| *** yadnesh|away is now known as yadnesh | 04:03 | |
| opendevreview | Gregory Thiemonge proposed openstack/octavia master: Add check for duplicate members in batch update https://review.opendev.org/c/openstack/octavia/+/864192 | 07:28 |
|---|---|---|
| skraynev | Hi folks. I have a question about Octavia API in yoga. | 08:22 |
| gthiemonge | skraynev: hi | 08:22 |
| skraynev | It uses pecan==1.4.1 and it always returns HTML response. I use the following header in HTTP request: 'Accept: application/json, text/plain. */* ' . In Octavia ussuri, it was pecan==1.3.2 and with this header it returns JSON, but now I get HTML. | 08:24 |
| skraynev | I ask about it, because in official Octavia document it says, that Accept header is ignored: https://docs.openstack.org/api-ref/load-balancer/v2/#response-format | 08:25 |
| skraynev | so I have the following behaviour with Octavia yoga and later: | 08:26 |
| skraynev | 'accept: json/application' -> JSON | 08:27 |
| skraynev | 'accept: json/application, */*' -> HTML | 08:28 |
| skraynev | 'accept: json/application, text/plain' -> JSON | 08:28 |
| skraynev | 'accept: json/application, text/plain, */*' -> HTML | 08:28 |
| skraynev | 'accept: */*' -> JSON | 08:28 |
| gthiemonge | skraynev: do you mean this: https://paste.opendev.org/show/bffD7qbpYPMs9tjIfJ8D/ ? | 08:29 |
| skraynev | What should I do to skip this header? Is it bug in doc/Octavia? | 08:30 |
| skraynev | will it be enough to drop this header on web server side for each request to Octavia? or I could do something else to fix it? | 08:30 |
| skraynev | I don't want to update each url and use suffix.json | 08:30 |
| skraynev | gthiemonge: yes! | 08:30 |
| gthiemonge | hmmm. I have no answer ATM. Obviously, the behavior described in the doc doesn't match what we observe here | 08:32 |
| skraynev | gthiemonge: indeed :( That's why I ask some advice. Ideally, I hoped on some magic option in config, but looks like this behaviour was not found before and most people use "correct" Accept header or do not use it at all. | 08:35 |
| gthiemonge | using "Accept: application/json, */*;q=0.8" would fix it, json would have an higher priority | 08:36 |
| *** yadnesh is now known as yadnesh|afk | 08:47 | |
| skraynev | gthiemonge: yes. it works, however I would prefer to fix it somewhere in Octavia. Because in this case I have to patch all HTTP requests. Or patch header on webserver side. Btw will it be enough to patch all such HTTP requests on webserver side? p.s. I also expected, that Accept header will be ignored by Octavia (according doc) or it's not possible? | 09:16 |
| gthiemonge | skraynev: I'm not sure, I think everything is processed in one of our deps, maybe johnsom will have more info about it | 09:21 |
| *** yadnesh|afk is now known as yadnesh | 09:22 | |
| skraynev | gthiemonge: yeah. additional opinion is good. let's wait johnsom feedback | 09:23 |
| opendevreview | Gregory Thiemonge proposed openstack/octavia master: Fix deletion of listeners and pools with deleted certs https://review.opendev.org/c/openstack/octavia/+/859387 | 09:59 |
| opendevreview | Gregory Thiemonge proposed openstack/octavia-tempest-plugin master: DNM: testing Octavia tempest jobs on Ubuntu jammy (22.04) https://review.opendev.org/c/openstack/octavia-tempest-plugin/+/861369 | 14:31 |
| opendevreview | Tom Weininger proposed openstack/octavia master: Add cpu-pinning element https://review.opendev.org/c/openstack/octavia/+/849129 | 15:29 |
| opendevreview | Tom Weininger proposed openstack/octavia master: Amphora vertical scaling optimization in Octavia https://review.opendev.org/c/openstack/octavia/+/849130 | 15:29 |
| opendevreview | Tom Weininger proposed openstack/octavia master: Set nohz_full in cpu-pinning element https://review.opendev.org/c/openstack/octavia/+/853802 | 15:29 |
| opendevreview | Tom Weininger proposed openstack/octavia master: Optimize HAProxy SSL cache size https://review.opendev.org/c/openstack/octavia/+/853606 | 15:29 |
| johnsom | The doc was referring to when we removed XML support and that all responses should be JSON. | 15:30 |
| johnsom | I think we did some work to force responses back to JSON, but when someone requests with a alternate "Accept" type we may not be forcing the issue, i.e. the test case. | 15:32 |
| johnsom | I would open a story for it and see if we can fix it to only allow application/json or return a 406 | 15:33 |
| gthiemonge | johnsom: thanks | 15:47 |
| gthiemonge | #startmeeting Octavia | 16:00 |
| opendevmeet | Meeting started Wed Nov 23 16:00:27 2022 UTC and is due to finish in 60 minutes. The chair is gthiemonge. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:00 |
| opendevmeet | The meeting name has been set to 'octavia' | 16:00 |
| gthiemonge | Hi! | 16:00 |
| johnsom | o/ | 16:00 |
| tweining | o/ | 16:00 |
| oschwart | o/ | 16:01 |
| gthiemonge | #topic Announcements | 16:01 |
| gthiemonge | ** Migrating devstack jobs to Jammy | 16:01 |
| gthiemonge | devstack/tempest are ready for Ubuntu Jammy (one of the official tested runtimes for Antelope) | 16:02 |
| gthiemonge | we need to prepare it, octavia-tempest-plugin explicitly uses the focal node-pool | 16:02 |
| gthiemonge | I rebased this patch from tweining to verify if everything is fine | 16:02 |
| gthiemonge | https://review.opendev.org/c/openstack/octavia-tempest-plugin/+/861369 | 16:02 |
| tweining | https://lists.openstack.org/pipermail/openstack-discuss/2022-November/031260.html | 16:03 |
| tweining | (for reference) | 16:03 |
| gthiemonge | so far, no error in octavia-v2-dsvm-scenario | 16:04 |
| gthiemonge | but we need to fix the nodepools for the stable branches | 16:04 |
| gthiemonge | anyways, it is still a WIP patch ATM | 16:04 |
| gthiemonge | any questions? any other announcements? | 16:05 |
| tweining | the other patch for testing with jammy seems to show no issues | 16:06 |
| tweining | https://review.opendev.org/c/openstack/octavia/+/861366 | 16:06 |
| gthiemonge | ok great! | 16:07 |
| oschwart | Great news folks | 16:07 |
| tweining | I will remove the DNM tag from the tempest patch | 16:08 |
| gthiemonge | yes | 16:08 |
| opendevreview | Tom Weininger proposed openstack/octavia-tempest-plugin master: Testing Octavia tempest jobs on Ubuntu jammy (22.04) https://review.opendev.org/c/openstack/octavia-tempest-plugin/+/861369 | 16:09 |
| gthiemonge | #topic CI Status | 16:10 |
| gthiemonge | Nothing to report here | 16:10 |
| gthiemonge | maybe you could take a look at | 16:10 |
| gthiemonge | https://review.opendev.org/c/openstack/octavia/+/864391 | 16:10 |
| gthiemonge | it splits the FIPS job into 2 jobs, the FIPS job is too long, it reaches the max timeout in zuul | 16:11 |
| gthiemonge | (this is a periodic job) | 16:11 |
| oschwart | Is there any openstack documentation regarding those CI jobs? | 16:13 |
| oschwart | (where to start from etc) | 16:13 |
| tweining | https://docs.openstack.org/infra/openstack-zuul-jobs/ | 16:14 |
| tweining | that maybe? | 16:14 |
| gthiemonge | I have no idea | 16:15 |
| oschwart | It looks like a list of many jobs and roles | 16:15 |
| oschwart | But anyway, I will take a look at the other related patches we had so far and will try to understand, thanks | 16:16 |
| johnsom | #link https://zuul-ci.org/docs/zuul/latest/config/job.html | 16:16 |
| johnsom | That is the official docs for defining a Zuul job | 16:16 |
| oschwart | Wow that looks like exactly what I was looking for, thanks! | 16:16 |
| gthiemonge | #topic Brief progress reports / bugs needing review | 16:18 |
| johnsom | #link https://review.opendev.org/c/openstack/octavia-tempest-plugin/+/812260 | 16:19 |
| gthiemonge | Folks, I have a first patch for an annoying issue, some exceptions are displayed by the amphorav2 driver (without jobboard), these exceptions should be ignored and hidden by Octavia: | 16:19 |
| johnsom | #link https://review.opendev.org/c/openstack/octavia-tempest-plugin/+/830015 | 16:19 |
| gthiemonge | https://review.opendev.org/c/openstack/octavia/+/861704 | 16:19 |
| johnsom | Were rebased, they have sat unreviewed so long they bit rotted. Would appreciate a review on those | 16:19 |
| gthiemonge | johnsom: sure | 16:20 |
| oschwart | johnsom: sure | 16:20 |
| oschwart | It could be nice if you guys review this one https://review.opendev.org/c/openstack/octavia/+/860211 which already has one +2 | 16:21 |
| gthiemonge | ack | 16:21 |
| tweining | in the past week there was great progress on reviewing backports. apart from that there was very little progress I must say. | 16:21 |
| gthiemonge | I have also updated 2 patches for recently reported bugs: | 16:21 |
| gthiemonge | Fix deletion of listeners and pools with deleted certs | 16:22 |
| gthiemonge | https://review.opendev.org/c/openstack/octavia/+/859387 | 16:22 |
| gthiemonge | Add check for duplicate members in batch update | 16:22 |
| gthiemonge | https://review.opendev.org/c/openstack/octavia/+/864192 | 16:22 |
| tweining | I've updated the cpu pinning element patch to disable that feature in diskimage-create.sh by default. | 16:22 |
| tweining | https://review.opendev.org/c/openstack/octavia/+/849129 | 16:22 |
| gthiemonge | QG: BTW thanks for your reviews on those patches! | 16:22 |
| johnsom | FYI https://review.opendev.org/c/openstack/octavia/+/864192 has a zuul -1 | 16:24 |
| QG | gthiemonge: no problem, i'm testing right now the patchset 2 on https://review.opendev.org/c/openstack/octavia/+/859387 | 16:24 |
| gthiemonge | argh... probably not related, but I might add this failure to our etherpad | 16:24 |
| tweining | I almost forgot: the patch that unblocks the ussuri gate still waits for a second +2 | 16:25 |
| tweining | https://review.opendev.org/c/openstack/octavia-tempest-plugin/+/864886 | 16:25 |
| gthiemonge | last patch from me: | 16:26 |
| gthiemonge | https://review.opendev.org/c/openstack/octavia/+/864998 | 16:26 |
| gthiemonge | it fixes a bug on centos 9 stream (and maybe now on Jammy, IDK) | 16:26 |
| oschwart | I got them all in my open tabs, I will see how I can help | 16:28 |
| tweining | I guess that is enough homework for this week ;) | 16:28 |
| gthiemonge | yeah, that's a lot of stuff | 16:28 |
| gthiemonge | #topic Open Discussion | 16:30 |
| tweining | I have nothing else to discuss I think | 16:31 |
| oschwart | Me neither | 16:31 |
| johnsom | tweining -1'd the ussuri patch, I think we should pin to the branch (stable/ussuri) and not a specific version tag | 16:31 |
| gthiemonge | barbican-tempest-plugin is branchless, right? | 16:32 |
| johnsom | Oh, opps, for some reason I read it as barbican, not tempest | 16:32 |
| tweining | ok, thanks for approving | 16:34 |
| gthiemonge | anything else? | 16:34 |
| johnsom | We have a lot of unreviewed tempest patches. Please have a look as you can. | 16:35 |
| gthiemonge | ack | 16:35 |
| oschwart | ack | 16:35 |
| johnsom | FYI, the next two days are a holiday in the US, so I will not be around. | 16:37 |
| gthiemonge | johnsom: enjoy! | 16:37 |
| tweining | see you next week then. enjoy | 16:37 |
| oschwart | Have fun in that break | 16:38 |
| gthiemonge | ok, I guess that's it | 16:38 |
| gthiemonge | thank you guys | 16:38 |
| gthiemonge | #endmeeting | 16:38 |
| opendevmeet | Meeting ended Wed Nov 23 16:38:28 2022 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:38 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/octavia/2022/octavia.2022-11-23-16.00.html | 16:38 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/octavia/2022/octavia.2022-11-23-16.00.txt | 16:38 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/octavia/2022/octavia.2022-11-23-16.00.log.html | 16:38 |
| *** yadnesh is now known as yadnesh|away | 16:41 | |
| opendevreview | Merged openstack/octavia-tempest-plugin master: Pin barbican-tempest-plugin to version 1.6.0 https://review.opendev.org/c/openstack/octavia-tempest-plugin/+/864886 | 18:28 |
| opendevreview | Michael Johnson proposed openstack/octavia master: Fix Octavia API HTTP Accept header handling https://review.opendev.org/c/openstack/octavia/+/865475 | 22:21 |
| opendevreview | Michael Johnson proposed openstack/octavia master: Fix Octavia API HTTP Accept header handling https://review.opendev.org/c/openstack/octavia/+/865475 | 22:26 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!