| *** sawblade6 has joined #openstack-ansible | 00:02 | |
| *** pmannidi has joined #openstack-ansible | 00:03 | |
| *** sawblade_ has quit IRC | 00:05 | |
| *** odyssey4me has quit IRC | 00:08 | |
| *** odyssey4me has joined #openstack-ansible | 00:08 | |
| *** kberger has joined #openstack-ansible | 00:08 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Update all references from the deprecated tenant_id to project_id. https://review.openstack.org/552178 | 00:09 |
|---|---|---|
| *** jbadiapa_ has joined #openstack-ansible | 00:10 | |
| *** KeithMnemonic has quit IRC | 00:12 | |
| *** jbadiapa has quit IRC | 00:12 | |
| *** kberger has quit IRC | 00:14 | |
| *** cjloader has joined #openstack-ansible | 00:14 | |
| *** cjloader has quit IRC | 00:14 | |
| *** cjloader has joined #openstack-ansible | 00:15 | |
| *** esberglu has quit IRC | 00:15 | |
| *** esberglu has joined #openstack-ansible | 00:16 | |
| *** cjloader has quit IRC | 00:19 | |
| *** sawblade_ has joined #openstack-ansible | 00:20 | |
| *** esberglu has quit IRC | 00:20 | |
| *** osnaya has joined #openstack-ansible | 00:20 | |
| *** sawblade6 has quit IRC | 00:22 | |
| *** lbragstad has quit IRC | 00:33 | |
| *** sawblade6 has joined #openstack-ansible | 00:35 | |
| *** sawblade_ has quit IRC | 00:37 | |
| RossSchulman[m] | I think a longer format question might help me describe my setup and pinpoint the issues I don't understand. Is there a forum for OSA that might be a more appropriate place for something like that? | 00:49 |
| *** cjloader has joined #openstack-ansible | 00:50 | |
| *** DanyC has joined #openstack-ansible | 01:00 | |
| *** KeithMnemonic has joined #openstack-ansible | 01:01 | |
| *** sawblade_ has joined #openstack-ansible | 01:04 | |
| *** KeithMnemonic has quit IRC | 01:06 | |
| *** sawblade6 has quit IRC | 01:06 | |
| *** DanyC has quit IRC | 01:07 | |
| *** sawblade6 has joined #openstack-ansible | 01:09 | |
| *** armaan has quit IRC | 01:09 | |
| *** sawblade_ has quit IRC | 01:10 | |
| *** sawblade_ has joined #openstack-ansible | 01:12 | |
| *** sawblade6 has quit IRC | 01:13 | |
| *** sawblade6 has joined #openstack-ansible | 01:16 | |
| *** sawblade_ has quit IRC | 01:17 | |
| logan- | RossSchulman[m]: openstack-dev of openstack-operators ML can be useful for that | 01:22 |
| logan- | or* | 01:25 |
| *** osnaya has quit IRC | 01:29 | |
| *** cjloader has quit IRC | 01:32 | |
| *** cjloader has joined #openstack-ansible | 01:33 | |
| *** armaan has joined #openstack-ansible | 01:35 | |
| RossSchulman[m] | Thanks, I'll give that a look. | 01:41 |
| *** cjloader has quit IRC | 01:46 | |
| *** flemingo_ has quit IRC | 01:51 | |
| *** lbragstad has joined #openstack-ansible | 02:20 | |
| *** dave-mccowan has joined #openstack-ansible | 02:23 | |
| *** sawblade_ has joined #openstack-ansible | 02:34 | |
| *** sawblade6 has quit IRC | 02:34 | |
| *** sawblade6 has joined #openstack-ansible | 02:40 | |
| *** sawblade_ has quit IRC | 02:42 | |
| *** armaan has quit IRC | 02:43 | |
| *** armaan has joined #openstack-ansible | 02:43 | |
| *** threestrands has quit IRC | 02:47 | |
| *** threestrands has joined #openstack-ansible | 02:47 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican master: Deprecate auth_uri option https://review.openstack.org/558382 | 02:48 |
| *** threestrands has quit IRC | 02:48 | |
| *** threestrands has joined #openstack-ansible | 02:49 | |
| *** sawblade_ has joined #openstack-ansible | 02:54 | |
| *** flemingo has joined #openstack-ansible | 02:56 | |
| *** sawblade6 has quit IRC | 02:56 | |
| *** flemingo has quit IRC | 03:00 | |
| *** dave-mccowan has quit IRC | 03:01 | |
| *** DanyC has joined #openstack-ansible | 03:03 | |
| *** DanyC has quit IRC | 03:08 | |
| *** sawblade6 has joined #openstack-ansible | 03:16 | |
| *** flemingo has joined #openstack-ansible | 03:17 | |
| *** sawblade_ has quit IRC | 03:18 | |
| *** sawblade_ has joined #openstack-ansible | 03:20 | |
| *** flemingo has quit IRC | 03:21 | |
| *** sawblade6 has quit IRC | 03:22 | |
| *** lbragstad has quit IRC | 03:25 | |
| *** flemingo has joined #openstack-ansible | 03:37 | |
| *** flemingo has quit IRC | 03:41 | |
| *** sawblade6 has joined #openstack-ansible | 03:52 | |
| *** flemingo has joined #openstack-ansible | 03:53 | |
| *** sawblade_ has quit IRC | 03:55 | |
| *** flemingo has quit IRC | 03:58 | |
| *** DanyC has joined #openstack-ansible | 04:05 | |
| *** flemingo has joined #openstack-ansible | 04:06 | |
| *** poopcat has quit IRC | 04:06 | |
| *** udesale has joined #openstack-ansible | 04:07 | |
| *** DanyC has quit IRC | 04:09 | |
| *** flemingo has quit IRC | 04:11 | |
| *** aruns has joined #openstack-ansible | 04:13 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_octavia master: Downloads amphora images from artifact storage https://review.openstack.org/555084 | 04:20 |
| *** jaden has joined #openstack-ansible | 04:20 | |
| *** pcaruana has joined #openstack-ansible | 04:27 | |
| *** sawblade_ has joined #openstack-ansible | 04:28 | |
| *** pcaruana has quit IRC | 04:30 | |
| *** sawblade6 has quit IRC | 04:31 | |
| *** sawblade6 has joined #openstack-ansible | 04:33 | |
| *** sawblade_ has quit IRC | 04:35 | |
| *** threestrands has quit IRC | 04:38 | |
| *** jaden has quit IRC | 04:39 | |
| *** sawblade_ has joined #openstack-ansible | 04:40 | |
| *** sawblade6 has quit IRC | 04:42 | |
| *** chhagarw has joined #openstack-ansible | 04:47 | |
| prometheanfire | oh, please | 04:53 |
| *** sawblade6 has joined #openstack-ansible | 04:59 | |
| *** sawblade_ has quit IRC | 05:01 | |
| *** DanyC has joined #openstack-ansible | 05:06 | |
| *** flemingo has joined #openstack-ansible | 05:06 | |
| *** DanyC has quit IRC | 05:10 | |
| *** flemingo has quit IRC | 05:11 | |
| *** flemingo has joined #openstack-ansible | 05:33 | |
| *** flemingo has quit IRC | 05:37 | |
| *** sawblade_ has joined #openstack-ansible | 05:42 | |
| *** dave-mccowan has joined #openstack-ansible | 05:43 | |
| *** sawblade6 has quit IRC | 05:45 | |
| *** pmannidi has quit IRC | 05:49 | |
| *** flemingo has joined #openstack-ansible | 05:49 | |
| *** flemingo has quit IRC | 05:54 | |
| *** DanyC has joined #openstack-ansible | 06:06 | |
| *** dave-mccowan has quit IRC | 06:09 | |
| *** DanyC has quit IRC | 06:10 | |
| *** flemingo has joined #openstack-ansible | 06:11 | |
| *** sar has joined #openstack-ansible | 06:12 | |
| *** flemingo has quit IRC | 06:16 | |
| *** sawblade6 has joined #openstack-ansible | 06:24 | |
| *** sawblade_ has quit IRC | 06:27 | |
| *** Sha0000 has quit IRC | 06:33 | |
| *** flemingo has joined #openstack-ansible | 06:36 | |
| *** shardy has quit IRC | 06:36 | |
| *** shardy has joined #openstack-ansible | 06:36 | |
| *** mpjetta has quit IRC | 06:39 | |
| *** flemingo has quit IRC | 06:40 | |
| *** mpjetta has joined #openstack-ansible | 06:51 | |
| *** flemingo has joined #openstack-ansible | 06:57 | |
| *** pcaruana has joined #openstack-ansible | 07:01 | |
| *** flemingo has quit IRC | 07:02 | |
| *** sc68cal has joined #openstack-ansible | 07:03 | |
| *** sc68cal- has quit IRC | 07:04 | |
| *** DanyC has joined #openstack-ansible | 07:04 | |
| *** sawblade_ has joined #openstack-ansible | 07:06 | |
| *** aruns__ has joined #openstack-ansible | 07:07 | |
| *** aruns has quit IRC | 07:07 | |
| *** sawblade6 has quit IRC | 07:09 | |
| *** DanyC has quit IRC | 07:09 | |
| *** DanyC has joined #openstack-ansible | 07:10 | |
| *** chyka has joined #openstack-ansible | 07:10 | |
| *** ivve has joined #openstack-ansible | 07:11 | |
| *** armaan has quit IRC | 07:12 | |
| *** prometheanfire has quit IRC | 07:12 | |
| *** fghaas has joined #openstack-ansible | 07:12 | |
| *** armaan has joined #openstack-ansible | 07:12 | |
| *** prometheanfire has joined #openstack-ansible | 07:13 | |
| *** chyka has quit IRC | 07:14 | |
| *** mbuil has joined #openstack-ansible | 07:15 | |
| *** flemingo has joined #openstack-ansible | 07:15 | |
| *** sawblade6 has joined #openstack-ansible | 07:15 | |
| *** ivve has quit IRC | 07:18 | |
| *** sawblade_ has quit IRC | 07:18 | |
| *** flemingo has quit IRC | 07:19 | |
| *** epalper has joined #openstack-ansible | 07:26 | |
| *** hamza21 has joined #openstack-ansible | 07:27 | |
| *** haad1 has joined #openstack-ansible | 07:29 | |
| *** flemingo has joined #openstack-ansible | 07:31 | |
| *** DanyC has quit IRC | 07:35 | |
| *** DanyC has joined #openstack-ansible | 07:35 | |
| *** flemingo has quit IRC | 07:36 | |
| *** electrofelix has joined #openstack-ansible | 07:39 | |
| *** DanyC has quit IRC | 07:40 | |
| *** shardy has quit IRC | 07:45 | |
| *** shardy has joined #openstack-ansible | 07:45 | |
| *** gkadam has joined #openstack-ansible | 07:49 | |
| *** sawblade_ has joined #openstack-ansible | 07:52 | |
| *** prometheanfire has quit IRC | 07:52 | |
| *** prometheanfire has joined #openstack-ansible | 07:54 | |
| *** sawblade6 has quit IRC | 07:54 | |
| *** flemingo has joined #openstack-ansible | 07:56 | |
| *** flemingo has quit IRC | 08:00 | |
| *** sawblade6 has joined #openstack-ansible | 08:04 | |
| *** smorant has joined #openstack-ansible | 08:04 | |
| *** DanyC has joined #openstack-ansible | 08:05 | |
| *** tinwood_ is now known as tinwood | 08:06 | |
| *** sawblade_ has quit IRC | 08:06 | |
| odyssey4me | evrardjp I wonder if https://review.openstack.org/#/q/Ibc191f6fc33ca9a1807c01a23f476c4dfe225f33 won't perhaps help stabilise tempest test execution in the gate | 08:08 |
| evrardjp | I think my patch is doing fine | 08:09 |
| evrardjp | https://review.openstack.org/#/c/557896/ | 08:10 |
| evrardjp | I do not say we need to merge this | 08:10 |
| evrardjp | but I've got 100% success in tempest itself | 08:10 |
| evrardjp | let me continue to do a few rechecks | 08:10 |
| evrardjp | then I set those overrides in gating | 08:11 |
| evrardjp | odyssey4me: I've checked at your rst patch | 08:12 |
| evrardjp | for automatic branch things in rdo/suse | 08:12 |
| evrardjp | it's hard because you need to pass the proper things to pigments | 08:12 |
| evrardjp | pygments* | 08:12 |
| evrardjp | I will have a look at what I can do | 08:13 |
| evrardjp | but we could probably say | 08:13 |
| evrardjp | replace <version> in the following code with ``|blabla|`` | 08:13 |
| evrardjp | and we put <version> in the code to pass in pygments | 08:13 |
| evrardjp | and the people should be smart :p | 08:13 |
| *** flemingo has joined #openstack-ansible | 08:14 | |
| evrardjp | the \ doesn't seem to escape properly | 08:15 |
| evrardjp | oh yes it works | 08:15 |
| evrardjp | it's just me | 08:15 |
| evrardjp | let me put that in the review | 08:15 |
| *** haad1 has quit IRC | 08:17 | |
| *** flemingo has quit IRC | 08:19 | |
| evrardjp | odyssey4me: the Pike merging is very annoying. | 08:25 |
| odyssey4me | yeah, it is rather a pain | 08:25 |
| odyssey4me | that's why I think moving to use a stable tempest version may help | 08:25 |
| evrardjp | I propose, when we'll have the new testing structure, that we backport it all the way to O. | 08:26 |
| evrardjp | my problem is timeouts | 08:26 |
| odyssey4me | sure | 08:26 |
| evrardjp | maybe we should increase it for P | 08:26 |
| evrardjp | because we can't backport the Queens change | 08:26 |
| odyssey4me | yes, I think that timeout patch should just go into the tempest role | 08:26 |
| evrardjp | except if we go for a gutsy decision | 08:26 |
| evrardjp | and say, okay we do a minor release for all Pike | 08:27 |
| odyssey4me | as you have it, let's get it in and port it back to q,p & o | 08:27 |
| *** goldenfri has quit IRC | 08:27 | |
| odyssey4me | I'm not sure what you're suggesting | 08:27 |
| *** aruns__ has quit IRC | 08:27 | |
| evrardjp | well the thing is I don't have the problem on my machine, so I have to do a few iterations on gates when some ppl don't break it :p | 08:27 |
| evrardjp | oh for the gutsy decision in Pike? | 08:28 |
| evrardjp | have a 16.2.0 | 08:28 |
| evrardjp | where we backport a major series of things from Q | 08:28 |
| odyssey4me | ok, my suggestion is to merge the timeout increase into the tempest role - the changes should be hard=set in the templates as you have it | 08:28 |
| odyssey4me | we then port those back through all the stable branches | 08:28 |
| evrardjp | yes that's fine, that's an easy fix | 08:28 |
| odyssey4me | then we sha bump in the integrated repo to include it | 08:28 |
| evrardjp | it would help Q/O | 08:29 |
| evrardjp | but P will continue to fail | 08:29 |
| evrardjp | because P is sloooooooooooow | 08:29 |
| odyssey4me | for p, we combine the two sha bumps (pip_install and os_tempest) | 08:29 |
| evrardjp | which one do you mean? | 08:29 |
| evrardjp | for pip install | 08:29 |
| evrardjp | the constraints? | 08:29 |
| odyssey4me | https://review.openstack.org/558229 | 08:29 |
| odyssey4me | yes | 08:30 |
| evrardjp | that's peanuts | 08:30 |
| evrardjp | well | 08:30 |
| evrardjp | that's important peanuts | 08:30 |
| evrardjp | :) | 08:30 |
| odyssey4me | the point is that patch can be combined with the patch to increase the tempest timeouts | 08:30 |
| evrardjp | but for the time it takes to build Pike, it won't change much | 08:30 |
| *** ibmko_ has joined #openstack-ansible | 08:30 | |
| odyssey4me | which will help get it merged | 08:30 |
| evrardjp | yeah | 08:30 |
| evrardjp | but I have to do a few iterations again | 08:30 |
| evrardjp | it improves reliability but increases the chance of timeouts | 08:31 |
| evrardjp | I'd say "Thanks Intel for those spectre/meltdown" | 08:31 |
| evrardjp | the "gutsy" decision was more about backporting radical changes we made in Q | 08:32 |
| evrardjp | I do not think it would be an easy project | 08:34 |
| openstackgerrit | Merged openstack/ansible-role-systemd_service master: Add a documentation list default https://review.openstack.org/558553 | 08:37 |
| *** flemingo has joined #openstack-ansible | 08:37 | |
| *** indistylo has joined #openstack-ansible | 08:38 | |
| *** Sha0000 has joined #openstack-ansible | 08:41 | |
| *** flemingo has quit IRC | 08:41 | |
| odyssey4me | I think doing that presents more risk than it's worth, especially given how long ago it was when it was done. | 08:42 |
| odyssey4me | We can perhaps do it, but you need to release what we have first, so that we have a full two week timeframe to figure out the issues. | 08:43 |
| evrardjp | yeah | 08:43 |
| evrardjp | quick question | 08:43 |
| evrardjp | for this: https://review.openstack.org/#/c/557988/2 | 08:43 |
| evrardjp | is there anything else pending? | 08:44 |
| evrardjp | My memory is telling me there is something else | 08:44 |
| odyssey4me | nope, for the role test it uses the release tempest in u-c | 08:44 |
| odyssey4me | I'm not sure if the integrated build needs changes to do the same. mattt tells me that the role defaults were being used, but I haven't validated it | 08:44 |
| evrardjp | I will merge this in Queens, so I can bring the stestr | 08:45 |
| evrardjp | because right now we are stabbing in the dark for tempest on Q | 08:45 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_tempest stable/queens: Switch from testrepository to stestr https://review.openstack.org/557335 | 08:47 |
| openstackgerrit | Merged openstack/openstack-ansible master: bootstrap-host: Correct SuSE repo name https://review.openstack.org/558593 | 08:47 |
| evrardjp | odyssey4me: could you revote ^ I added the depends on | 08:47 |
| *** mathlin has joined #openstack-ansible | 08:49 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Developer docs refactor https://review.openstack.org/557650 | 08:51 |
| evrardjp | odyssey4me: haha: https://github.com/openstack/openstack-ansible-ceph_client/blob/newton-eol/tasks/ceph_auth_extra.yml#L19 https://github.com/openstack/openstack-ansible-ceph_client/blob/newton-eol/tasks/ceph_auth_extra.yml#L24 | 08:54 |
| evrardjp | never caught by testing until now! | 08:54 |
| evrardjp | just before EOLing, when the role is eoled already | 08:54 |
| evrardjp | wooot | 08:54 |
| evrardjp | it's because now we are properly linting since we merged the change with run_tests.sh I guess | 08:57 |
| evrardjp | ahah found the issue | 08:59 |
| evrardjp | oh no my bad | 09:00 |
| *** ibmko_ has quit IRC | 09:01 | |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: Set new parameter in OVS to support ODL Oxygen https://review.openstack.org/553333 | 09:06 |
| *** vnogin has joined #openstack-ansible | 09:07 | |
| odyssey4me | heh, double with-items... how nice | 09:10 |
| *** aruns has joined #openstack-ansible | 09:11 | |
| odyssey4me | if the lint test is failing for newton, we might just have to turn it off for the last EOL patch | 09:11 |
| odyssey4me | unless we recreate the newton branch, then fix the issue, then replace the eol tag | 09:11 |
| *** vnogin has quit IRC | 09:11 | |
| *** indistylo has quit IRC | 09:11 | |
| evrardjp | odyssey4me: I won't do this dance if not required | 09:14 |
| evrardjp | I am building on my machine, and the with_items is just a warning | 09:14 |
| evrardjp | (still working on it) | 09:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Ensure that RDO/SuSE series repo instructions stay up to date https://review.openstack.org/558558 | 09:14 |
| evrardjp | if it's not impacting the deploy, I won't fix it, just ignore it | 09:14 |
| evrardjp | odyssey4me: it's SUSE not SuSE :p | 09:14 |
| evrardjp | hwoarang: told you in a commit :) | 09:14 |
| *** ibmko_ has joined #openstack-ansible | 09:14 | |
| hwoarang | wait wait i have something to show you | 09:15 |
| odyssey4me | yeah, yeah - sorry - will update now | 09:15 |
| odyssey4me | I'm old school. :p | 09:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558558 | 09:15 |
| hwoarang | odyssey4me: for you https://www.youtube.com/watch?v=N4lFHP-UxTk | 09:16 |
| odyssey4me | haha | 09:17 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/queens: Do not generate the html ARA report any more https://review.openstack.org/558586 | 09:17 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/pike: Do not generate the html ARA report any more https://review.openstack.org/558587 | 09:17 |
| *** sawblade_ has joined #openstack-ansible | 09:17 | |
| evrardjp | hahah | 09:18 |
| *** ibmko_ has quit IRC | 09:19 | |
| *** sawblade6 has quit IRC | 09:20 | |
| *** sawblade6 has joined #openstack-ansible | 09:23 | |
| *** sawblade_ has quit IRC | 09:25 | |
| *** pmannidi has joined #openstack-ansible | 09:25 | |
| *** vnogin has joined #openstack-ansible | 09:28 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/queens: Disable developer mode for tempest role tests https://review.openstack.org/557988 | 09:30 |
| *** ibmko_ has joined #openstack-ansible | 09:34 | |
| fghaas | SuSE SuSE SuSE RedHat RedHat RackSpace RackSpace Openstack CEPH have I annoyed everyone yet? | 09:39 |
| *** boxrick_ has quit IRC | 09:39 | |
| *** boxrick has joined #openstack-ansible | 09:40 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558558 | 09:40 |
| odyssey4me | fghaas it's OpenStack :p | 09:41 |
| fghaas | odyssey4me: whooosh? | 09:41 |
| odyssey4me | evrardjp the newton-eol patch failed lint again | 09:43 |
| evrardjp | I tried on my machine by doing tox -e linters and it worked | 09:46 |
| evrardjp | so it's the way the job is inherited I guess | 09:47 |
| evrardjp | let me check that | 09:47 |
| odyssey4me | ¯\(°_o)/¯ | 09:47 |
| evrardjp | I will compare the outputs and if I don't see something really different, I will just move it to nv | 09:48 |
| evrardjp | no need to spend that much time on this | 09:48 |
| odyssey4me | it won't be able to be -nv because there is no such job, you'll have to remove it from the check queue | 09:48 |
| odyssey4me | chec/gate | 09:48 |
| odyssey4me | unless you want to add a -nv job, but that's just wasting time IMO | 09:49 |
| odyssey4me | evrardjp ok, looks like https://review.openstack.org/558558 is doing all the right things, thanks for the help on that one! | 09:50 |
| odyssey4me | hwoarang if you could peek at it too please, that'd be nice | 09:50 |
| *** flemingo has joined #openstack-ansible | 09:51 | |
| odyssey4me | hwoarang evrardjp my only thing for that patch is how we deal with master, where neither of those configurations will work until we hit the RC | 09:51 |
| odyssey4me | I guess I could make it use previous_series_name if the branch is master, then current series name if it is not? | 09:52 |
| odyssey4me | ja, I think I should do that. | 09:52 |
| mbuil | hwoarang: what do you mean by a release note in your comment? | 09:53 |
| *** flemingo has quit IRC | 09:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558558 | 09:56 |
| hwoarang | mbuil: ehh an openstack release note... | 09:56 |
| hwoarang | mbuil: openstack-ansible-os_neutron/releasenotes/ and https://docs.openstack.org/reno/latest/ | 09:57 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558558 | 09:57 |
| *** Sha0000 has quit IRC | 10:04 | |
| *** foutatoro has joined #openstack-ansible | 10:04 | |
| odyssey4me | hwoarang evrardjp made a suggestion in https://review.openstack.org/#/c/558599/2 | 10:05 |
| *** evin has joined #openstack-ansible | 10:07 | |
| evrardjp | odyssey4me: will have a look | 10:07 |
| *** vnogin_ has joined #openstack-ansible | 10:07 | |
| *** armaan_ has joined #openstack-ansible | 10:07 | |
| *** haad1 has joined #openstack-ansible | 10:09 | |
| *** _ioni has joined #openstack-ansible | 10:10 | |
| *** bradm_ has joined #openstack-ansible | 10:11 | |
| *** flemingo has joined #openstack-ansible | 10:12 | |
| *** pcaruana|afk| has joined #openstack-ansible | 10:12 | |
| *** gokhan__ has joined #openstack-ansible | 10:12 | |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: Set new parameter in OVS to support ODL Oxygen https://review.openstack.org/553333 | 10:13 |
| *** stuartgr1 has joined #openstack-ansible | 10:13 | |
| *** olivierb- has joined #openstack-ansible | 10:14 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558558 | 10:15 |
| *** sar has quit IRC | 10:15 | |
| *** ansmith has quit IRC | 10:15 | |
| *** brad[] has quit IRC | 10:15 | |
| *** geb has quit IRC | 10:15 | |
| *** stuartgr has quit IRC | 10:15 | |
| *** ioni has quit IRC | 10:15 | |
| *** vnogin has quit IRC | 10:15 | |
| *** mbuil has quit IRC | 10:15 | |
| *** armaan has quit IRC | 10:15 | |
| *** pcaruana has quit IRC | 10:15 | |
| *** olivierbourdon38 has quit IRC | 10:15 | |
| *** gokhan_ has quit IRC | 10:15 | |
| *** admin0 has quit IRC | 10:15 | |
| *** dcdamien has quit IRC | 10:15 | |
| *** dmsimard has quit IRC | 10:15 | |
| *** bradm has quit IRC | 10:15 | |
| *** geb_ has joined #openstack-ansible | 10:15 | |
| *** admin0 has joined #openstack-ansible | 10:16 | |
| evrardjp | Giving up on the newton thing. It's only on setup everything, and only in gates. | 10:16 |
| evrardjp | requirements seem the same | 10:16 |
| *** jwitko has quit IRC | 10:16 | |
| *** flemingo has quit IRC | 10:16 | |
| *** ansmith has joined #openstack-ansible | 10:17 | |
| *** dmsimard has joined #openstack-ansible | 10:17 | |
| *** mbuil has joined #openstack-ansible | 10:18 | |
| *** dcdamien has joined #openstack-ansible | 10:19 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/newton: Update all SHAs for newton-eol https://review.openstack.org/558539 | 10:19 |
| odyssey4me | evrardjp you should add why the linters check is remove to the commit msg | 10:19 |
| *** udesale has quit IRC | 10:23 | |
| *** udesale has joined #openstack-ansible | 10:24 | |
| *** sawblade_ has joined #openstack-ansible | 10:26 | |
| *** sawblade6 has quit IRC | 10:28 | |
| *** pcaruana|afk| has quit IRC | 10:28 | |
| *** pcaruana|afk| has joined #openstack-ansible | 10:28 | |
| *** flemingo has joined #openstack-ansible | 10:32 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/queens: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558777 | 10:35 |
| *** flemingo has quit IRC | 10:38 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558778 | 10:39 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/newton: Update all SHAs for newton-eol https://review.openstack.org/558539 | 10:39 |
| *** pmannidi has quit IRC | 10:40 | |
| odyssey4me | evrardjp something we can do for the pike gate is switch it to doing compute kit only | 10:43 |
| *** sawblade6 has joined #openstack-ansible | 10:43 | |
| evrardjp | odyssey4me: we don't have compute kit in P | 10:43 |
| evrardjp | we discussed this :) | 10:43 |
| evrardjp | but yes | 10:44 |
| evrardjp | that would solve the timing issues | 10:44 |
| *** sawblade_ has quit IRC | 10:45 | |
| odyssey4me | that's easy to add | 10:46 |
| evrardjp | is there a clean backport for this? | 10:48 |
| evrardjp | let me check | 10:48 |
| *** sawblade6 has quit IRC | 10:49 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/queens: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558777 | 10:49 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558778 | 10:50 |
| evrardjp | odyssey4me: we should probably have a ubuntu basekit in master too before doing that | 10:50 |
| odyssey4me | evrardjp probably not, but it doesn't matter too much - we can still do it | 10:50 |
| evrardjp | yeah it's test coverage it's not touching code | 10:50 |
| evrardjp | but I've been there | 10:51 |
| evrardjp | adding tests and boops we find out stuff is wrong! | 10:51 |
| evrardjp | (which is great!) | 10:51 |
| *** sawblade6 has joined #openstack-ansible | 10:51 | |
| *** Sha000000 has joined #openstack-ansible | 10:57 | |
| *** brad[] has joined #openstack-ansible | 11:01 | |
| *** sawblade_ has joined #openstack-ansible | 11:07 | |
| *** haad1 has quit IRC | 11:07 | |
| *** sawblade6 has quit IRC | 11:09 | |
| foutatoro | hi all, I'm trying to deploy openstack with ceph as backend storage. The install process finished correctly by it seems that ceph is not running correctly in the storage node. | 11:10 |
| foutatoro | here is my configs http://paste.openstack.org/show/718371/ | 11:10 |
| foutatoro | any suggestion about what's wrong ? | 11:10 |
| *** sawblade6 has joined #openstack-ansible | 11:17 | |
| *** _ioni is now known as ioni | 11:17 | |
| *** ioni has quit IRC | 11:18 | |
| *** ioni has joined #openstack-ansible | 11:18 | |
| *** vnogin_ has quit IRC | 11:20 | |
| *** sawblade_ has quit IRC | 11:21 | |
| *** haad1 has joined #openstack-ansible | 11:22 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/queens: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558777 | 11:25 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558778 | 11:28 |
| *** flemingo has joined #openstack-ansible | 11:31 | |
| fghaas | foutatoro: why would you want to use a single storage host for Ceph? | 11:32 |
| *** flemingo has quit IRC | 11:36 | |
| foutatoro | fghaas: it's just for testing purpose before using a real ceph cluster | 11:37 |
| foutatoro | fghaas: I used LVM in my previous deployment | 11:37 |
| evrardjp | foutatoro: which branch? | 11:39 |
| foutatoro | evrardjp: 15.1.15 | 11:39 |
| evrardjp | ocata | 11:40 |
| evrardjp | mmmm | 11:40 |
| fghaas | your public_network and cluster_network settings can't work, I think | 11:40 |
| evrardjp | seemed to have worked in gates: http://logs.openstack.org/periodic/git.openstack.org/openstack/openstack-ansible/stable/ocata/openstack-ansible-deploy-ceph-ubuntu-xenial/abe3601/ | 11:41 |
| evrardjp | I'd say a config issue, but fghaas just seemed to have confirmed it :) | 11:41 |
| fghaas | the OSA storage network would have to be the Ceph public network, because that's what Ceph clients (like Glance, Cinder and Qemu/KVM) use to connect to the Ceph cluster. The Ceph cluster network, *if* you want to separate it, would have to be yet another network — though for a test environment with just one node, splitting public and cluster networks strikes me as an exercise in futility. | 11:43 |
| fghaas | foutatoro: if you want to read up on Ceph public vs. cluster networks, http://docs.ceph.com/docs/master/rados/configuration/network-config-ref/#ceph-networks is your definitive reference | 11:45 |
| *** vnogin has joined #openstack-ansible | 11:45 | |
| Miouge | It should work to run OSA with “ --user ubuntu --become --become-user root” flags right? Not required to use the root account over SSH? | 11:50 |
| *** sawblade_ has joined #openstack-ansible | 11:52 | |
| foutatoro | fghaas evrardjp: you mean that I do make this change public_network: "{{ cidr_networks.storage }}" in user_variables.yml ? | 11:53 |
| fghaas | yes, and you should drop the cluster_network entry altogether. Note, however, that you may want to completely start over with your deployment, because: | 11:53 |
| *** sawblade6 has quit IRC | 11:54 | |
| fghaas | Unless your playbook actually failed, I am guessing that you now have an entry in your monmap with the wrong mon IP. If you now redeploy your mon with a different public_network, you'll have two mons in the monmap, one of which is always down. That means you never get mon quorum, and your cluster can never be functional. | 11:55 |
| evrardjp | Miouge: we have everything setup for root user | 11:55 |
| odyssey4me | Miouge I don't think we're yet in a place where that works across the board just yet. We're moving towards that, but are not quite there yet. | 11:55 |
| odyssey4me | So yes, the root user is used to access hosts/containers. | 11:56 |
| evrardjp | Miouge: it's not a feature asked a lot but if you're willing to help you can :) | 11:56 |
| odyssey4me | evrardjp hwoarang the implementations for pike/queens are also done: https://review.openstack.org/#/q/Id28540d335390a52e0eb65a734a1d594235265d6 | 11:56 |
| odyssey4me | (of the rdo/suse deploy host docs thing) | 11:57 |
| *** sawblade_ has quit IRC | 11:57 | |
| evrardjp | oh yeah I renamed things | 11:57 |
| *** vnogin has quit IRC | 11:57 | |
| *** sawblade6 has joined #openstack-ansible | 11:58 | |
| Miouge | evrardjp: I guess I would start by testing that in an AIO deploy to see what fails and work my way through the errors? | 11:58 |
| hwoarang | thank you odyssey4me | 11:59 |
| *** flemingo has joined #openstack-ansible | 11:59 | |
| *** Sha000000 has quit IRC | 12:00 | |
| evrardjp | Miouge: yup | 12:00 |
| evrardjp | that's the easiest | 12:00 |
| evrardjp | we can help you better that way | 12:00 |
| evrardjp | try on a recent branch too | 12:01 |
| Miouge | I am working on an ansible playbook to do the “Prepare the target hosts” stage (https://docs.openstack.org/project-deploy-guide/openstack-ansible/draft/targethosts-prepare.html#configure-the-operating-system-ubuntu). I have a working PoC using ansible “raw” module. Something worth contributing? | 12:01 |
| evrardjp | like master | 12:01 |
| evrardjp | Miouge: probably in ops yes | 12:01 |
| evrardjp | Because we don't expect ppl to have ansible already | 12:02 |
| Miouge | Until now I had a hacky “for srv in $hosts; do ssh $srv sh < prepare-hosts.sh; done” | 12:02 |
| odyssey4me | Miouge I think that could be done in the openstack-hosts playbook | 12:02 |
| *** sawblade_ has joined #openstack-ansible | 12:02 | |
| evrardjp | Well it's good enough when you think of it | 12:02 |
| Miouge | Well that’s the point of the “raw” module, you don’t need python in the target node | 12:02 |
| evrardjp | yes I know but you still need ansible | 12:03 |
| Miouge | So you can use the “raw” module to install python, then run ansible as usual :) | 12:03 |
| evrardjp | :p | 12:03 |
| odyssey4me | someone already started it: https://github.com/openstack/openstack-ansible/blob/master/playbooks/openstack-hosts-setup.yml#L22 | 12:03 |
| evrardjp | I am surprised it's not me :) | 12:03 |
| evrardjp | if it's only for target hosts indeed that could make sense | 12:04 |
| *** flemingo has quit IRC | 12:04 | |
| evrardjp | networking config could be done too | 12:04 |
| evrardjp | but I think it would be wise to do that on a separate playbook | 12:04 |
| odyssey4me | yes, absolutely | 12:04 |
| *** sawblade6 has quit IRC | 12:04 | |
| odyssey4me | but given that we now have systemd networking being implemented across all platforms, it's possible to now do that | 12:05 |
| evrardjp | indeed | 12:05 |
| evrardjp | networkd is not 100% multi platform if you want a working networking stack | 12:05 |
| evrardjp | it expects systemd-resolved for example | 12:05 |
| evrardjp | which is not shipped by all distros | 12:06 |
| odyssey4me | ah yes, we're still waiting for there to be a suse package for it | 12:06 |
| evrardjp | ofc we don't care if we have a resolved role | 12:06 |
| evrardjp | or only static things | 12:06 |
| Miouge | good thing I asked then, odyssey4me it looks like it’s already done :) | 12:07 |
| odyssey4me | Miouge only for ubuntu though - it could be extended to work for other platforms too | 12:07 |
| evrardjp | not 100% is done | 12:07 |
| evrardjp | if you check, we don't install any package | 12:07 |
| evrardjp | we don't load modules | 12:08 |
| evrardjp | (which we could with openstack_hosts) | 12:08 |
| evrardjp | we don't configure storage or networking | 12:08 |
| evrardjp | and that's where I think it would be helpful | 12:08 |
| Miouge | On ubuntu the doc says to do “apt-get install bridge-utils debootstrap ifenslave ifenslave-2.6 lsof lvm2 ntp ntpdate openssh-server sudo tcpdump vlan”, that openstack-hosts-setup.yml#L22 openstack-hosts-setup.yml#L22 installs only python. Do we can to close the gap? Less manual action is always a good thing in my book :) | 12:08 |
| Miouge | s/Do we can/Do we want | 12:09 |
| evrardjp | Miouge: yes, see what I just said above :) | 12:09 |
| odyssey4me | hmm, I think ntp shouldn't be in that list given that we install chrony | 12:09 |
| evrardjp | odyssey4me: that's not 100% true | 12:09 |
| evrardjp | chrony is installed if hardening is used | 12:09 |
| evrardjp | iirc | 12:09 |
| odyssey4me | openssh-server needs to be there before you can use ansible | 12:09 |
| odyssey4me | not sure why tcpdump/lsof are there | 12:10 |
| evrardjp | odyssey4me: Miouge my point there: we can bridge some of the gap with the openstack_hosts role, which is supposed to run on every single host (physical or container) of the environement | 12:10 |
| *** jbadiapa_ is now known as jbadiapa | 12:10 | |
| odyssey4me | ifenslave/bridge-utils, etc would need to be put there before ansible runs because you need to implement your bridges beforehand | 12:10 |
| evrardjp | the openstack_hosts role already take care to know if you're on a physical host to not run things for physical hosts in containers | 12:11 |
| evrardjp | for the rest we can classify in two things | 12:11 |
| evrardjp | requirements (like openssh/networking/storage) and optional things | 12:12 |
| evrardjp | optional could be ntp if chrony is not installed | 12:12 |
| evrardjp | or that no ntp server is provided through dhcp | 12:12 |
| evrardjp | but one step at a time, I'd say the docs need change to follow the eventual changes in openstack_hosts packages install | 12:13 |
| Miouge | Now I wonder why that package list is so long. The only requirements should be openssh-server, then from that point ansible can take over and install all needed packages no? | 12:14 |
| odyssey4me | if we automate the bridge setup, yes - if not, no | 12:14 |
| evrardjp | yes | 12:14 |
| Miouge | OK, I’ll see what I can do | 12:15 |
| evrardjp | Miouge: as you can see, all the packages listed there are used in that same page | 12:15 |
| evrardjp | Miouge: you might want to check how to automate the network configuration with our systemd-networkd role | 12:16 |
| *** sawblade6 has joined #openstack-ansible | 12:17 | |
| evrardjp | but that's a stretch goal probably :) | 12:17 |
| evrardjp | (also cloudnull has started to implement that already) | 12:17 |
| Miouge | I deploy physical nodes with Ubuntu maas, so it takes care of the bridges and storage already. A manual deploy would need to fix those things manually hence why you need some of those packages, I see. | 12:17 |
| *** sawblade_ has quit IRC | 12:19 | |
| *** odyssey4me has quit IRC | 12:23 | |
| *** odyssey4me has joined #openstack-ansible | 12:23 | |
| *** sawblade6 has quit IRC | 12:23 | |
| evrardjp | odyssey4me: I am very sad | 12:24 |
| *** flemingo has joined #openstack-ansible | 12:24 | |
| evrardjp | pip is the gift that keeps on giving | 12:24 |
| evrardjp | https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L138-L141 https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L174 | 12:24 |
| evrardjp | so ofc https://github.com/openstack/openstack-ansible/blob/stable/ocata/scripts/bootstrap-ansible.sh#L98-L104 in ocata installs 10.x | 12:25 |
| *** sawblade6 has joined #openstack-ansible | 12:26 | |
| odyssey4me | evrardjp ah yeah, we didn't port all those adjustments back to ocata | 12:28 |
| *** flemingo has quit IRC | 12:28 | |
| *** lbragstad has joined #openstack-ansible | 12:29 | |
| odyssey4me | 'cos https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L164-L165 should have been done early | 12:29 |
| odyssey4me | I guess that https://github.com/openstack/openstack-ansible/commit/5f7000e185fca71a5077ae7e2bcd29450c4cb70f needs to be ported back to ocata ? | 12:29 |
| evrardjp | oh I see | 12:29 |
| evrardjp | odyssey4me: yup we can backport this one, but then we need to backport the other ones too | 12:30 |
| evrardjp | I mean the more recent ones | 12:31 |
| odyssey4me | yep, it'll very likely need to be a combined chain | 12:31 |
| odyssey4me | want me to do that? | 12:31 |
| evrardjp | omg I wonder if I should just not re-implement the global-requirements-pins there | 12:31 |
| evrardjp | if you have time yes, so I can do something else | 12:31 |
| odyssey4me | ok, I can do that | 12:32 |
| *** flemingo has joined #openstack-ansible | 12:34 | |
| *** sawblade_ has joined #openstack-ansible | 12:35 | |
| evrardjp | odyssey4me: if you remove the pip_packages like the old patch, don't forget its gonna break https://github.com/openstack/openstack-ansible-pip_install/blob/stable/ocata/tasks/install_source.yml#L21 | 12:35 |
| evrardjp | so the rest of the patches have to be included :( | 12:35 |
| *** sawblade6 has quit IRC | 12:37 | |
| *** adrianreza has quit IRC | 12:37 | |
| evrardjp | odyssey4me: wait because I see on the gates they have NOT 10. | 12:38 |
| *** flemingo has quit IRC | 12:38 | |
| evrardjp | I wonder how it landed on my machine | 12:39 |
| odyssey4me | evrardjp no it won't | 12:39 |
| odyssey4me | https://github.com/openstack/openstack-ansible-pip_install/blob/stable/ocata/defaults/main.yml#L87 | 12:39 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Set new parameter in OVS to support ODL Oxygen https://review.openstack.org/553333 | 12:39 |
| odyssey4me | what'll happen is that the bootstrap will put 10 there, then pip_install will downgrade it | 12:40 |
| evrardjp | odyssey4me: we don't have any version there | 12:40 |
| *** vnogin has joined #openstack-ansible | 12:41 | |
| evrardjp | and we don't have the change for using the constraints in ocata | 12:41 |
| odyssey4me | evrardjp yes, because constraints provides the version | 12:41 |
| odyssey4me | and yes, that means we need the constraints patch to ocata - naturally | 12:41 |
| evrardjp | https://github.com/openstack/openstack-ansible-pip_install/blob/stable/ocata/tasks/install_source.yml#L19 | 12:42 |
| evrardjp | ok | 12:42 |
| evrardjp | if we do these that's fine :) | 12:42 |
| evrardjp | but that's a lot | 12:42 |
| evrardjp | maybe it's worth just passing the global-requirement-pins in the bootstrap ansible | 12:42 |
| evrardjp | so that it's a minimal change | 12:42 |
| evrardjp | let's pray for EOL of ocata soon | 12:43 |
| evrardjp | but it won't happen | 12:43 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install master: Specify defaults for pip_install_upper_constraints https://review.openstack.org/558599 | 12:50 |
| *** sawblade6 has joined #openstack-ansible | 12:51 | |
| *** sawblade_ has quit IRC | 12:53 | |
| *** sawblade_ has joined #openstack-ansible | 12:56 | |
| *** sawblade6 has quit IRC | 12:57 | |
| *** shardy is now known as shardy_mtg | 12:58 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Skip host pip installs for ansible bootstrap https://review.openstack.org/558807 | 12:59 |
| *** fghaas has quit IRC | 13:01 | |
| *** harryhuang_ has quit IRC | 13:01 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Improve ansible host & venv bootstrap https://review.openstack.org/558809 | 13:02 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Skip host pip installs for ansible bootstrap https://review.openstack.org/558810 | 13:02 |
| *** sawblade_ has quit IRC | 13:02 | |
| odyssey4me | mattt mind porting https://review.openstack.org/#/c/558599/ back to queens/pike ? | 13:04 |
| *** evin has quit IRC | 13:05 | |
| evrardjp | d34dh0r53: what's the status with ironic deprecations? | 13:06 |
| *** smorant has quit IRC | 13:07 | |
| evrardjp | and what about ELK? | 13:10 |
| mattt | odyssey4me: sure, looking now | 13:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install stable/ocata: Use constraints with get-pip.py https://review.openstack.org/558811 | 13:10 |
| *** sawblade6 has joined #openstack-ansible | 13:10 | |
| *** smorant has joined #openstack-ansible | 13:11 | |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-pip_install stable/queens: Specify defaults for pip_install_upper_constraints https://review.openstack.org/558813 | 13:12 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Improve ansible host & venv bootstrap https://review.openstack.org/558809 | 13:12 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-pip_install stable/pike: Specify defaults for pip_install_upper_constraints https://review.openstack.org/558814 | 13:12 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Skip host pip installs for ansible bootstrap https://review.openstack.org/558810 | 13:12 |
| mattt | odyssey4me: the original issue didn't hit ocata ? | 13:12 |
| odyssey4me | mattt nope, but given that we're looking to push similar changes which caused the issue into ocata, I've combined both patches into https://review.openstack.org/558811 | 13:13 |
| odyssey4me | evrardjp there we go, I found another relevant patch as well to port back to pike | 13:14 |
| odyssey4me | see https://review.openstack.org/#/q/topic:get-pip-constraints+status:open | 13:14 |
| mattt | odyssey4me: ok cool | 13:14 |
| odyssey4me | mattt evrardjp so as far as I recall, the two major reasons for pike integrated build failures are tempest timeouts, and build timeouts - is that correct? | 13:19 |
| *** udesale has quit IRC | 13:20 | |
| *** hamza21 has quit IRC | 13:20 | |
| evrardjp | odyssey4me: correct | 13:22 |
| *** yolanda_ has joined #openstack-ansible | 13:22 | |
| *** armaan_ has quit IRC | 13:22 | |
| evrardjp | I am working on the tempest ones | 13:23 |
| evrardjp | to bring more resiliency to tempest, but will add more to build timeouts | 13:23 |
| mattt | odyssey4me: the last # i looked at yesterday were actual tempest failures | 13:23 |
| mattt | however i ran the test manually and it succeeded fine | 13:23 |
| *** armaan has joined #openstack-ansible | 13:23 | |
| evrardjp | mattt: I've seen one of those lately: so tempest run fine, but the assertion failed just after ? | 13:23 |
| mattt | evrardjp: there were 3 or 4 actual failures in each instance i looked at ... in most cases 4 | 13:24 |
| d34dh0r53 | evrardjp: I'm not sure about ironic deprecations, I believe prometheanfire was looking at that. As far as ELK goes, since we're moving to using journald for log shipping I'm looking at the possibility of using those logs on the logging node rather that having filebeat running in every container/host. So a bit of refactoring there, unfortunately the spectre/meltdown fix has nuked my multi-node aios and I'm currently trying to get them | 13:24 |
| d34dh0r53 | back up and running. | 13:24 |
| evrardjp | well I mean you don't have gibberish things in tempest output, but there is a real issue | 13:24 |
| ArchiFleKs | hi,I m trying to upgrade to latest stable/pike, I get pip error like this : http://paste.openstack.org/show/718381/ | 13:24 |
| *** yolanda has quit IRC | 13:24 | |
| evrardjp | ArchiFleKs: we are fixing Pike actively | 13:25 |
| mattt | evrardjp: tons of jibberish output, which reminds me actually trying to read that is near on impossible :) | 13:25 |
| ArchiFleKs | on the lxc-setup-host and all the other follwing playbooks | 13:25 |
| evrardjp | ArchiFleKs: https://review.openstack.org/#/c/558229/ | 13:25 |
| *** throwsb1 has joined #openstack-ansible | 13:25 | |
| ArchiFleKs | evrardjp: ok this is know, thanks :) | 13:25 |
| evrardjp | mattt: is the reason a failure to teardown or to answer in a timely fashion? | 13:26 |
| *** kstev has joined #openstack-ansible | 13:26 | |
| evrardjp | ArchiFleKs: yes pip 10 was released and we have not constrained it | 13:26 |
| evrardjp | (summarized version) | 13:26 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Improve ansible host & venv bootstrap https://review.openstack.org/558809 | 13:26 |
| evrardjp | mattt: on that gibberish thing, it's probably the now usual timeout of doing an action in tempest | 13:27 |
| *** jwitko has joined #openstack-ansible | 13:27 | |
| *** b1tsh1ft3r has joined #openstack-ansible | 13:27 | |
| d34dh0r53 | cloudnull: what is the non spectred kernel for ubuntu? 4.4.0-103? | 13:28 |
| mattt | evrardjp: the last readable message i see in that tempest blob for a failed build is "Unable to complete operation on subnet 6070a8a8-e85f-4ea6-a6e0-ca427e769994: One or more ports have an IP allocation from this subnet." | 13:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Skip host pip installs for ansible bootstrap https://review.openstack.org/558810 | 13:28 |
| b1tsh1ft3r | Is there a basic overview / guide on setting up designate utilizing powerDNS to run in containers via OSA in an HA environment? I realize this is somewhat specific scenario, but figured id ask as it might be more common than im aware of | 13:29 |
| *** esberglu has joined #openstack-ansible | 13:29 | |
| odyssey4me | b1tsh1ft3r assuming there's a well established role out there to setup powerdns, the rest wouldn't be too hard - and no, I'm not aware of anyone putting something together to show how | 13:30 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install stable/queens: Specify defaults for pip_install_upper_constraints https://review.openstack.org/558813 | 13:31 |
| foutatoro | hi all, I got this error trying to rebuild OSA deployment http://paste.openstack.org/show/718382/ | 13:33 |
| *** kstev has quit IRC | 13:33 | |
| foutatoro | someone knows what's wrong ? because I follow the rebuild proccess decribed in the wiki | 13:33 |
| *** mugsie_ is now known as mugsie | 13:35 | |
| *** mugsie has quit IRC | 13:36 | |
| *** mugsie has joined #openstack-ansible | 13:36 | |
| evrardjp | b1tsh1ft3r: additional question, do you have powerdns already set up? | 13:36 |
| evrardjp | b1tsh1ft3r: I think it's worth documenting when you've done it :) | 13:37 |
| evrardjp | not sure which instructions you are talking about foutatoro | 13:38 |
| b1tsh1ft3r | I dont tbh. Just scoping out the initial overview of what it takes before diving in. | 13:38 |
| evrardjp | ok | 13:38 |
| b1tsh1ft3r | But yes, if i get something working here today ill share :) | 13:38 |
| evrardjp | maybe worth asking the designate team too | 13:38 |
| odyssey4me | evrardjp https://review.openstack.org/#/c/558523/1/inventory/group_vars/all/all.yml ? | 13:39 |
| evrardjp | odyssey4me: yeah :p it's the first bump here | 13:39 |
| evrardjp | we didn't bump after .0 was released | 13:39 |
| evrardjp | we faffled around for rc2 and put the .0 inside the .rc2 | 13:39 |
| evrardjp | so that .0 would be tagged .0 | 13:40 |
| evrardjp | and we forgot once again to bump | 13:40 |
| evrardjp | every release we forget. | 13:40 |
| odyssey4me | :/ | 13:40 |
| evrardjp | every new branch* | 13:40 |
| *** kstev has joined #openstack-ansible | 13:40 | |
| *** sawblade_ has joined #openstack-ansible | 13:40 | |
| evrardjp | you know what? | 13:40 |
| evrardjp | I will write it down. | 13:41 |
| *** evin has joined #openstack-ansible | 13:42 | |
| evrardjp | could someone vote on https://review.openstack.org/#/c/557650/ ? | 13:42 |
| *** sawblade6 has quit IRC | 13:43 | |
| odyssey4me | evrardjp if we can get https://review.openstack.org/558811 merged then I can sha bump the role in https://review.openstack.org/558809 | 13:44 |
| *** vnogin has quit IRC | 13:45 | |
| *** vnogin has joined #openstack-ansible | 13:45 | |
| *** flemingo has joined #openstack-ansible | 13:45 | |
| *** armaan has quit IRC | 13:49 | |
| *** armaan has joined #openstack-ansible | 13:49 | |
| *** flemingo has quit IRC | 13:50 | |
| evrardjp | I force merge it because it blocks O | 13:52 |
| evrardjp | I'll re-release O | 13:52 |
| evrardjp | I don't think it's good to merge the sha bump patch with this one | 13:53 |
| odyssey4me | I think it may be required | 13:55 |
| *** lbragstad has quit IRC | 13:56 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/ocata: Update all SHAs for 15.1.19 https://review.openstack.org/558532 | 13:59 |
| evrardjp | odyssey4me: you are right it's better | 14:00 |
| *** flemingo has joined #openstack-ansible | 14:03 | |
| *** aruns has quit IRC | 14:04 | |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install stable/ocata: Use constraints with get-pip.py https://review.openstack.org/558811 | 14:04 |
| *** sawblade6 has joined #openstack-ansible | 14:05 | |
| *** flemingo has quit IRC | 14:08 | |
| *** sawblade_ has quit IRC | 14:08 | |
| evrardjp | jmccrory: are you there? | 14:08 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Improve ansible host & venv bootstrap https://review.openstack.org/558809 | 14:09 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Skip host pip installs for ansible bootstrap https://review.openstack.org/558810 | 14:10 |
| *** armaan has quit IRC | 14:25 | |
| *** flemingo has joined #openstack-ansible | 14:25 | |
| *** flemingo has quit IRC | 14:29 | |
| *** hvvclint has joined #openstack-ansible | 14:30 | |
| *** aedan9902 has quit IRC | 14:31 | |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install stable/pike: Specify defaults for pip_install_upper_constraints https://review.openstack.org/558814 | 14:33 |
| *** sawblade_ has joined #openstack-ansible | 14:35 | |
| *** sawblade6 has quit IRC | 14:35 | |
| *** lbragstad has joined #openstack-ansible | 14:35 | |
| foutatoro | evrardjp: I was talking about this section explain how to destroy and reduild a deployment https://docs.openstack.org/openstack-ansible/ocata/developer-docs/quickstart-aio.html#rebuilding-an-aio | 14:44 |
| Bico_Fino | Good morning! | 14:46 |
| Bico_Fino | I'm trying to add new compute host, openstack-ansible setup-hosts.yml --limit NEW_HOST_NAME it will create the /openstack/venvs/*? | 14:46 |
| *** vnogin has quit IRC | 14:47 | |
| evrardjp | are you running ocata foutatoro ? | 14:47 |
| evrardjp | Bico_Fino: I think we have docs for new compute hosts in our operators guide | 14:48 |
| *** vnogin has joined #openstack-ansible | 14:48 | |
| Bico_Fino | evrardjp checking | 14:48 |
| evrardjp | https://docs.openstack.org/openstack-ansible/latest/admin/scale-environment.html | 14:48 |
| evrardjp | yes we have | 14:49 |
| evrardjp | :) | 14:49 |
| Bico_Fino | testing. | 14:49 |
| Bico_Fino | also evrardjp https://bugs.launchpad.net/openstack-ansible/+bug/1748740 | 14:49 |
| openstack | Launchpad bug 1748740 in openstack-ansible "Documentation missing for adding compute host with ceph" [Medium,Confirmed] | 14:49 |
| *** flemingo has joined #openstack-ansible | 14:50 | |
| *** Rhys has quit IRC | 14:51 | |
| *** b1tsh1ft3r has quit IRC | 14:51 | |
| *** vnogin has quit IRC | 14:53 | |
| Bico_Fino | evrardjp "Error while linking: [Errno 2] No such file or directory", "path": "/openstack/venvs/nova-16.0.10/lib/python2.7/libguestfsmod.so", "state": "absent"} | 14:54 |
| Bico_Fino | setup-hosts should install /openstack/venvs right? | 14:54 |
| odyssey4me | Bico_Fino nope, the venvs are put in place by their respective service playbooks | 14:54 |
| *** flemingo has quit IRC | 14:55 | |
| Bico_Fino | os-nova-install then | 14:55 |
| foutatoro | evrardjp: yes 15.1.15 | 14:56 |
| Bico_Fino | odyssey4me so openstack-ansible os-nova-install.yml --limit NEWHOST, should install /openstack/venvs/nova-xxx right? | 14:58 |
| evrardjp | foutatoro: then I guess that's the documentation -- you found a bug? | 14:58 |
| foutatoro | evrardjp: yes I think there are some missing information. Things are not running as expected | 14:59 |
| cloudnull | mornings | 15:00 |
| Bico_Fino | morning cloudnull ! | 15:00 |
| cloudnull | hows it Bico_Fino | 15:02 |
| *** DanyC has quit IRC | 15:02 | |
| Bico_Fino | Not bad and you? | 15:02 |
| openstackgerrit | Merged openstack/openstack-ansible stable/queens: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558777 | 15:02 |
| odyssey4me | Bico_Fino it should, but it may not work as expected with limits | 15:02 |
| cloudnull | doing ok, just getting started doing typie typie work :) | 15:03 |
| Bico_Fino | so, run os-nova-install playbook without limit odyssey4me ? | 15:03 |
| odyssey4me | Bico_Fino sure, unless you can figure out why it's failing | 15:03 |
| Bico_Fino | testing. | 15:04 |
| odyssey4me | I do know that using limits can get complicated | 15:04 |
| Bico_Fino | Just trying to add a new fresh host as a compute node. :P | 15:04 |
| *** rcarrillocruz has quit IRC | 15:05 | |
| *** flemingo has joined #openstack-ansible | 15:05 | |
| *** vnogin has joined #openstack-ansible | 15:09 | |
| *** flemingo has quit IRC | 15:10 | |
| *** sawblade6 has joined #openstack-ansible | 15:17 | |
| *** sawblade_ has quit IRC | 15:19 | |
| foutatoro | hi all, any suggestion about solving that error http://paste.openstack.org/show/718382/ | 15:25 |
| odyssey4me | foutatoro there should be a log file on the host which will show what went wrong | 15:26 |
| odyssey4me | oh actually, looks like you're using something old | 15:27 |
| odyssey4me | foutatoro the error is pretty clearly showin in the ansible output: E: Failed to fetch http://archive.ubuntu.com/ubuntu/dists/xenial-updates/main/source/Sources 503 Service Unavailable | 15:27 |
| odyssey4me | foutatoro did you wipe the containers and are trying to rebuild? | 15:28 |
| dmsimard | odyssey4me: there's an odd issue with the OSA ara-report thing | 15:29 |
| odyssey4me | dmsimard oh? | 15:29 |
| dmsimard | odyssey4me: I was testing mod_wsgi tuning because I noticed some performance issues | 15:29 |
| dmsimard | I'm not sure I understand why this is providing different results | 15:30 |
| odyssey4me | foutatoro if you did, then the apt proxy is configured on the host and it's trying to use the repo container, which doesn't exist - make sure you wipe out the apt.conf.d file as specified in https://docs.openstack.org/openstack-ansible/latest/user/aio/quickstart.html#rebuilding-an-aio on all hosts | 15:30 |
| dmsimard | - http://logs.openstack.org/21/557921/6/check/openstack-ansible-functional-ubuntu-xenial/6411906/logs/ara-report/ <-- logs.o.o | 15:30 |
| dmsimard | - http://logs-dev.openstack.org/21/557921/6/check/openstack-ansible-functional-ubuntu-xenial/6411906/logs/ara-report/ <-- logs-dev.o.o | 15:30 |
| dmsimard | maybe something at the vhost level .. | 15:30 |
| odyssey4me | dmsimard how can I help? | 15:31 |
| dmsimard | not sure yet, mostly pointing out to pay attention if you see inconsistent results while I figure this out | 15:32 |
| odyssey4me | dmsimard I thought it was a little slower, but that's ok and kinda makes sense seeing as it's reading it live | 15:33 |
| *** flemingo has joined #openstack-ansible | 15:33 | |
| dmsimard | odyssey4me: the slow part will be addressed by https://review.openstack.org/#/q/topic:logs-wsgi | 15:33 |
| odyssey4me | I see that you've got a few new repositories in place, getting ready for the component split for v1? :) I know we've got quite a few operators waiting for that to use ARA for operations. | 15:33 |
| dmsimard | which is what I've been working on | 15:33 |
| dmsimard | odyssey4me: 1.0 is still some time out | 15:34 |
| dmsimard | odyssey4me: not before vancouver for sure but maybe usable in some shape or form by then | 15:34 |
| dmsimard | but yes, things are officially progressing and there's a new core contributor | 15:34 |
| *** cjloader has joined #openstack-ansible | 15:34 | |
| odyssey4me | ah, nice! | 15:34 |
| *** evrardjp_ has joined #openstack-ansible | 15:35 | |
| dmsimard | I'm due for a blog post/status update | 15:35 |
| dmsimard | sometime in the near future | 15:35 |
| dmsimard | You can see the new backend/API take shape here https://github.com/openstack/ara-server | 15:35 |
| *** evrardjp has quit IRC | 15:36 | |
| *** dariko has joined #openstack-ansible | 15:36 | |
| johnsom | Hi, I was looking at https://review.openstack.org/#/c/558229/ and the openstack-ansible-deploy-aio-centos-7 gate failure. There is a bad URL somewhere that has "centos" in it twice: | 15:36 |
| johnsom | fatal: [aio1]: FAILED! => {"attempts": 5, "changed": true, "failed": true, "msg": "http://mirror.iad.rax.openstack.org/centos/centos/7/cloud/x86_64/openstack-pike/repodata/repomd.xml: [Errno 14] HTTP Error 404 - Not Found\nTrying other mirror.\nTo address this issue please refer to the below wiki article \n\nhttps://wiki.centos.org/yum-errors\n\nIf above article doesn't help to resolve this issue please use | 15:37 |
| johnsom | https://bugs.centos.org/.\n\nhttp://mirror.iad.rax.openstack.org/centos/centos/7/virt/x86_64/kvm-common/repodata/repomd.xml: [Errno 14] HTTP Error 404 - Not Found\nTrying other mirror.\nhttp://mirror.iad.rax.openstack.org/centos/centos/7/cloud/x86_64/openstack-pike/common/debootstrap-1.0.81-1.el7.noarch.rpm: [Errno 14] HTTP Error 404 - Not Found | 15:37 |
| johnsom | Hmmm, sorry that didn't pastebin, it's usually automagic for me | 15:37 |
| *** flemingo has quit IRC | 15:37 | |
| odyssey4me | johnsom hmm, nice find | 15:37 |
| foutatoro | odyssey4me: yes I knew that OSA was trying to use the container repo. I clear all apt cache and pip before renning again the setup-hosts but the error still araise | 15:37 |
| odyssey4me | johnsom this is in the integrated build, or in the tests repo? | 15:38 |
| johnsom | odyssey4me I just looked at the check gate on https://review.openstack.org/#/c/558229/ | 15:38 |
| odyssey4me | foutatoro if you've cleared the apt cacher file and it's still an issue, then you have a dns resolution error or a mirror access error | 15:38 |
| *** DanyC has joined #openstack-ansible | 15:39 | |
| odyssey4me | johnsom it'd help if you could point to the exact line or task in the ara report? | 15:40 |
| *** admin0 has quit IRC | 15:40 | |
| johnsom | http://logs.openstack.org/29/558229/1/check/openstack-ansible-deploy-aio-centos-7/6cefb67/job-output.txt.gz#_2018-04-04_08_30_45_065473 | 15:40 |
| johnsom | Yep, sorry, still starting my coffee for the day | 15:40 |
| johnsom | ara didn't seem to render right for that job/run | 15:40 |
| odyssey4me | johnsom FYI our ara report is in logs/ara, not in the log root for the job (that's the zuul ara report) | 15:42 |
| Bico_Fino | odyssey4me I ran os-nova-install and setup-hosts with no limit, still no /openstack/venvs on the new host. | 15:42 |
| johnsom | Ah, gotcha. Pretty sure that is one I re-learn every time I look at these jobs. Too used to our gates. | 15:43 |
| odyssey4me | Bico_Fino if you have the output from the ansible run, check the tasks and see why they're either skipped or erroring, because that's weird | 15:43 |
| Bico_Fino | Going run again and save the output | 15:43 |
| *** cjloader has quit IRC | 15:44 | |
| odyssey4me | Bico_Fino you have a saved output log already | 15:44 |
| *** cjloader has joined #openstack-ansible | 15:44 | |
| odyssey4me | Bico_Fino look at /openstack/log/ansible-logging/ansible.log | 15:44 |
| *** pcaruana|afk| has quit IRC | 15:45 | |
| Bico_Fino | checking | 15:45 |
| *** goldenfri has joined #openstack-ansible | 15:45 | |
| *** armaan has joined #openstack-ansible | 15:46 | |
| *** admin0 has joined #openstack-ansible | 15:46 | |
| Miouge | Getting some <markers 'python_version == "3.4"' don't match your environment> on the pip install task. It’s non-blocking, but is it bad? | 15:46 |
| *** ianychoi__ is now known as ianychoi | 15:50 | |
| *** Rhys has joined #openstack-ansible | 15:52 | |
| *** Rhys has quit IRC | 15:52 | |
| *** mardim is now known as mardim_PTO | 15:53 | |
| *** chyka has joined #openstack-ansible | 15:53 | |
| *** flemingo has joined #openstack-ansible | 15:54 | |
| *** Rhys has joined #openstack-ansible | 15:54 | |
| johnsom | On the https://review.openstack.org/#/c/558229/ patch, it looks like designate is broken, causing a bunch of retries that is tipping it over the timeout: http://logs.openstack.org/29/558229/1/check/openstack-ansible-deploy-aio-opensuse-423/c2b3478/job-output.txt.gz#_2018-04-04_10_17_05_171604 | 15:54 |
| *** DanyC_ has joined #openstack-ansible | 15:55 | |
| odyssey4me | looks like that's the same issue solved by https://review.openstack.org/#/c/533650/ - but I'm not aware of us porting that back to pike at all | 15:56 |
| odyssey4me | both those problems are on non-voting gates though | 15:57 |
| odyssey4me | Miouge no, that's just because there are constraints which are python version specific - they're being ignored, as they should | 15:57 |
| johnsom | Yeah, true, I just started at the top. See above coffee statement | 15:57 |
| *** vnogin has quit IRC | 15:57 | |
| odyssey4me | haha | 15:57 |
| *** DanyC has quit IRC | 15:57 | |
| odyssey4me | the issue that's been hurting the patch on the voting gates has been two fold - one is that tempest tests fail due to timing out, the other is that the build hits the 2 hr timeout | 15:58 |
| *** flemingo has quit IRC | 15:58 | |
| odyssey4me | the first evrardjp is looking at increasing timeouts in tempest | 15:59 |
| odyssey4me | the timeouts are being hit, it seems due to resource exhaustion | 15:59 |
| odyssey4me | both problems could be solved by reducing the number of components we execute in the test, but that would require mixing up the current fix with a bunch of other changes, which we'd prefer not to do | 16:00 |
| d34dh0r53 | looking at the last gate for 558229 tempest initially failed on a timeout (300s) waiting for the test VM to come up, then the remaining tests failed | 16:03 |
| evrardjp_ | d34dh0r53: yes there seems to be an issue with tempest not respecting what I am telling it to do | 16:05 |
| *** evrardjp_ is now known as evrardjp | 16:05 | |
| d34dh0r53 | tempest.lib.exceptions.TimeoutException: Request timed out\n Details: (TestMinimumBasicScenario:test_minimum_basic_scenario) Server 3945352f-7b2e-4c1a-a537-183f298ac2ee failed to reach ACTIVE status and task state \"None\" within the required time (300 s). Current status: BUILD. Current task state: spawning.\n | 16:05 |
| d34dh0r53 | oops, sorry, meant to paste a link :/ | 16:05 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install stable/pike: Correct CentOS mirror URL https://review.openstack.org/558862 | 16:06 |
| evrardjp | d34dh0r53: 300s is the default timeout | 16:06 |
| odyssey4me | johnsom ^ should fix the first issue you picked up | 16:06 |
| d34dh0r53 | hmm | 16:06 |
| *** DanyC_ has quit IRC | 16:08 | |
| *** DanyC has joined #openstack-ansible | 16:08 | |
| *** aludwar has quit IRC | 16:17 | |
| *** aludwar has joined #openstack-ansible | 16:19 | |
| johnsom | Nice | 16:21 |
| johnsom | Glad I was remotely helpful. Grin | 16:21 |
| *** Leo_m has joined #openstack-ansible | 16:21 | |
| *** smorant has quit IRC | 16:22 | |
| mbuil | evardjp: I have a solution for the fwaas bug. How do you guys create the commit message of a patch that fixes a bug? Should I add BUG: 1752073 in the second line? Or I added to the title? or...? Every community has its ways :P | 16:22 |
| openstack | bug 1752073 in openstack-ansible "Neutron firewall_v2 driver missing" [Undecided,New] https://launchpad.net/bugs/1752073 - Assigned to Manuel Buil (mbuil) | 16:22 |
| johnsom | Ok, so, Your other issue on https://review.openstack.org/#/c/558229/ openstack-ansible-deploy-ceph-ubuntu-xenial | 16:23 |
| johnsom | http://logs.openstack.org/29/558229/1/gate/openstack-ansible-deploy-ceph-ubuntu-xenial/aa549d5/logs/host/libvirt/qemu/instance-00000001.log.txt.gz#_2018-04-04_12_35_01_728 | 16:23 |
| evrardjp | mbuil: we have something for you | 16:23 |
| mbuil | evrardjp: oh... I hope something nice ;) | 16:23 |
| evrardjp | https://docs.openstack.org/openstack-ansible/latest/contributor/contribute.html#general-guidelines-for-submitting-code | 16:23 |
| johnsom | So, you have KVM enabled. OVH has a buggy version of their hypervisor on some hosts. This means that even cirros will fail to boot. We have the same problem with octavia and excluded OVH from KVM | 16:23 |
| johnsom | See this: https://review.openstack.org/#/c/543034/24/playbooks/Octavia-DSVM/pre.yaml | 16:24 |
| johnsom | So your tempest tests are failing to start up the test servers.... | 16:24 |
| johnsom | The smoking gun is the "hardware error 0x0" | 16:25 |
| johnsom | We have been talking with OVH and have an upstream kernel bug for it too | 16:25 |
| evrardjp | we check that and decide to run qemu or kvm | 16:25 |
| evrardjp | where can I see if I have this hardware error? | 16:26 |
| johnsom | Yeah, everything looks great on those hosts, but the guest kernel blows up. | 16:26 |
| evrardjp | is that new with the spectre things? | 16:26 |
| johnsom | /logs/host/libvirt/qemu/instance-00000002.log.txt.gz | 16:26 |
| *** udesale has joined #openstack-ansible | 16:26 | |
| johnsom | Look at your instance logs. If you have the crash you will have the hardware error and a dump. | 16:26 |
| odyssey4me | interesting | 16:26 |
| johnsom | Also check that "accel=kvm" is on for qemu | 16:27 |
| evrardjp | odyssey4me: I think that's what we use to decide whether kvm or qemu | 16:27 |
| evrardjp | except ovh :p | 16:27 |
| evrardjp | maybe we should check if ovh in gates, then force an user_* override to qemu | 16:27 |
| johnsom | All the tests come back fine on OVH, but they fail. We had to explicitly exclude them from KVM until it's fixed. It's also only certain hosts there. | 16:27 |
| evrardjp | well I think our timeouts are not only on ovh, but that definitely is helpful | 16:28 |
| johnsom | Yeah, this is a nasty random failure... | 16:28 |
| odyssey4me | evrardjp heh, interesting - for the role tests we force qemu: http://git.openstack.org/cgit/openstack/openstack-ansible-tests/tree/test-vars.yml#n198 | 16:28 |
| johnsom | Both Trove and Octavia got bit by this | 16:29 |
| evrardjp | mmm | 16:29 |
| evrardjp | odyssey4me: I didn't know that | 16:29 |
| odyssey4me | for the integrated build, we rely on detection: https://github.com/openstack/openstack-ansible-os_nova/blob/master/tasks/nova_virt_detect.yml | 16:29 |
| evrardjp | we should remove that | 16:29 |
| evrardjp | yeah | 16:29 |
| evrardjp | nova_virt_type should be set for ovh nodes | 16:29 |
| odyssey4me | that'll mean we have to add something into the bootstrap host to detect which provider it's on | 16:30 |
| evrardjp | yes | 16:30 |
| odyssey4me | this sucks | 16:30 |
| odyssey4me | lemme work something out - thanks for the hint johnsom ! | 16:30 |
| johnsom | NP | 16:30 |
| mbuil | evrardjp: that link is unfortunately not answering my question. I'll try the BUG:#NUMBER approach | 16:30 |
| evrardjp | you deal with bootstrap-host, I deal with the tests removal odyssey4me ? | 16:30 |
| johnsom | We have been e-mailing with OVH since the PTG, but so far no luck. I think we are low priority... | 16:30 |
| *** sawblade_ has joined #openstack-ansible | 16:31 | |
| *** flemingo_ has joined #openstack-ansible | 16:31 | |
| evrardjp | odyssey4me: haha we mess up with hostnames too. That's not a good thing | 16:32 |
| evrardjp | maybe we should use something else, like the zuul data | 16:32 |
| odyssey4me | that's exactly my plan ;) | 16:33 |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: Bug fix: Parenthesis needed when using ternary https://review.openstack.org/558871 | 16:33 |
| evrardjp | odyssey4me: good | 16:33 |
| evrardjp | do you plan to pass a var? | 16:33 |
| evrardjp | in the job? | 16:33 |
| evrardjp | Or simply consume what's exported in some way | 16:33 |
| evrardjp | (I don't know the interface there) | 16:33 |
| *** sawblade6 has quit IRC | 16:33 | |
| evrardjp | is that information stored in the /etc/nodepool ? | 16:34 |
| odyssey4me | I'm not entirely sure what the cleanest option is right now. | 16:34 |
| odyssey4me | I might just do a dirty hack for now to get us by, given that I'm already doing work on making a thing that reads the jobs and spits out things from them. | 16:35 |
| evrardjp | or the /etc/ci/? | 16:35 |
| evrardjp | let me try something real quick with tests/ | 16:36 |
| evrardjp | that's easy | 16:36 |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: Bug fix: fwaas_v2 was missing in the pip task https://review.openstack.org/558873 | 16:36 |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: Bug fix: fwaas_v2 was missing in the pip task https://review.openstack.org/558873 | 16:39 |
| *** sawblade6 has joined #openstack-ansible | 16:39 | |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: Bug fix: fwaas_v2 was missing in the pip task https://review.openstack.org/558873 | 16:40 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-tests master: [WIP] Run with KVM if we can https://review.openstack.org/558876 | 16:41 |
| *** sawblade_ has quit IRC | 16:41 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-tests master: [WIP] Run with KVM if we can https://review.openstack.org/558876 | 16:45 |
| *** epalper has quit IRC | 16:46 | |
| *** udesale has quit IRC | 16:46 | |
| foutatoro | fghaas: do you run that error before "fatal: [ceph_ceph-mon_container-c4280548]: FAILED! => {"failed": true, "msg": "{{ cidr_networks.storage }}: 'cidr_networks' is undefined"}" | 16:47 |
| evrardjp | foutatoro: cidr_networks should be defined in your openstack_user_config | 16:49 |
| mnaser | time to try deploying queens and seeing how much of it we can break :D | 16:49 |
| *** Sha000000 has joined #openstack-ansible | 16:50 | |
| evrardjp | mnaser: it's still early in the cycle ! | 16:50 |
| evrardjp | :) | 16:50 |
| evrardjp | things break :p | 16:50 |
| mnaser | i expect production level everything >:) | 16:50 |
| evrardjp | but I'd rather deploy Q than P right now. | 16:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Work around nested virt issues on OVH test nodes https://review.openstack.org/558878 | 16:50 |
| evrardjp | mnaser: good, us too. | 16:50 |
| evrardjp | :p | 16:50 |
| spotz | you guys give me ulcers:) | 16:50 |
| foutatoro | evrardjp: yes it's already defined into openstack_user_config | 16:50 |
| evrardjp | odyssey4me: how did you know NODEPOOL_PYPI_MIRROR ? | 16:51 |
| evrardjp | has ovh | 16:51 |
| evrardjp | sounds good | 16:51 |
| *** ibmko_ has quit IRC | 16:51 | |
| evrardjp | DARN YOU SMRT | 16:51 |
| evrardjp | let me do the thing tests | 16:51 |
| foutatoro | evrardjp: "storage: 172.29.244.0/22" | 16:51 |
| odyssey4me | evrardjp they all do, but I figured that the pypi mirror is most universal | 16:51 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-tests master: [WIP] Run with KVM if we can https://review.openstack.org/558876 | 16:53 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-tests master: Run with KVM if we can in tests https://review.openstack.org/558876 | 16:54 |
| evrardjp | odyssey4me: ^ | 16:54 |
| evrardjp | foutatoro: could you show us your o_u_c please? | 16:54 |
| odyssey4me | evrardjp should work, let's see what the gate says | 16:55 |
| evrardjp | yup when I saw your commit I thought why not :) | 16:56 |
| evrardjp | mbuil: it is | 16:56 |
| evrardjp | check on the links there | 16:56 |
| foutatoro | evrardjp: here is the openstack_user_config http://paste.openstack.org/show/718406/ | 16:56 |
| evrardjp | foutatoro: that's weird, your user_*.yml ? | 16:58 |
| mnaser | has anyone ever experimented with running OSA remotely rather than on a 'control' node | 16:59 |
| mnaser | like say.. just put it on my laptop and run it from here | 16:59 |
| foutatoro | evrardjp:: this is one is for user_variable and openstack_user_config http://paste.openstack.org/show/718408/ | 17:01 |
| evrardjp | foutatoro: try the [' '] notation instead of the . ? cidr_networks['storage'] | 17:03 |
| evrardjp | not sure it would change a thing | 17:04 |
| evrardjp | but I'd do this because it doesn't look like a property | 17:04 |
| *** shan5464_ is now known as shananigans | 17:05 | |
| shananigans | mnaser: We use a separate deploy box in some of our environment. It needs connectivity to the host, management(container) and storage networks. | 17:07 |
| mnaser | yeah but i have cases where i have access to all those networks directly via vpn | 17:07 |
| shananigans | mnaser: also beware that the ansible bootstrap wraps ansible which may cause some issues on your laptop. | 17:09 |
| mnaser | shananigans: yeah that's the concern | 17:09 |
| mnaser | shananigans: im considering a docker container to run this stuff inside of | 17:09 |
| *** jonher has quit IRC | 17:10 | |
| goldenfri | Anyone know why I would be able to ping google etc from an instance but not be able to get out over port 80 on an default AIO install before I go too deep down this rabbit hole? | 17:15 |
| *** haad1 has quit IRC | 17:16 | |
| mnaser | sounds like security groups | 17:16 |
| goldenfri | I also had to create a DNS forwarder on the host so the instance could get DNS, not sure if that is normal | 17:16 |
| mnaser | can you ping 8.8.8.8 | 17:16 |
| goldenfri | yea I can't ping eveything | 17:17 |
| goldenfri | *can | 17:17 |
| *** mathlin has quit IRC | 17:17 | |
| goldenfri | security group is wide open, first thing I checked | 17:17 |
| mnaser | can you ping google.com | 17:17 |
| goldenfri | yes | 17:17 |
| goldenfri | but curling anything just hangs | 17:17 |
| mnaser | mtu | 17:18 |
| mnaser | check your interfaces mtus | 17:18 |
| foutatoro | evrardjp: after changing to cidr_networks['storage'] I get this error fatal: [ceph_ceph-mon_container-c4280548]: FAILED! => {"failed": true, "msg": "ipaddr: unknown filter type: cidr_networks['container']"} | 17:19 |
| *** mbuil has quit IRC | 17:19 | |
| evrardjp | you have something weird going on there. | 17:20 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest stable/queens: Switch from testrepository to stestr https://review.openstack.org/557335 | 17:20 |
| evrardjp | it's like you're passing something wrong to a filter -- are you sure you're doing the right thing? Are you sure these are the overrides you should define, and you define them properly? | 17:20 |
| evrardjp | woot mattt ! stestr! | 17:20 |
| goldenfri | mtus all seem normal 1500 or 1450 on the taps and vxlan | 17:22 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Developer docs refactor https://review.openstack.org/557650 | 17:23 |
| foutatoro | evrardjp: Yes these variables should be defined according to use ceph as backend https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/user_variables.yml.prod-ceph.example | 17:25 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/ocata: Update all SHAs for 15.1.19 https://review.openstack.org/558532 | 17:25 |
| evrardjp | goldenfri: mnaser after mtus, check checksum drops :p | 17:26 |
| evrardjp | foutatoro: this is master, are you deploying master? | 17:27 |
| evrardjp | are you sure your roles are up to date too? | 17:27 |
| spotz | evrardjp: Do we have a bugs team? | 17:27 |
| evrardjp | yes | 17:28 |
| evrardjp | why? | 17:28 |
| goldenfri | ahh I think I figured it out, it is actually working just super slow which probably means its an IPv6 issue... :( | 17:28 |
| spotz | I'm reviewing your doc | 17:28 |
| evrardjp | anyone that subscribes | 17:28 |
| evrardjp | https://launchpad.net/~openstack-ansible-bugs | 17:28 |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Ensure that RDO/SUSE series repo instructions stay up to date https://review.openstack.org/558778 | 17:36 |
| *** adriano has joined #openstack-ansible | 17:37 | |
| foutatoro | evrardjp: I'm running ocata 15.1.15 | 17:37 |
| foutatoro | evrardjp: I don't see doc explain variables for cep | 17:38 |
| foutatoro | ceph | 17:38 |
| evrardjp | maybe that's because ceph isn't supported well on ocata? | 17:38 |
| evrardjp | (I am telling bad things htere, we have jobs in Ocata for ceph) | 17:39 |
| evrardjp | but that is probably a different docs than master | 17:39 |
| foutatoro | evrardjp: you suggest using pike ? | 17:39 |
| evrardjp | https://docs.openstack.org/project-deploy-guide/openstack-ansible/ocata/app-ceph.html | 17:40 |
| odyssey4me | evrardjp it worked :) http://logs.openstack.org/76/558876/4/check/openstack-ansible-functional-centos-7/c2296f4/logs/ara-report/result/e57edbe7-ee21-4782-846a-ee6012973eff/ | 17:40 |
| logan- | foutatoro: configuring ceph-ansible is mostly the same on ocata, but the docs were only added in pike: https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-config-prod-ceph.html | 17:40 |
| evrardjp | I am not familiar with ceph | 17:40 |
| evrardjp | odyssey4me: great | 17:41 |
| evrardjp | maybe that will help nova gates | 17:41 |
| evrardjp | and tempest gates | 17:41 |
| evrardjp | your patch for the integrated should do the trick | 17:42 |
| evrardjp | but I doubt that's all | 17:42 |
| *** sawblade_ has joined #openstack-ansible | 17:43 | |
| *** vnogin has joined #openstack-ansible | 17:44 | |
| odyssey4me | cjloader please do not recheck a patch which has not yet received results from the previous recheck | 17:46 |
| *** sawblade6 has quit IRC | 17:46 | |
| odyssey4me | while it should make no difference, I have seen it stop the existing test and start again once - and https://review.openstack.org/558229 is in the gate queue now, so if it resets then a bunch of people are gonna be mad | 17:47 |
| cjloader | odyssey4me my fault | 17:47 |
| *** electrofelix has quit IRC | 17:47 | |
| *** mathlin has joined #openstack-ansible | 17:48 | |
| odyssey4me | cjloader no worries, this is how we learn ;) | 17:49 |
| odyssey4me | cjloader FYI you can see the progress for all current tests in motion at http://zuul.openstack.org/status.html | 17:54 |
| cjloader | ah | 17:54 |
| cjloader | okay | 17:54 |
| *** mathlin has quit IRC | 17:54 | |
| *** mathlin has joined #openstack-ansible | 17:55 | |
| *** hamza21 has joined #openstack-ansible | 17:56 | |
| *** sawblade6 has joined #openstack-ansible | 17:56 | |
| *** sawblade_ has quit IRC | 17:58 | |
| *** mathlin has quit IRC | 18:00 | |
| *** adriano has quit IRC | 18:02 | |
| *** poopcat has joined #openstack-ansible | 18:03 | |
| *** jonher has joined #openstack-ansible | 18:06 | |
| *** shardy_mtg has quit IRC | 18:06 | |
| *** jonher_ has joined #openstack-ansible | 18:06 | |
| *** vnogin has quit IRC | 18:08 | |
| *** jonher has quit IRC | 18:11 | |
| *** armaan has quit IRC | 18:16 | |
| *** mathlin has joined #openstack-ansible | 18:17 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_octavia master: Downloads amphora images from artifact storage https://review.openstack.org/555084 | 18:17 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_nova master: Remove default for nova_glance_api_servers https://review.openstack.org/558888 | 18:17 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Skip host pip installs for ansible bootstrap https://review.openstack.org/558807 | 18:22 |
| *** vnogin has joined #openstack-ansible | 18:24 | |
| *** vnogin has quit IRC | 18:28 | |
| *** armaan has joined #openstack-ansible | 18:30 | |
| *** flemingo_ has quit IRC | 18:31 | |
| *** armaan has quit IRC | 18:35 | |
| *** armaan has joined #openstack-ansible | 18:36 | |
| d34dh0r53 | are we happy with https://review.openstack.org/#/c/558876/4? | 18:49 |
| d34dh0r53 | odyssey4me: evrardjp ^ | 18:49 |
| odyssey4me | evrardjp would you look at that, all green on the first try: https://review.openstack.org/558878 | 18:49 |
| odyssey4me | d34dh0r53 yep, I think that's a go - the one I pointed to is the same | 18:50 |
| evrardjp | yup | 18:50 |
| d34dh0r53 | ok, nice to see green :) | 18:50 |
| evrardjp | great | 18:51 |
| odyssey4me | johnsom for prez! | 18:51 |
| johnsom | Ugh-Oh, what did I do now? | 18:51 |
| d34dh0r53 | woot +1 | 18:51 |
| evrardjp | that can be backported | 18:51 |
| odyssey4me | yep, I'll definitely be porting those back once they merge | 18:52 |
| evrardjp | johnsom: change the world | 18:52 |
| * d34dh0r53 makes a note to check instance logs on persistent failures | 18:52 | |
| johnsom | Please don't tell anyone... grin | 18:52 |
| *** epalper has joined #openstack-ansible | 18:53 | |
| odyssey4me | I'm impressed that both CentOS and SUSE completed on time. | 18:53 |
| *** osnaya has joined #openstack-ansible | 18:54 | |
| odyssey4me | d34dh0r53 try checking out the http://zuul.openstack.org/builds.html screen to find the periodics :) that'd be most helpful for evrardjp | 18:55 |
| evrardjp | odyssey4me: SUSE is a good citizen | 18:56 |
| evrardjp | now | 18:56 |
| evrardjp | :) | 18:56 |
| odyssey4me | hahaha | 18:56 |
| evrardjp | yeah that page is basically my go to everyday | 18:56 |
| evrardjp | you don't see the branch directly, but you can see it in the log url | 18:57 |
| evrardjp | I thought I should write a parser. | 18:57 |
| evrardjp | does that work today? with a big NO/YES | 18:57 |
| odyssey4me | or write an extension to allow branch filtering too | 18:57 |
| evrardjp | pointing ppl to it | 18:57 |
| *** haad1 has joined #openstack-ansible | 18:58 | |
| odyssey4me | ah yes, I believe they're looking to make an API which will make stuff like that much easier to do | 18:58 |
| evrardjp | yeah IIRC | 18:58 |
| osnaya | @cloudnull @admin0 @evrardjp @odyssey4me quick q - I have a user "user1" for which I have generated a key and copied it to target host1, target host2. In /etc/ansible/hosts I have configured ansible_user=user1..... When I run setup-hosts.yml it throws error "ESTABLISH SSH CONNECTION FOR USER: root".... Where is the user=root configured other than /etc/ansible/hosts.... I have restriction on root account. I can use "user1" | 18:59 |
| odyssey4me | everyone hold your breath, https://review.openstack.org/558229 is doing tempest tests | 18:59 |
| odyssey4me | osnaya no, the root user needs to be used at this time | 19:00 |
| *** gkadam has quit IRC | 19:00 | |
| odyssey4me | we're working on resolving that, but it will be some time yet - if you'd like to help out, that'd be great | 19:00 |
| osnaya | On my setups, I am not allowed to use user=root, but can use user1 with sudo privileges.... which is effective root | 19:01 |
| odyssey4me | osnaya well, I guess that means you'll need to help us work through where the problems are and to push patches to fix them then :) | 19:01 |
| odyssey4me | I know this was one patch submitted: https://review.openstack.org/557901 | 19:02 |
| odyssey4me | that appears that it might be breaking the integrated gate though | 19:03 |
| osnaya | @odyssey4me I am using Pike... so there is no way around thru configuration? Is root hardcoded or configured some place? | 19:04 |
| odyssey4me | it's hard coded into the playbooks at least, not sure where else | 19:05 |
| odyssey4me | note that we only use ssh key based access, no password access is allowed - so it's safe enough | 19:05 |
| odyssey4me | osnaya it's entirely possible to change where it's hard coded in playbooks to rather use become & become_user... then see what else breaks - we'd love to see some patches, or bug reports with some triage/solution suggestions | 19:06 |
| *** epalper has quit IRC | 19:07 | |
| *** sawblade_ has joined #openstack-ansible | 19:08 | |
| odyssey4me | :( failure for pike again | 19:09 |
| odyssey4me | evrardjp thoughts on combining a backport of https://review.openstack.org/558878 and https://review.openstack.org/558229 ? | 19:09 |
| *** sawblade6 has quit IRC | 19:10 | |
| *** mathlin has quit IRC | 19:16 | |
| *** epalper has joined #openstack-ansible | 19:23 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Update pip_install role and work around OVH nested-virt issues https://review.openstack.org/558904 | 19:23 |
| odyssey4me | evrardjp d34dh0r53 ^ A separate combined patch - let's see which merges first, shall we? | 19:24 |
| *** flemingo_ has joined #openstack-ansible | 19:24 | |
| *** sawblade6 has joined #openstack-ansible | 19:24 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Update pip_install role and work around OVH nested-virt issues https://review.openstack.org/558904 | 19:24 |
| *** sawblade_ has quit IRC | 19:26 | |
| *** sawblade_ has joined #openstack-ansible | 19:27 | |
| admin0 | osnaya, if you are allowed to use sudo su and be root, but no direct root via auth keys .its the same thing | 19:28 |
| admin0 | unless sudo is logged | 19:28 |
| admin0 | you can try --become ? | 19:29 |
| *** sawblade6 has quit IRC | 19:29 | |
| odyssey4me | logan- jmccrory anyone around to take a look at https://review.openstack.org/558904 ? | 19:31 |
| *** armaan has quit IRC | 19:31 | |
| odyssey4me | hwoarang cloudnull ^ ? | 19:32 |
| *** adriano_ has joined #openstack-ansible | 19:34 | |
| *** vnogin has joined #openstack-ansible | 19:34 | |
| spotz | odyssey4me: looking | 19:34 |
| odyssey4me | thanks spotz | 19:35 |
| spotz | odyssey4me: I like little ones:) Still going through evrardjp's doc trying to decide on plurals/singulars:( | 19:35 |
| *** anonymous has joined #openstack-ansible | 19:36 | |
| odyssey4me | haha, I thought there was an openstack guideline for that? | 19:36 |
| *** anonymous is now known as Guest44281 | 19:36 | |
| *** epalper has quit IRC | 19:37 | |
| odyssey4me | need another review for https://review.openstack.org/558539 if anyone's available | 19:37 |
| *** armaan has joined #openstack-ansible | 19:38 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: Run with KVM if we can in tests https://review.openstack.org/558876 | 19:40 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/queens: Run with KVM if we can in tests https://review.openstack.org/558907 | 19:40 |
| *** haad1 has quit IRC | 19:41 | |
| *** gkadam has joined #openstack-ansible | 19:42 | |
| *** osnaya has quit IRC | 19:47 | |
| odyssey4me | I'm out for the night - cheerio all. | 19:48 |
| d34dh0r53 | ciao odyssey4me | 19:49 |
| *** sawblade6 has joined #openstack-ansible | 19:51 | |
| *** sawblade_ has quit IRC | 19:53 | |
| *** armaan has quit IRC | 19:54 | |
| *** sawblade6 has quit IRC | 19:59 | |
| *** esberglu has quit IRC | 20:03 | |
| *** sawblade6 has joined #openstack-ansible | 20:05 | |
| *** esberglu_ has joined #openstack-ansible | 20:08 | |
| *** sawblade_ has joined #openstack-ansible | 20:09 | |
| *** sawblade6 has quit IRC | 20:11 | |
| *** dariko has quit IRC | 20:16 | |
| *** sawblade6 has joined #openstack-ansible | 20:17 | |
| *** sawblade_ has quit IRC | 20:19 | |
| *** vnogin has quit IRC | 20:21 | |
| *** evin has quit IRC | 20:22 | |
| *** vnogin has joined #openstack-ansible | 20:25 | |
| cloudnull | odyssey4me: sorry been afk most of the day | 20:31 |
| cloudnull | looking now | 20:32 |
| *** vnogin has quit IRC | 20:35 | |
| *** sawblade_ has joined #openstack-ansible | 20:41 | |
| *** sawblade6 has quit IRC | 20:41 | |
| *** sawblade6 has joined #openstack-ansible | 20:49 | |
| *** sawblade_ has quit IRC | 20:51 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/queens: Update all SHAs for 17.0.2 https://review.openstack.org/558523 | 20:53 |
| openstackgerrit | Merged openstack/openstack-ansible master: Developer docs refactor https://review.openstack.org/557650 | 20:53 |
| *** kstev has quit IRC | 20:54 | |
| *** idlemind has joined #openstack-ansible | 21:04 | |
| prometheanfire | has osa raised timeouts recently? wondering if I should up reqs timeout | 21:07 |
| *** foutatoro has quit IRC | 21:09 | |
| evrardjp | nope we shouldn't -- we are already at max for most jobs | 21:09 |
| evrardjp | we should shorten the builds instead | 21:09 |
| cloudnull | evrardjp: anything in need of babysitting ? | 21:10 |
| prometheanfire | evrardjp: I thought gate got slower or something | 21:10 |
| evrardjp | prometheanfire: spectre/meltdown | 21:11 |
| prometheanfire | ya, within the last week | 21:11 |
| evrardjp | cloudnull: yes the Pike patch :( | 21:11 |
| cloudnull | the gift that keeps on giving :) | 21:11 |
| evrardjp | haha true | 21:11 |
| cloudnull | evrardjp: https://review.openstack.org/#/c/558229/ ? | 21:11 |
| *** tomtomtom has joined #openstack-ansible | 21:11 | |
| prometheanfire | I may increase our timeout then, since it's our cross gates that are slow (not our actual gates) | 21:11 |
| evrardjp | I will work on the tempest timeouts tomorrow, so that we can backport | 21:11 |
| evrardjp | cloudnull: yes | 21:12 |
| evrardjp | prometheanfire: oh | 21:12 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible stable/pike: Bump pip_install SHA to include constraints fix https://review.openstack.org/558229 | 21:12 |
| prometheanfire | evrardjp: I'm asking about it from a reqs perspective | 21:12 |
| evrardjp | prometheanfire: what do you mean? where is the job definition? | 21:12 |
| tomtomtom | hello, running the run-upgrade script for ocata->pike, running into CONFIG_BRIDGE_NF_EBTABLES= is not a valid pattern, anyone run into this before and have any pointers? | 21:12 |
| evrardjp | prometheanfire: do you exercice a full build in all the branches? | 21:13 |
| evrardjp | if it's only for master, you can probably run a lighter build | 21:13 |
| evrardjp | like our unit tests things | 21:13 |
| prometheanfire | evrardjp: we do | 21:13 |
| evrardjp | prometheanfire: do you mind if we talk about that tomorrow? | 21:13 |
| prometheanfire | we don't want to run the lighter build because the cost of failure is a bit higher for us | 21:13 |
| prometheanfire | sure | 21:13 |
| evrardjp | ok let's discuss that tomorrow when I am fresher | 21:13 |
| prometheanfire | evrardjp: check .zuul.d/cross-jobs.yaml in reqs | 21:14 |
| evrardjp | ok | 21:14 |
| evrardjp | will do for tomorrow | 21:14 |
| evrardjp | cloudnull: else you can do whatever you want, the world is your oyster! | 21:14 |
| evrardjp | :p | 21:14 |
| cloudnull | tomtomtom: never seen that. | 21:14 |
| evrardjp | cloudnull: why the rebase? You just lost 30 minutes :p | 21:15 |
| cloudnull | evrardjp: the test had failed. | 21:15 |
| evrardjp | it was rechecked? | 21:15 |
| evrardjp | tomtomtom: I heard about it long ago | 21:15 |
| tomtomtom | cloudnull, yeah it happening on the controllers but not the other physical hosts. | 21:16 |
| evrardjp | are you running the latest code of ocata and pike ? | 21:16 |
| evrardjp | tomtomtom: I think this is used to define what to load in the openstack_hosts role | 21:16 |
| evrardjp | check there :p | 21:16 |
| evrardjp | I have to go | 21:16 |
| evrardjp | ttyl everyone! | 21:16 |
| cloudnull | evrardjp: it was failed since last recheck, just waiting for all of the other jobs to complete | 21:16 |
| cloudnull | rebase to save 30 min :) | 21:17 |
| cloudnull | tomtomtom: do you happen to have any overrides set for kernel modules? | 21:17 |
| evrardjp | hahah ok | 21:17 |
| evrardjp | cloudnull: just fyi: this is a competing commit: https://review.openstack.org/#/c/558904/ any of the two merging is fine. | 21:17 |
| tomtomtom | I did a git checkout stable/pike before running run-upgrade | 21:18 |
| cloudnull | tomtomtom: https://github.com/openstack/openstack-ansible/blob/master/releasenotes/notes/specific_kernel_modules_with_group_vars-8d169f564ffd450c.yaml | 21:18 |
| tomtomtom | I don't have any overrides set for kernel modules that I'm aware of. | 21:18 |
| evrardjp | that's probably for master | 21:18 |
| cloudnull | ah fair enoug | 21:18 |
| evrardjp | no I just think the check is not applicable to tomtomtom 's environement | 21:19 |
| evrardjp | but I don't have time to check that -- tomtomtom have a look in the role, it might help you | 21:20 |
| evrardjp | ok I am off for good this time! | 21:20 |
| evrardjp | ciao! | 21:20 |
| cloudnull | mind pasting the error ? | 21:21 |
| *** hamza21 has quit IRC | 21:25 | |
| *** kstev has joined #openstack-ansible | 21:29 | |
| tomtomtom | failed: [infra1] (item={u'pattern': u'CONFIG_BRIDGE_NF_EBTABLES=', u'group': u'network_hosts', u'name': u'ebtables'}) => {"failed": true, "item": {"group": "network_hosts", "name": "ebtables", "pattern": "CONFIG_BRIDGE_NF_EBTABLES="}, "msg": "CONFIG_BRIDGE_NF_EBTABLES= is not a valid pattern"} failed: [infra2] (item={u'pattern': u'CONFIG_BRIDGE_NF_EBTABLES=', u'group': u'network_hosts', u'name': u'ebtables'}) => {"faile | 21:32 |
| *** ibmko_ has joined #openstack-ansible | 21:33 | |
| *** gkadam has quit IRC | 21:36 | |
| *** kstev1 has joined #openstack-ansible | 21:36 | |
| *** kstev has quit IRC | 21:38 | |
| *** jwitko has quit IRC | 21:39 | |
| tomtomtom | hmmm.... did the user config file format change from ocata to pike? | 21:40 |
| *** Sha000000 has quit IRC | 21:41 | |
| *** DanyC has quit IRC | 21:44 | |
| *** esberglu_ has quit IRC | 21:57 | |
| *** chhagarw has quit IRC | 22:00 | |
| *** sawblade_ has joined #openstack-ansible | 22:03 | |
| *** cjloader has quit IRC | 22:03 | |
| *** sawblade6 has quit IRC | 22:05 | |
| *** xgerman_ has quit IRC | 22:15 | |
| *** xgerman_ has joined #openstack-ansible | 22:15 | |
| *** throwsb1 has quit IRC | 22:16 | |
| *** lbragstad has quit IRC | 22:18 | |
| *** Leo_m has quit IRC | 22:23 | |
| idlemind | so the pip-install in stable/pike seems to be still awaiting merging? | 22:28 |
| *** sawblade6 has joined #openstack-ansible | 22:30 | |
| cloudnull | tomtomtom: it there shouldnt be any changes there. | 22:30 |
| cloudnull | idlemind: yup. | 22:31 |
| cloudnull | tomtomtom: Ive not seen any issues in that area when doing ocata > pike upgrades. though Ive not done one in a while . | 22:32 |
| cloudnull | bgmccollum_: did not long ago | 22:32 |
| cloudnull | maybe he has? | 22:32 |
| *** sawblade_ has quit IRC | 22:33 | |
| cloudnull | tomtomtom: what pike release are you checking out ? | 22:33 |
| *** vnogin has joined #openstack-ansible | 22:36 | |
| *** ibmko_ has quit IRC | 22:40 | |
| *** vnogin has quit IRC | 22:41 | |
| tomtomtom | cloudnull; I'm running git checkout stable/pike | 22:44 |
| tomtomtom | is there a different one i should get? | 22:44 |
| *** bradm_ is now known as bradm | 22:44 | |
| cloudnull | no, that's good. | 22:44 |
| cloudnull | I was just making sure I'm looking at the right thigns | 22:44 |
| *** kstev1 has quit IRC | 22:47 | |
| *** sawblade_ has joined #openstack-ansible | 22:53 | |
| *** sawblade6 has quit IRC | 22:55 | |
| *** sawblade6 has joined #openstack-ansible | 23:07 | |
| *** sawblade_ has quit IRC | 23:10 | |
| *** flemingo has joined #openstack-ansible | 23:10 | |
| *** flemingo_ has quit IRC | 23:14 | |
| *** flemingo has quit IRC | 23:15 | |
| *** sawblade_ has joined #openstack-ansible | 23:16 | |
| *** sawblade6 has quit IRC | 23:18 | |
| *** chyka has quit IRC | 23:30 | |
| *** sawblade6 has joined #openstack-ansible | 23:37 | |
| *** sawblade_ has quit IRC | 23:39 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_octavia master: Downloads amphora images from artifact storage https://review.openstack.org/555084 | 23:47 |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_octavia master: Downloads amphora images from artifact storage https://review.openstack.org/555084 | 23:49 |
| *** sawblade6 has quit IRC | 23:49 | |
| *** sawblade6 has joined #openstack-ansible | 23:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!