59 | |
| *** JAYPIPES is now known as jaypipes | 21:59 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:00 | |
| *** sdake__ has joined #openstack-ansible | 22:33 | |
| *** sdake_ has quit IRC | 22:37 | |
| *** nosleep77 has quit IRC | 22:50 | |
| daneyon | I am able to deploy v9 w/o any errors. I am at the step of verifying my deployment and I am unable to find openrc in the utility container. I try to run find but i get permission denied. | 23:05 |
| daneyon | nm.. i found openrc in the keystone container | 23:10 |
| *** sdake__ has quit IRC | 23:11 | |
| *** vmtrooper has joined #openstack-ansible | 23:16 | |
| *** vmtrooper has quit IRC | 23:21 | |
| *** sdake_ has joined #openstack-ansible | 23:39 | |
| *** sdake_ has quit IRC | 23:44 | |
| *** britthouser has joined #openstack-ansible | 23:45 | |
| *** jmccrory has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!