| *** ysandeep|out is now known as ysandeep | 03:24 | |
| *** ysandeep is now known as ysandeep|afk | 04:34 | |
| *** ysandeep|afk is now known as ysandeep | 05:45 | |
| opendevreview | Gregory Thiemonge proposed openstack/octavia stable/train: WIP Testing DIB on stable/train https://review.opendev.org/c/openstack/octavia/+/845812 | 05:56 |
|---|---|---|
| opendevreview | Gregory Thiemonge proposed openstack/octavia stable/train: WIP Testing DIB on stable/train https://review.opendev.org/c/openstack/octavia/+/845812 | 06:48 |
| opendevreview | Gregory Thiemonge proposed openstack/octavia stable/train: WIP Testing DIB on stable/train https://review.opendev.org/c/openstack/octavia/+/845812 | 06:58 |
| manuvakery1 | johnsom: We use aodh for a lot of event notifications such as instance created,deleted,resized .. if octavia can sent out notifications for lb its very useful .. I hope https://review.opendev.org/c/openstack/octavia/+/831051 will get merged and with some minor changes in ceilometer I can get the required resources in gnocchi | 07:11 |
| gthiemonge | manuvakery1: this commit introduces 3 event types: https://review.opendev.org/c/openstack/octavia/+/831051/17/doc/source/admin/event-notifications.rst#41 | 07:12 |
| gthiemonge | manuvakery1: do you think you will need more events? | 07:12 |
| manuvakery1 | gthiemonge: I think these are enough | 07:13 |
| manuvakery1 | gthiemonge: the stats API is enough for other details .. Here I am more interested when an lb is created, updated, and deleted | 07:15 |
| gthiemonge | ok | 07:16 |
| *** ysandeep is now known as ysandeep|afk | 07:28 | |
| opendevreview | Gregory Thiemonge proposed openstack/octavia stable/train: WIP Testing DIB on stable/train https://review.opendev.org/c/openstack/octavia/+/845812 | 08:11 |
| *** ysandeep|afk is now known as ysandeep | 09:38 | |
| opendevreview | Gregory Thiemonge proposed openstack/octavia stable/train: Fix amphora build on stable/train https://review.opendev.org/c/openstack/octavia/+/845812 | 09:57 |
| gthiemonge | EmilienM: ^ it should fix the gate for stable/train | 09:59 |
| EmilienM | gthiemonge: thanks a lot! | 11:10 |
| EmilienM | gthiemonge: can we also apply it to ussuri? I believe we have the issue there as well | 11:10 |
| gthiemonge | EmilienM: yep, I'll do it | 11:20 |
| EmilienM | gthiemonge: merci ;-) | 11:22 |
| Namrata | Hi Folks, I want to create octavia L7 policy redirecting URL. I am not able to figure out how can I redirect https://abc.com/foo to https://efg.com/foo Can someone help me with this? | 11:31 |
| gthiemonge | Namrata: Hi, did you take a look at the l7 cookbook? https://docs.openstack.org/octavia/latest/user/guides/l7-cookbook.html | 11:36 |
| Namrata | @gthiemonge Yes, using the --redirect-url parameter I can redirect https://abc.com/ to https://efg.com/ but if I want to redirect https://abc.com/foo to https://efg.com/foo it dosen't help | 11:38 |
| gthiemonge | Namrata: in your case, /foo might be any URL? or is it always /foo? | 11:42 |
| Namrata | any url /courses, /dashboard, /about | 11:42 |
| gthiemonge | Namrata: so if I add a REDIRECT_PREFIX rule, for instance: openstack loadbalancer l7policy create --action REDIRECT_PREFIX --redirect-prefix https://www.example.com/ --name policy1 listener1 | 11:56 |
| gthiemonge | and then a rule: openstack loadbalancer l7rule create --compare-type STARTS_WITH --type HOST_NAME --value foo.bar policy1 | 11:56 |
| gthiemonge | the LB redirects any requests to foo.bar/my/path to https://www.example.com/my/path | 11:57 |
| gthiemonge | hmm compare-type should be EQUAL_TO | 12:01 |
| Namrata | @gthiemonge I also figured out REDIRECT_PREFIX, but you say I should set compare-type for every URL /courses, /dashboard, /about | 12:03 |
| gthiemonge | Namrata: you don't need to do it for every URL | 12:07 |
| gthiemonge | https://paste.opendev.org/show/bXSq6Ktq7F6Ra8QdQ3Ub/ | 12:08 |
| gthiemonge | EmilienM: I don't see the same issue with DIB in ussuri, do you have logs on your side? | 12:11 |
| gthiemonge | EmilienM: my bad, there's something wrong with dib in ussuri, but maybe not the same error | 12:13 |
| opendevreview | Gregory Thiemonge proposed openstack/octavia stable/ussuri: Fix amphora build on stable/ussuri https://review.opendev.org/c/openstack/octavia/+/845979 | 12:15 |
| gthiemonge | ^ should fix it too | 12:16 |
| Namrata | @gthiemonge thanks for your help, this works fine. Thanks a lot | 12:19 |
| gthiemonge | Namrata: cool! np | 12:21 |
| *** ysandeep is now known as ysandeep|afk | 14:08 | |
| *** ysandeep|afk is now known as ysandeep|out | 14:57 | |
| prometheanfire | hi, is anyone able to review https://review.opendev.org/c/openstack/octavia/+/845171 it's blocking a requirements update to pyroute2, thanks | 15:28 |
| johnsom | prometheanfire Short answer, we need the taskflow release to land (networkx broke it), then we land a pylint fix, then we can land that pyroute2 patch. | 15:28 |
| johnsom | I think the taskflow thing is +w for upper-constraints, so the dominoes should start falling today. | 15:29 |
| prometheanfire | lol, kk | 15:31 |
| johnsom | Thanks for the priority ping | 15:31 |
| opendevreview | Dr. Jens Harbott proposed openstack/octavia master: Fix pyroute2.netns mock in some tests https://review.opendev.org/c/openstack/octavia/+/845171 | 15:39 |
| opendevreview | Gregory Thiemonge proposed openstack/octavia master: Fix pyroute2.netns mock in some tests https://review.opendev.org/c/openstack/octavia/+/845171 | 15:47 |
| gthiemonge | #startmeeting Octavia | 16:00 |
| opendevmeet | Meeting started Wed Jun 15 16:00:11 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 |
| gthiemonge | #topic Announcements | 16:01 |
| gthiemonge | I don't have any announcements... anyone? johnsom? | 16:01 |
| johnsom | Just that the taskflow fix is almost landed. Upper-constraints is in the gate jobs. | 16:02 |
| johnsom | So we should be able to start landing the gate fixes today | 16:02 |
| tweining | hm, the naming scheme for the next cycle changed, right? I read somthing on the ML I think | 16:02 |
| gthiemonge | johnsom: thanks for working on it ;-) | 16:02 |
| johnsom | Yes | 16:03 |
| tweining | year.sth | 16:03 |
| gthiemonge | yeah it is in "[all][tc] Change OpenStack release naming policy proposal" on the ML | 16:03 |
| johnsom | #link https://governance.openstack.org/tc/resolutions/20220524-release-identification-process.html | 16:04 |
| tweining | k, seems not really new though | 16:04 |
| gthiemonge | yeah but I don't think we mentioned it in a previous meeting | 16:05 |
| johnsom | Basically removes the public vote process for picking a "name". | 16:05 |
| tweining | also I think tick-tock got a new name as well, but I don't remember | 16:05 |
| johnsom | I need to re-read that as I still don't have tick-tock in my brain. | 16:06 |
| gthiemonge | yeah same here | 16:07 |
| gthiemonge | https://review.opendev.org/c/openstack/governance/+/840354/4/resolutions/20220210-release-cadence-adjustment.rst | 16:08 |
| gthiemonge | "SLURP" | 16:08 |
| tweining | right. an awful name if you ask me ;) | 16:08 |
| gthiemonge | haha, I agree | 16:09 |
| johnsom | some of these recent TC docs have been hard to read. The rbac goal is rough reading too | 16:09 |
| gthiemonge | #topic CI Status | 16:11 |
| gthiemonge | johnsom already mentioned the taskflow/pylint issues on master and stable branches | 16:11 |
| gthiemonge | there's also a DIB issue on train and ussuri | 16:11 |
| johnsom | Some of the pylint fixes have one +2 already | 16:11 |
| johnsom | #link https://review.opendev.org/c/openstack/octavia/+/845044 | 16:11 |
| johnsom | #link https://review.opendev.org/c/openstack/octavia/+/845043 | 16:12 |
| gthiemonge | johnsom: yeah I will CR+2/W+1 them when the one on mater is merged/ready | 16:12 |
| johnsom | The stable branches are not impacted by the networkx/taskflow issue | 16:12 |
| gthiemonge | ah ok | 16:13 |
| gthiemonge | thanks for the patches johnsom ;-) | 16:13 |
| gthiemonge | regarding train and ussuri, we're pulling DIB from the master branch and DIB has removed the support for py36 | 16:13 |
| gthiemonge | the workaround/fix is to checkout the latest commit that supports py36 | 16:13 |
| gthiemonge | #link https://review.opendev.org/q/Id763dee0cc048d3d5fe601ce51097581d3c1c4e3 | 16:13 |
| johnsom | DIB is branchless sadly | 16:14 |
| gthiemonge | yeah, old DIB releases have this git issue (permission issue), and new releases don't work with py36 | 16:14 |
| johnsom | There are tags though | 16:14 |
| gthiemonge | yeah we could use tags | 16:15 |
| gthiemonge | let me know in the review if you prefer another solution | 16:15 |
| johnsom | Not sure it matters that much. The SHA is probably fine. | 16:16 |
| gthiemonge | ok | 16:17 |
| gthiemonge | #topic Recheck policy | 16:17 |
| gthiemonge | another CI/gerrit related topic | 16:17 |
| gthiemonge | we were asked to be really careful with "bare" rechecks in gerrit | 16:17 |
| gthiemonge | during the PTG we had a discussion on adding a reason to the recheck messages | 16:18 |
| gthiemonge | ex: recheck octavia-v2-dsvm-scenario story-2009887 | 16:18 |
| gthiemonge | (don't need a story, reason could be: mirror issue, nova timeout, etc...) | 16:18 |
| gthiemonge | in the last 30 days, it seems that many rechecks didn't include a reason | 16:19 |
| gthiemonge | there's an etherpad that gathers all the recurrent CI issues in Octavia: | 16:19 |
| gthiemonge | #link https://etherpad.opendev.org/p/octavia-ci-failures | 16:19 |
| gthiemonge | if we find new "random" issues, we shouldadd them to this etherpad | 16:19 |
| gthiemonge | and don't hesitate, let me know if you don't know why a job has failed | 16:20 |
| tweining | hm, my impression was that there were no recheck without reason since PTG | 16:20 |
| gthiemonge | tweining: they have a script :D | 16:20 |
| gthiemonge | but I need to check the output of the script before sharing the numbers :D | 16:21 |
| johnsom | Yeah | 16:21 |
| gthiemonge | and now... | 16:23 |
| gthiemonge | #topic Octavia Forum at the OpenInfra summit | 16:23 |
| johnsom | Surprise! | 16:23 |
| gthiemonge | none of us (I mean the active contributors here) were at the OpenInfra summit but there was a Operator Forum on Octavia | 16:24 |
| gthiemonge | and there's an etherpad with the notes: | 16:24 |
| gthiemonge | #link https://etherpad.opendev.org/p/oif-summit-berlin-2022-forum-lbaas | 16:24 |
| gthiemonge | quick summary | 16:24 |
| gthiemonge | lot of complaints towards octavia | 16:24 |
| gthiemonge | many non-sense issues | 16:24 |
| johnsom | It wasn't on the official forum list either, so maybe just a session? | 16:25 |
| gthiemonge | (no UDP support, no security between the control plane and the amphora VMs) | 16:25 |
| johnsom | Yeah, there was definitely some mis-information captured there. | 16:25 |
| gthiemonge | yeah, and we were not asked to participate to this session | 16:25 |
| gthiemonge | one good thing: it seesm they have found the link to storyboard | 16:26 |
| johnsom | #link https://wiki.openstack.org/wiki/Forum/Berlin2022 | 16:26 |
| johnsom | I was tracking this page, which it was not on | 16:26 |
| tweining | forum organizers should align those sessions better with the project teams in advance. | 16:26 |
| johnsom | Anyway, I added some after-the-fact comments to the etherpad to try to clarify some of the comments there. | 16:27 |
| gthiemonge | johnsom: thanks | 16:27 |
| gthiemonge | Let's hope those people will at least create stories if they have issues with Octavia | 16:28 |
| johnsom | Yeah, it would be great if we got more operator feedback. | 16:29 |
| johnsom | We have one core reviewer that is an operator, and a few others drop in every once in a while. | 16:29 |
| gthiemonge | #topic Brief progress reports / bugs needing review | 16:32 |
| tweining | well, I could repeat my list from last time. | 16:33 |
| tweining | ;) since then nothing has changed. | 16:33 |
| gthiemonge | I was not really active this last week... I need to check my review backlog | 16:33 |
| tweining | that one would be a good start: | 16:34 |
| tweining | https://review.opendev.org/c/openstack/octavia-lib/+/835049 | 16:34 |
| gthiemonge | tweining: I reviewed your failover threshold patch | 16:34 |
| tweining | cool. thanks | 16:34 |
| gthiemonge | yeah I gave it +1 because we're waiting for this octavia-lib patch | 16:34 |
| gthiemonge | I aalso need to take another look at the notification patch | 16:36 |
| johnsom | FYI, taskflow upper-constraints just merged. I will recheck pylint | 16:36 |
| gthiemonge | Nice! | 16:37 |
| johnsom | Oh, bummer, someone rechecked early, so have to wait.... | 16:37 |
| gthiemonge | :'( | 16:38 |
| tweining | it's good that the whole release process is automated using Gerrit and Zuul, but it can take some time. | 16:38 |
| gthiemonge | #topic Open Discussion | 16:39 |
| gthiemonge | any other topics folks? | 16:39 |
| gthiemonge | nop? | 16:40 |
| gthiemonge | ack | 16:41 |
| tweining | not really | 16:41 |
| gthiemonge | thanks everyone | 16:41 |
| gthiemonge | #endmeeting | 16:41 |
| opendevmeet | Meeting ended Wed Jun 15 16:41:24 2022 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:41 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/octavia/2022/octavia.2022-06-15-16.00.html | 16:41 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/octavia/2022/octavia.2022-06-15-16.00.txt | 16:41 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/octavia/2022/octavia.2022-06-15-16.00.log.html | 16:41 |
| opendevreview | Michael Johnson proposed openstack/octavia master: Fix pyroute2.netns mock in some tests https://review.opendev.org/c/openstack/octavia/+/845171 | 17:39 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!