*** chkumar|off is now known as chkumar|ruck | 06:44 | |
*** zoli is now known as zoli|wfh | 07:12 | |
*** zoli|wfh is now known as zoli | 07:13 | |
*** jpena|off is now known as jpena | 07:51 | |
sfbender | Merged logreduce master: tokenizer: handle hexa numbers https://softwarefactory-project.io/r/13602 | 08:04 |
---|---|---|
sfbender | Tristan de Cacqueray created software-factory/sf-config master: zuul/nodepool: switch to logrotate for log rotation https://softwarefactory-project.io/r/13772 | 08:17 |
sfbender | Jakub Ružička created rpmreq master: WIP: Dep Tree Traversal (TOO SLOW) https://softwarefactory-project.io/r/13773 | 08:32 |
sfbender | Merged rdopkg master: doc: fix spelling of rdopkg https://softwarefactory-project.io/r/13767 | 08:39 |
sfbender | Merged rdopkg master: tests: use pytest.parametrize in spec tests https://softwarefactory-project.io/r/13769 | 08:42 |
sfbender | Merged rdopkg master: specfile: document release_parts() return value https://softwarefactory-project.io/r/13770 | 08:45 |
sfbender | Merged rdopkg master: specfile: add comments to recognized_release() https://softwarefactory-project.io/r/13771 | 08:49 |
*** chkumar|ruck has quit IRC | 08:58 | |
*** chandankumar has joined #softwarefactory | 08:59 | |
*** chandankumar is now known as chkumar|ruck | 09:00 | |
*** jpena is now known as jpena|lunch | 11:21 | |
*** zoli is now known as zoli|lunch | 12:27 | |
*** zoli|lunch is now known as zoli | 12:27 | |
*** jpena|lunch is now known as jpena | 12:28 | |
pabelanger | tristanC: mhu: fbo: nhicher: Can I get a review of https://softwarefactory-project.io/r/13761 and https://softwarefactory-project.io/r/13768 please | 12:39 |
pabelanger | cleans up old nodepool tenant, and add max-ready-age for labels | 12:40 |
dmsimard | Ansible to adopt molecule and ansible-lint projects, https://groups.google.com/forum/m/#!topic/ansible-project/ehrb6AEptzA | 12:48 |
pabelanger | dmsimard: indeed! | 12:56 |
dmsimard | looks like the idea is to automatically run them on all galaxy roles or something | 12:57 |
dmsimard | heh, it would actually be easy with zuul ? | 12:57 |
dmsimard | well, "easy" | 12:57 |
pabelanger | Yah, details still to be worked out I believe. Being driven by the galaxy team | 12:58 |
pabelanger | but yah, basic PTI for role and scoring system | 12:58 |
pabelanger | more for local users to test then anything | 12:58 |
pabelanger | dmsimard: http://git.openstack.org/cgit/openstack/ansible-role-nodepool/tree/tox.ini#n38 | 13:01 |
pabelanger | example of how you can molecule with zuul today | 13:01 |
pabelanger | works okay from what I've seen | 13:02 |
*** zoli is now known as zoli|wfh | 13:06 | |
dmsimard | oh I remember seeing that patch | 13:07 |
*** sshnaidm is now known as sshnaidm|mtg | 13:07 | |
pabelanger | nhicher: tristanC: how do we modify clouds.yaml again in sf.io? I don't see it in config repo | 13:20 |
pabelanger | Want to add scj1 region to vexxhost-ansible-network | 13:20 |
nhicher | pabelanger: it not stored on git repo. If you provide the data I can update it | 13:21 |
pabelanger | nhicher: http://git.openstack.org/cgit/openstack-infra/system-config/tree/playbooks/templates/clouds/nodepool_clouds.yaml.j2#n58 | 13:23 |
pabelanger | we need to update regions key | 13:23 |
pabelanger | to include both ca-ymq-1 and sjc1 | 13:23 |
*** chkumar|ruck is now known as chandankumar | 13:26 | |
nhicher | ok | 13:28 |
nhicher | done pabelanger | 13:32 |
pabelanger | nhicher: thanks | 13:32 |
pabelanger | nhicher: and that is called vexxhost-ansible-networking for the cloud name right? | 13:34 |
nhicher | pabelanger: yes | 13:35 |
nhicher | pabelanger: I got a deprecation warning message https://softwarefactory-project.io/paste/show/1216/ | 13:35 |
pabelanger | nhicher: yah, needs to be list, not comma | 13:36 |
nhicher | I don't have comma https://softwarefactory-project.io/paste/show/1217/ | 13:37 |
pabelanger | nhicher: also, once warning is fixed, can we do the dance of creating vexxhost-ansible-network entry (copypasta), then I'll update nodepool.yaml to use it, then we can delete vexxhost-ansible-networking. This is to help save some characters in the hostname for nodes | 13:37 |
pabelanger | also ansible-networking is quite the same | 13:38 |
pabelanger | nhicher: Hmm is there another region_name? | 13:38 |
pabelanger | oh | 13:39 |
pabelanger | nhicher: regions | 13:39 |
pabelanger | not region_name | 13:39 |
pabelanger | nhicher: actually | 13:39 |
pabelanger | is this clouds.yaml | 13:40 |
pabelanger | or something that SF uses to generate clouds.yaml | 13:40 |
nhicher | ok, I will check how to fix that | 13:40 |
nhicher | so, I copy/paste vexxhost-ansible-networkink to rename it to vexxhost-ansible-network ? | 13:41 |
nhicher | networking* | 13:41 |
pabelanger | nhicher: yah, but we need to keep vexxhost-ansible-networking for a while (few hours) | 13:41 |
pabelanger | then I'll send the all clear to delete it | 13:41 |
nhicher | Ok, I prepare the change | 13:42 |
pabelanger | otherwise we'll break nodepool | 13:42 |
pabelanger | thanks | 13:42 |
nhicher | done pabelanger, new clouds.yaml deployed | 13:43 |
pabelanger | nhicher: thanks, will propose nodepool changes now | 13:43 |
pabelanger | remote: https://softwarefactory-project.io/r/13775 Switch to vexxhost-ansible-network for clouds.yaml | 13:45 |
pabelanger | nhicher: ^ | 13:45 |
spredzy | pabelanger: Could we get a new review on this one please https://github.com/ansible/zuul-config/pull/22 | 13:54 |
sfbender | Fabien Boucher created software-factory/managesf master: Fix issue where we expect to have the group resources https://softwarefactory-project.io/r/13776 | 13:57 |
pabelanger | spredzy: sure, will look shortly | 13:58 |
pabelanger | remote: https://softwarefactory-project.io/r/13777 Add vexxhost-ansible-network-sjc1 provider to nodepool | 13:58 |
pabelanger | nhicher: ^do you mind a review | 13:58 |
pabelanger | note the yaml anchors we are using to reduce copypasta :) | 13:59 |
*** zoli|wfh is now known as zoli | 14:02 | |
*** zoli is now known as zoli|wfh | 14:02 | |
*** zoli|wfh is now known as zoli | 14:02 | |
nhicher | pabelanger: done, yes, that's useful | 14:11 |
pabelanger | nhicher: thanks, if you don't mind watching to see if we upload images properly to the new region | 14:15 |
nhicher | pabelanger: it's in progress: Uploading DIB image build 0000000056 from /var/opt/rh/rh-python35/lib/nodepool/dib/ansible-centos-7-0000000056.raw to vexxhost-ansible-network-sjc1 | 14:17 |
nhicher | ansible-fedora-28-0000000015 in vexxhost-ansible-network-sjc1 is ready | 14:17 |
pabelanger | woot | 14:17 |
nhicher | ansible-centos-7-0000000056 in vexxhost-ansible-network-sjc1 is ready | 14:18 |
pabelanger | let me try to test | 14:18 |
pabelanger | Oh | 14:21 |
pabelanger | ha | 14:21 |
pabelanger | I didn't run cloud-launcher | 14:21 |
pabelanger | so, think port 22 will fail for SSH | 14:21 |
pabelanger | nhicher: do you mind check launcher debug log, haven't gotten a node yet | 14:28 |
nhicher | pabelanger: I'm checking | 14:40 |
nhicher | pabelanger: keypair, flavors and images are present | 14:45 |
pabelanger | nhicher: in launcher logs, is there any errors | 14:50 |
pabelanger | I wonder if launcher / zookeeper having capacity issues | 14:50 |
nhicher | I don't see any error pabelanger | 14:51 |
pabelanger | k, I've created new PR | 14:51 |
pabelanger | okay, building now | 14:51 |
nhicher | which label do you use for your test ? | 14:52 |
nhicher | I grep ansible-fedora-28-4vcpu on launcher.log but did not find anything | 14:52 |
pabelanger | ansible-fedora-28-1vcpu | 14:57 |
pabelanger | I see it currently building | 14:57 |
zoli | hi | 14:58 |
nhicher | hi zoli =) | 14:58 |
pabelanger | nhicher: it looks like sjc1 is also having issues upstream | 14:58 |
pabelanger | nhicher: so, might be issue with region | 14:58 |
nhicher | pabelanger: ok, waiting for your update =) | 14:58 |
nhicher | pabelanger: https://softwarefactory-project.io/paste/show/1218/ | 14:59 |
pabelanger | nhicher: thanks, see that too. I think there might be issues on compute side | 15:00 |
pabelanger | remote: https://softwarefactory-project.io/r/13779 Set launch-timeout for vexxhost-ansible-network providers | 15:10 |
pabelanger | nhicher: optimization to help in this case^ | 15:10 |
*** chandankumar is now known as chkumar|off | 15:26 | |
*** sshnaidm|mtg is now known as sshnaidm | 15:29 | |
nhicher | pabelanger: when using cloud-image, you only have to add a zuul user or there is another steps to prepare the image ? | 15:33 |
pabelanger | nhicher: right, you either need to add a zuul user / ssh key to the image or tell nodepool to use the cloud user the image ships | 15:34 |
pabelanger | for ansible-network cloud-images, we are rebuilding them and bootstraping a zuul user | 15:34 |
nhicher | pabelanger: ok, so add zuul user and populate .ssh/authorized key with zuul pub key should do the job | 15:35 |
pabelanger | yah | 15:36 |
nhicher | thanks pabelanger | 15:36 |
pabelanger | woot | 16:05 |
pabelanger | https://ansible-network.softwarefactory-project.io/logs/an/58/58/561744e6d94a2a234b8a9cc0ec0c7e6d33cd6a10/check/tox-linters/4664cb0/job-output.txt.gz#_2018-09-26_15_52_20_108476 | 16:05 |
pabelanger | nhicher: thanks for help, sjc1 worked | 16:05 |
pabelanger | nhicher: I think it is safe to delete the vexxhost-ansible-networking settings now on your side | 16:05 |
nhicher | pabelanger: great | 16:06 |
sfbender | Ken Dreyer created rdopkg master: specfile: final ".Z" digit is part of the version https://softwarefactory-project.io/r/13782 | 16:42 |
*** zoli is now known as zoli|gone | 17:08 | |
*** zoli|gone is now known as zoli | 17:08 | |
*** jpena is now known as jpena|off | 17:09 | |
pabelanger | mhu: nhicher: are you able to confirm nodepool-launcher is working properly for ansible-network-vexxhost-mtl1 provider | 17:58 |
pabelanger | https://softwarefactory-project.io/grafana/d/ddLmFcAmk/provider-vexxhost-ansible-network-mtl1?orgId=1 | 17:58 |
pabelanger | no longer see requests happening | 17:58 |
pabelanger | remote: https://softwarefactory-project.io/r/13784 Fix issue with provider layout for ansible-network | 18:25 |
pabelanger | nhicher: dmsimard: ^ is why we lost ansible-network-vexxhost-mtl1 | 18:25 |
pabelanger | providers don't merge | 18:25 |
pabelanger | I'll self approve now | 18:25 |
pabelanger | nhicher: Yup, that fixed it. Thanks dmsimard for looking at logs | 18:30 |
nhicher | pabelanger: I was afk, everything is ok now ? | 18:38 |
pabelanger | nhicher: yup, dmsimard tagged in to help | 18:39 |
pabelanger | guess he was able to still access nodes | 18:39 |
nhicher | cool, thanks dmsimard =) | 18:40 |
pabelanger | nhicher: did you discuss access for soembody in ansible-network? | 18:40 |
spredzy | guys, on zuul we have a job that is started and then stopped but nothing is reported to Github | 18:40 |
pabelanger | if not, that is okay | 18:40 |
spredzy | This is the PR affected https://github.com/ansible/awx/pull/2309 | 18:40 |
spredzy | Plus nothing shows up here https://ansible.softwarefactory-project.io/zuul/builds.html | 18:40 |
nhicher | pabelanger: no, I will discuss it tomorow | 18:41 |
pabelanger | spredzy: I have to run now, but can help in a bit | 18:41 |
spredzy | pabelanger: ack thanks. Make sure you cc matburt I might be off by then | 18:42 |
sfbender | Merged software-factory/cauth-distgit master: Add python-jwt dependency https://softwarefactory-project.io/r/13620 | 18:42 |
pabelanger | spredzy: recheck again, if fails we need to look at logs | 18:42 |
pabelanger | bbiab | 18:42 |
spredzy | we did, 2 times already | 18:44 |
spredzy | ack | 18:44 |
spredzy | I suspect some zuul configuration issue, job is not being picked up I'd say | 19:06 |
spredzy | tox job should be picked up being part of the python template | 19:20 |
spredzy | Stepping out for dinner, maybe when I get back issue will look obvious | 19:21 |
sfbender | Nicolas HICHER created software-factory/sf-config master: sf-repos: ensure all templated files contain a warning https://softwarefactory-project.io/r/13786 | 20:23 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!