| opendevreview | OpenStack Proposal Bot proposed openstack/octavia-dashboard master: Imported Translations from Zanata https://review.opendev.org/c/openstack/octavia-dashboard/+/990780 | 03:17 |
|---|---|---|
| opendevreview | Adam Harwell proposed openstack/octavia master: Add ASGI entry point for Octavia API https://review.opendev.org/c/openstack/octavia/+/995230 | 04:09 |
| opendevreview | Adam Harwell proposed openstack/octavia master: Add ASGI entry point for Octavia API https://review.opendev.org/c/openstack/octavia/+/995230 | 06:03 |
| opendevreview | Sergey Kraynev proposed openstack/octavia master: Validate duplicate l7rules https://review.opendev.org/c/openstack/octavia/+/995661 | 10:58 |
| gthiemonge | #startmeeting Octavia | 16:00 |
| opendevmeet | Meeting started Wed Jul 1 16:00:22 2026 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 | o/ | 16:00 |
| viniciusr | o/ | 16:00 |
| raineszm | O/ | 16:00 |
| YushoYamaguchi | o/ | 16:00 |
| gthiemonge | #topic Announcements | 16:02 |
| gthiemonge | * 2026.2 Hibiscus Release Schedule | 16:02 |
| gthiemonge | we are at Hibiscus-2 milestone | 16:02 |
| gthiemonge | nothing special to do here | 16:03 |
| gthiemonge | but just a quick heads-up, the feature freeze will be on Aug 24-28 (time flies) and in case we have some changes for octavia-lib they'll have to merge the week before | 16:03 |
| gthiemonge | that's all for my announcements, anything else folks? | 16:04 |
| gthiemonge | #topic Brief progress reports / bugs needing review | 16:06 |
| gthiemonge | i was on PTO last week, so not much to report | 16:07 |
| gthiemonge | my focus is on reviewing patches | 16:07 |
| YushoYamaguchi | Sorry, this is my first time dealing with this, so I'm not very familiar with the process. Would it be better to discuss the RFE later? | 16:08 |
| gthiemonge | YushoYamaguchi: we can discuss it during the open discussion section in a few minutes | 16:09 |
| YushoYamaguchi | Okay. Thank you! | 16:09 |
| gthiemonge | #topic Open Discussion | 16:11 |
| gthiemonge | YushoYamaguchi: your turn ^^ | 16:11 |
| YushoYamaguchi | Thank you! I want to discuss two RFE. | 16:13 |
| YushoYamaguchi | https://bugs.launchpad.net/octavia/+bug/2158125 | 16:13 |
| YushoYamaguchi | https://bugs.launchpad.net/octavia/+bug/2158637 | 16:13 |
| YushoYamaguchi | If we reach an agreement on either RFE, I'd like to implement it myself. | 16:14 |
| gthiemonge | ack, so for the first one, some folks are working on a feature that allows users to change the Octavia flavor of a load balancer https://review.opendev.org/c/openstack/octavia/+/890215 | 16:15 |
| gthiemonge | it's different, but the result is similar: the nova flavor of an existing LB can be updated | 16:16 |
| gthiemonge | YushoYamaguchi: would it work for you? | 16:18 |
| YushoYamaguchi | I didn't know about this working Merge Request. Thank you. Do you think we can allow to update nova flavor, too? | 16:18 |
| gthiemonge | with this feature, admin creates a new Octavia flavor with the new nova flavor, then users can switch their LB to use the new Octavia flavor | 16:19 |
| gthiemonge | so it doesn't really update the nova flavor, it replaces the octavia flavor | 16:19 |
| YushoYamaguchi | For operational reasons, we'd prefer to provide the same Octavia flavor if possible. | 16:20 |
| gthiemonge | but yeah I see the difference, this feature aims to allow a user to switch for instance from a S or L flavor to an XL flavor, your feature sounds more like a feature for admins | 16:21 |
| YushoYamaguchi | same Octavia flavor with new Nova flavor is what I want. | 16:21 |
| gthiemonge | ack I see | 16:21 |
| YushoYamaguchi | Will it be able to allow? | 16:23 |
| gthiemonge | well, I like the idea of your RFE, I'm going to think about it and add a comment in the launchpad, my concern is that we've never had API calls that override immutable fields (and don't apply the change directly to the amphora), so we need to consider potential side effects or inconsistencies between the DB and the existing amphora instances | 16:24 |
| YushoYamaguchi | Thank you for your comment. | 16:25 |
| gthiemonge | for the 2nd RFE I already added a comment | 16:27 |
| gthiemonge | I feel that we could allow users to perform a failover of an amphora in ERROR status only if the other amphora is not in ERROR status (because that would trigger more mess) | 16:28 |
| YushoYamaguchi | I understand your concern. However, in my product environment, we have no choice but to rewrite DB directory, because there are multiple HyperVisor Type in my cluster. | 16:28 |
| YushoYamaguchi | ↑ for the 1st one. | 16:28 |
| gthiemonge | ack | 16:29 |
| gthiemonge | for this 2nd RFE, I also think that we need to evaluate the impact of the change | 16:30 |
| YushoYamaguchi | <second one> Thank you! I understand that it should not be allowed to failover when the other amphora is ERROR status. | 16:30 |
| YushoYamaguchi | So, should we insert the code to check it? | 16:31 |
| gthiemonge | FYI there's an open bug that triggers bad things when a failover is performed when the other amphora is in ERROR status: https://bugs.launchpad.net/octavia/+bug/2121370 | 16:32 |
| YushoYamaguchi | Thank you. I know this bug because it occur in my cluster multiple times. | 16:33 |
| YushoYamaguchi | (Fixing this bug is one of what I want do but I think the fix is not easy. ) | 16:34 |
| gthiemonge | I think this code would have to be replaced by something new: https://github.com/openstack/octavia/blob/master/octavia/api/v2/controllers/amphora.py#L157 | 16:34 |
| gthiemonge | YushoYamaguchi: (I proposed a patch this week for the VRRP issue, but it still needs to be tested: https://review.opendev.org/c/openstack/octavia/+/995455) | 16:35 |
| YushoYamaguchi | I agree whit you. This is very good news for me. Thank you! | 16:36 |
| YushoYamaguchi | It is okay to wait for merging this VRRP fix and then I try to do https://bugs.launchpad.net/octavia/+bug/2158637. | 16:36 |
| gthiemonge | ok! | 16:37 |
| YushoYamaguchi | Thank you! I'll watch this PR. | 16:38 |
| opendevreview | Alicja Filipek proposed openstack/octavia master: Fix ZooKeeper session churn in ZookeeperTaskFlowDriver https://review.opendev.org/c/openstack/octavia/+/993556 | 16:38 |
| YushoYamaguchi | So please organize about https://bugs.launchpad.net/octavia/+bug/2158125 lastly. Could I continue or discuss about other topic? | 16:40 |
| gthiemonge | yeah you can continue | 16:41 |
| YushoYamaguchi | Thank you. so your concern about in mismatch between old amphora and DB is so serious? | 16:43 |
| YushoYamaguchi | Or, you think we could implement it? | 16:44 |
| gthiemonge | I don't think it's serious, it's just that AFAIK we always try to be consistent, so that would be new, but it's probably acceptable | 16:45 |
| gthiemonge | give me one week, I will put a new comment in the launchpad | 16:45 |
| YushoYamaguchi | Okay, thank you. | 16:46 |
| YushoYamaguchi | After I see it, I'll decide whether or not to implement it. | 16:47 |
| gthiemonge | ok | 16:47 |
| YushoYamaguchi | Thank you so much for listening to me and giving me so much advice! | 16:47 |
| gthiemonge | hey, thank you | 16:48 |
| gthiemonge | any other topics for this week guys? | 16:48 |
| viniciusr | I would like to give a brief update on our bgp-aa deployment in lab at CERN | 16:49 |
| gthiemonge | viniciusr: sure | 16:50 |
| viniciusr | I did few patches involving failover, the newly created amphora were trying to create to use VIP subnet as mgmt ip, and was failing. Another fix was in our deployment we have 2 different netmasks for VIP subnet and vrrp subnet and this was breaking the setup | 16:52 |
| viniciusr | My changes are based in the patchset from December 2025, I was looking to the new patchsets and will rebase upon it, then we can discuss further those scenarios | 16:53 |
| gthiemonge | I think I broke the latest patchchain, I noticed some issues in the tests, so don't focus on it | 16:54 |
| viniciusr | hmmmm ok, In this case I can send my patches anyway with a description of why it was breaking in our deployment | 16:56 |
| viniciusr | this is everything from me, thank you :) | 16:56 |
| gthiemonge | ok thank you viniciusr! | 16:56 |
| gthiemonge | we are at the end of the meeting, I guess we can close it | 16:58 |
| gthiemonge | thank you guys, great discussions! | 16:58 |
| gthiemonge | have a good week! | 16:58 |
| viniciusr | Cheers! Thank you'all | 16:58 |
| YushoYamaguchi | Thank you! | 16:58 |
| gthiemonge | #endmeeting | 16:58 |
| opendevmeet | Meeting ended Wed Jul 1 16:58:34 2026 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:58 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/octavia/2026/octavia.2026-07-01-16.00.html | 16:58 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/octavia/2026/octavia.2026-07-01-16.00.txt | 16:58 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/octavia/2026/octavia.2026-07-01-16.00.log.html | 16:58 |
| raineszm | Have a good one | 16:58 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!