| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #airshipit | 04:33 | |
| *** roman_g has joined #airshipit | 06:18 | |
| *** seaneagan has quit IRC | 11:00 | |
| *** rajatSharma has joined #airshipit | 12:12 | |
| *** ianychoi_ has joined #airshipit | 12:13 | |
| *** rajatSharma has quit IRC | 12:14 | |
| *** ianychoi has quit IRC | 12:17 | |
| *** AshuKumar_ has joined #airshipit | 13:33 | |
| *** rajatSharma21 has joined #airshipit | 13:40 | |
| *** rajatSharma21 has quit IRC | 13:41 | |
| *** rajatSharma has joined #airshipit | 13:42 | |
| mattmceuen | Would appreciate review of the initial networking catalogue changeset for airshipctl -- it's been out there for a while and is currently happily rebased :) https://review.opendev.org/#/c/749611/ | 13:43 |
|---|---|---|
| AshuKumar_ | Hi mattmceuen Regarding https://review.opendev.org/#/c/748421 | 13:46 |
| *** diga has joined #airshipit | 13:46 | |
| mattmceuen | Hey @AshuKumar_ | 13:47 |
| AshuKumar_ | thanks for review comments first, need to discuss your second comment | 13:47 |
| mattmceuen | This one? "I don't think we want to kubectl apply the profiles, we want them to be included in a normal phase payload." | 13:47 |
| AshuKumar_ | file tools/deployment/34_deploy_worker_node.sh line:37 | 13:47 |
| AshuKumar_ | yes, to accomplish this we have to introduce a new phase worker-provision | 13:48 |
| AshuKumar_ | reason is HWCC need node in "ready state" to apply sample profiles | 13:49 |
| mattmceuen | I see | 13:49 |
| mattmceuen | remind me, is HWCC deployed in the initinfra phase? | 13:50 |
| rajatSharma | Yes | 13:50 |
| rajatSharma | It is deployed with BMO in worker initinfra. | 13:51 |
| mattmceuen | What happens if you include the profiles in `workers` phase? | 13:52 |
| mattmceuen | said differently, what happens if you apply a profile before the node is in ready state? | 13:54 |
| AshuKumar_ | profiles will fail | 13:54 |
| mattmceuen | I understand them failing at first, but then once the node becomes ready, they should then succeed, right? | 13:54 |
| rajatSharma | No if profile fails once then it cannot be | 13:56 |
| rajatSharma | applied as there is no change in the | 13:56 |
| rajatSharma | profile kubernetes says yaml unchanged | 13:56 |
| AshuKumar_ | Actually in worker phase all profiles are applied at once through kustomization and node takes time in introspection so what happens is if we apply profile at that time hwcc will not find any host in ready state. | 13:56 |
| mattmceuen | AshuKumar_: yep that makes sense, but I would have expected the profile to be reevaluated after more nodes reached ready state | 13:57 |
| mattmceuen | But that could just be me misunderstanding something. Let's separate these things out: | 13:57 |
| mattmceuen | For now, yeah, let's add a new phase (how do you feel about worker-classification?) | 13:58 |
| rajatSharma | It will also include provision part | 13:58 |
| mattmceuen | Ok - I don't understand yet, but it'll make more sense during code review :) if not I'll leave a comment | 13:59 |
| mattmceuen | And then let's also discuss in the Airship design call to see if I'm crazy or if this should be an enhancement to the HWCC to re-evaluate profiles when nodes change state | 13:59 |
| mattmceuen | Does that sound good? | 13:59 |
| AshuKumar_ | it is designed like that, if profile is changed then only reconciliation happened | 14:00 |
| mattmceuen | Understood, that's why I was thinking of it as a potential enhancement (not necessarily a bug) | 14:00 |
| rajatSharma | Yess | 14:01 |
| mattmceuen | Ok cool. Thanks for your quick response to the feedback | 14:01 |
| AshuKumar_ | yes, we can discuss in design call. Thanks | 14:01 |
| rajatSharma | Thanks mattmceuen | 14:02 |
| mattmceuen | any time rajatSharma, Ashu | 14:02 |
| *** rajatSharma has quit IRC | 14:02 | |
| *** AshuKumar_ has quit IRC | 14:07 | |
| *** awander has joined #airshipit | 14:15 | |
| *** diga has quit IRC | 14:15 | |
| airship-irc-bot | <se6518> Please, review proxy config/replacement transformer change: https://review.opendev.org/755085 | 17:29 |
| airship-irc-bot | <mf4716> Please review PS related to airshipctl phase validate subcommand: https://review.opendev.org/#/c/749638/ [on behalf of, Srinivasa Muly] | 17:34 |
| *** roman_g has quit IRC | 20:33 | |
| *** roman_g has joined #airshipit | 20:34 | |
| *** roman_g has quit IRC | 20:38 | |
| airship-irc-bot | <sidney.shiba> Hello, after the rebase I started getting the error `"Missing repository name."`. Do I need to add something in a config file, e.g., $HOME/.airship/config? | 21:03 |
| airship-irc-bot | <sidney.shiba> I think I found where but seems the yaml syntax is not correct: `subPath: manifests/site/az-test-site` `targetPath: /home/esidshi/projects/phase2/airshipctl` `repositories:` `- manifests/site/az-test-site` `- manifests/site/gcp-test-site` `- manifests/site/docker-test-site` `- manifests/site/openstack-test-site` `- manifests/site/test-site` `phaseRepositoryName: manifests/site/az-test-site` | 21:13 |
| airship-irc-bot | `metadataPath: manifests/metadata.yaml` | 21:13 |
| airship-irc-bot | <sidney.shiba> What is the expected syntax, please? | 21:14 |
| airship-irc-bot | <sidney.shiba> This patch set (https://review.opendev.org/#/c/756400/) seems to be related to the issue I am experiencing but the commit message is blank and no reference to any documentation. Can someone help? | 22:05 |
| airship-irc-bot | <sidney.shiba> I guess everybody is busy so closing shop for this week. Enjoy your weekend and stay safe. | 22:07 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!