| *** rcernin has joined #openstack-lbaas | 00:21 | |
| *** sapd1_x has quit IRC | 00:40 | |
| openstackgerrit | YuehuiLei proposed openstack/octavia master: setup.cfg: Replace dashes with underscores https://review.opendev.org/c/openstack/octavia/+/788124 | 00:45 |
|---|---|---|
| haleyb | rm_work: this is the doc, https://docs.openstack.org/neutron/wallaby/admin/config-service-subnets.html | 00:50 |
| haleyb | and i'm not stuck on it, just came to mind when looking at this | 00:51 |
| haleyb | the original intention was to be able to designate subnets for routers, floating ips, and computes all on the same network, and neutron would choose the right one, allowing operators to not consume public IP space (think dvr routers on every compute) | 00:52 |
| haleyb | it does say it requires admin privileges which i don't remember, but maybe just since it's typically on an external network | 00:53 |
| haleyb | using it would require more work on port creation, and now thinking out loud maybe won't work for octavia | 00:54 |
| rm_work | haleyb: it DOES seem like this is ... exactly the purpose of this feature | 01:09 |
| rm_work | but yeah `Each service type must correspond to a valid device owner within the port model in order for it to be used.` STILL worries me | 01:09 |
| rm_work | why would port creation require more work? | 01:10 |
| rm_work | i assumed just SETTING this value or reading it maybe would require admin? | 01:10 |
| haleyb | i'm just thinking if we have to set the device_owner for it to pick that subnet. did i say it was late here? :) | 01:10 |
| haleyb | and my brain is swapping | 01:10 |
| rm_work | yeah if so, it wouldn't work -- as we need to allow general purpose stuff also | 01:18 |
| rm_work | on the same subnets | 01:18 |
| haleyb | and the name i'm using is hitting the thing you were worried about | 01:19 |
| haleyb | $ openstack subnet create --network e12fb43f-67a2-4f9c-b3b3-b78b04038c8c --service-type octavia --subnet-pool shared-default-subnetpool-v4 lb-subnet | 01:20 |
| haleyb | BadRequestException: 400: Client Error for url: http://172.16.0.155:9696/v2.0/subnets, Invalid input for operation: Failed to create port on network e12fb43f-67a2-4f9c-b3b3-b78b04038c8c, because fixed_ips included invalid subnet 6f97bfb6-983b-4849-a60d-fcb15486b26c. | 01:20 |
| haleyb | super non-obvious error, but removing --service-type octavia helps | 01:20 |
| haleyb | so i guess forget i mentioned it | 01:20 |
| rm_work | ah, damn | 01:22 |
| rm_work | you almost had me excited about doing something like... service-type=octavia:flavorname | 01:23 |
| haleyb | or there's a bug, i tried again with another string and it said the device_owner was invalid | 01:23 |
| rm_work | <_< | 01:24 |
| haleyb | rm_work: yeah, it's only supposed to support network:*, neutron:*, compute:* prefixes | 01:33 |
| rm_work | hmm ok | 01:33 |
| rm_work | tags it is I guess | 01:33 |
| rm_work | unless we... like... register with neutron? by adding code to support lbaas:*? | 01:33 |
| haleyb | i could have sworn i saw other strings there, but maybe it's a test where it mocks out the validator | 01:33 |
| rm_work | which ... would not work in a backwards-compatible way | 01:33 |
| rm_work | so ... yeah :/ tags | 01:33 |
| haleyb | there are the deprecated neutron:LOADBALANCER owners... | 01:34 |
| haleyb | it was worth a noodle | 01:35 |
| haleyb | now you just have to fight over the name of the option, vip_subnet_tag or valid_vip_subnet_tag | 01:36 |
| rm_work | lol | 01:52 |
| rm_work | or `vip_subnet_selection_tag=`? | 01:52 |
| rm_work | lol | 01:52 |
| *** priteau has quit IRC | 02:03 | |
| johnsom | Tags is cleaner imo | 02:05 |
| johnsom | The owner fields are trouble from my experience years ago messing with them. | 02:06 |
| rm_work | so ... varname? | 02:07 |
| rm_work | then i'll start tests and docs | 02:07 |
| johnsom | That is fine with me. I just didn’t want future rm_work to come back and be like “why valid? The is lame” | 02:07 |
| johnsom | The->that | 02:08 |
| rm_work | seriously, not married to it, just picked SOMETHING to get the patch out the door at 3am | 02:08 |
| rm_work | suggestions welcome :D | 02:08 |
| rm_work | what "makes sense" | 02:08 |
| johnsom | `vip_subnet_selection_tag works | 02:09 |
| *** sapd1_x has joined #openstack-lbaas | 02:40 | |
| *** xgerman has quit IRC | 02:52 | |
| *** sapd1_x has quit IRC | 04:26 | |
| *** sapd1_x has joined #openstack-lbaas | 04:34 | |
| *** gcheresh has joined #openstack-lbaas | 04:37 | |
| *** vishalmanchanda has joined #openstack-lbaas | 04:40 | |
| rm_work | i'm kinda at an impasse for docs | 04:48 |
| rm_work | it's hard for me to write about this flavor setting, because we don't talk about any provider specific flavor options *at all*, seemingly on purpose | 04:48 |
| johnsom | Sounds like a new doc page, amphora provider flavors | 05:14 |
| johnsom | My flavors page is intentionally non-provider specific for sure, but we can always create a new doc page | 05:15 |
| *** sapd1_x has quit IRC | 05:39 | |
| rm_work | ok, so not AGAINST having a doc page describing the options for the amphora provider *in the docs* | 05:56 |
| rm_work | rather than just "use provider capabilities list" | 05:57 |
| rm_work | I would love that anyway | 05:57 |
| rm_work | also debating writing a page on "using octavia with routed networks" lol | 05:57 |
| rm_work | but I'd prolly have to be like "here's some cool unmerged patches" XD | 05:57 |
| *** damien_r has joined #openstack-lbaas | 06:06 | |
| *** damien_r has quit IRC | 06:11 | |
| *** luksky has joined #openstack-lbaas | 06:34 | |
| *** QG has joined #openstack-lbaas | 07:06 | |
| openstackgerrit | Gregory Thiemonge proposed openstack/octavia master: Spare pool removal https://review.opendev.org/c/openstack/octavia/+/787722 | 07:15 |
| *** rcernin has quit IRC | 07:25 | |
| *** marlinc has quit IRC | 07:29 | |
| *** rpittau|afk is now known as rpittau | 07:36 | |
| *** marlinc has joined #openstack-lbaas | 07:38 | |
| *** rcernin has joined #openstack-lbaas | 08:21 | |
| zigo | johnsom: Sorry for my stupidity yesterday, the issue is that *I* wasn't using the correct subnet for the VIP of my LB. Otherwise, if creating with the correct subnet, everything works perfectly in our setup ! :) | 08:29 |
| *** rcernin has quit IRC | 08:32 | |
| *** sapd1_x has joined #openstack-lbaas | 08:37 | |
| *** wolsen has quit IRC | 08:41 | |
| *** wolsen has joined #openstack-lbaas | 08:48 | |
| openstackgerrit | Gregory Thiemonge proposed openstack/octavia stable/wallaby: Cap hacking and pylint https://review.opendev.org/c/openstack/octavia/+/788433 | 09:01 |
| openstackgerrit | Merged openstack/octavia master: Fix pylint errors https://review.opendev.org/c/openstack/octavia/+/788047 | 09:16 |
| *** priteau has joined #openstack-lbaas | 09:22 | |
| openstackgerrit | Gregory Thiemonge proposed openstack/octavia stable/victoria: Cap hacking https://review.opendev.org/c/openstack/octavia/+/788458 | 09:24 |
| *** sapd1_x has quit IRC | 09:26 | |
| *** rcernin has joined #openstack-lbaas | 09:39 | |
| *** sapd1_x has joined #openstack-lbaas | 09:49 | |
| openstackgerrit | Gregory Thiemonge proposed openstack/octavia stable/ussuri: Cap hacking https://review.opendev.org/c/openstack/octavia/+/788464 | 09:53 |
| *** vishalmanchanda has quit IRC | 10:00 | |
| *** sapd1_x has quit IRC | 10:06 | |
| *** sapd1_x has joined #openstack-lbaas | 10:06 | |
| *** vishalmanchanda has joined #openstack-lbaas | 10:11 | |
| *** sapd1_x has quit IRC | 11:14 | |
| *** rcernin has quit IRC | 15:01 | |
| openstackgerrit | Merged openstack/octavia-tempest-plugin master: Fix owner_or_admin API tests https://review.opendev.org/c/openstack/octavia-tempest-plugin/+/787177 | 15:10 |
| *** rcernin has joined #openstack-lbaas | 15:12 | |
| *** xgerman has joined #openstack-lbaas | 15:17 | |
| *** rcernin has quit IRC | 15:17 | |
| *** damien_r has joined #openstack-lbaas | 15:41 | |
| *** rcernin has joined #openstack-lbaas | 15:59 | |
| gthiemonge | #startmeeting Octavia | 16:01 |
| openstack | Meeting started Wed Apr 28 16:01:00 2021 UTC and is due to finish in 60 minutes. The chair is gthiemonge. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:01 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:01 |
| *** openstack changes topic to " (Meeting topic: Octavia)" | 16:01 | |
| openstack | The meeting name has been set to 'octavia' | 16:01 |
| gthiemonge | Hi | 16:01 |
| QG | Hello | 16:01 |
| johnsom | o/ | 16:01 |
| haleyb | hi | 16:01 |
| gthiemonge | #topic Announcements | 16:01 |
| *** openstack changes topic to "Announcements (Meeting topic: Octavia)" | 16:01 | |
| gthiemonge | PEP8 issues: new releases of hacking and pylint have broken our pep8 job | 16:02 |
| johnsom | Short agenda this week, nice! | 16:02 |
| gthiemonge | it is now fixed on master | 16:02 |
| gthiemonge | johnsom: we discussed too much during the PTG | 16:02 |
| gthiemonge | But fixes are waiting for reviews on stable branches: | 16:02 |
| gthiemonge | #link https://review.opendev.org/c/openstack/octavia/+/788433 | 16:03 |
| johnsom | Thanks for the quick patch on the pep8 issue | 16:03 |
| gthiemonge | #link https://review.opendev.org/c/openstack/octavia/+/788458 | 16:03 |
| gthiemonge | #link https://review.opendev.org/c/openstack/octavia/+/788464 | 16:03 |
| gthiemonge | older stable branches already have the cap on pylint | 16:03 |
| gthiemonge | I don't have any other announcements for this week. Does anyone have anything? | 16:04 |
| johnsom | Yeah, I agree we should do that | 16:04 |
| johnsom | Thanks to everyone for a good PTG session. I think we had some good discussions. | 16:04 |
| gthiemonge | Yes, that was really interesting | 16:05 |
| johnsom | There is a mailing list discussion about sqlalchemy breakage with the new version. | 16:05 |
| *** rcernin has quit IRC | 16:05 | |
| johnsom | #link http://lists.openstack.org/pipermail/openstack-discuss/2021-April/022094.html | 16:05 |
| johnsom | That may be of interest to folks | 16:06 |
| gthiemonge | johnsom: I launched the unit tests in a env with the new sqlalchemy release, and I didn't notice anything wrong | 16:06 |
| johnsom | Ok cool. I think we also did some of that work already. | 16:06 |
| johnsom | I remember it coming up before | 16:06 |
| gthiemonge | cool | 16:06 |
| gthiemonge | anything else? | 16:07 |
| gthiemonge | ok | 16:07 |
| gthiemonge | #topic Brief progress reports / bugs needing review | 16:07 |
| *** openstack changes topic to "Brief progress reports / bugs needing review (Meeting topic: Octavia)" | 16:07 | |
| gthiemonge | Following our discussions on the CentOS 8/stream jobs during the PTG, I deployed my dev environemnt on CentOS Stream. | 16:08 |
| johnsom | I think I have mostly been doing reviews and minor patches. I did complete the items I signed up for at the PTG and updated the etherpad with my findings. | 16:08 |
| gthiemonge | I patched devstack to not use the current release of RDO (which is still Victoria) but instead I'm using packages from RDO Wallaby. | 16:08 |
| gthiemonge | And everything is working fine so far, so I'm confident that we will be able to fix our CentOS jobs after the RDO release. | 16:09 |
| gthiemonge | johnsom: yeah thanks for the update on the TLS features that are missing in openstacksdk | 16:09 |
| gthiemonge | I also tried to continue the work on amphorav2, I found a couple of bugs when enabling persistence | 16:10 |
| gthiemonge | One in Octavia, the max_workers config is not used when jobboard is enabled: | 16:10 |
| gthiemonge | #link https://review.opendev.org/c/openstack/octavia/+/788307 | 16:10 |
| gthiemonge | The second bug in taskflow, I proposed a patch: | 16:10 |
| gthiemonge | #link https://review.opendev.org/c/openstack/taskflow/+/788328 | 16:11 |
| gthiemonge | Oh and there's a last bug with the failover flow: | 16:11 |
| gthiemonge | #link https://review.opendev.org/c/openstack/octavia/+/787667 | 16:11 |
| gthiemonge | I'm also working on the removal of the spare feature (deprecated since Victoria) | 16:11 |
| gthiemonge | it will help us to avoid some known issues with amphorav2: | 16:12 |
| johnsom | I think it would be nice to get the focal periodic test image build job going for the deployment projects that use the test image: | 16:12 |
| johnsom | #link https://review.opendev.org/c/openstack/octavia/+/763092 | 16:12 |
| gthiemonge | #link https://review.opendev.org/c/openstack/octavia/+/787722 | 16:12 |
| johnsom | We did finally get the xenial image removed | 16:12 |
| gthiemonge | ok, cgoncalves sent a CR-1 for a previous patchset | 16:14 |
| gthiemonge | johnsom: I will review it | 16:14 |
| gthiemonge | #topic Open Discussion | 16:16 |
| *** openstack changes topic to "Open Discussion (Meeting topic: Octavia)" | 16:16 | |
| johnsom | Yeah, I addressed his comment by removing the other changes we needed at the time that was posted. | 16:16 |
| cgoncalves | +2'd | 16:16 |
| gthiemonge | johnsom: ok! | 16:16 |
| gthiemonge | thanks | 16:16 |
| gthiemonge | johnsom: one question for you: during the spare pool removal, I noticed a lot of unused controller tasks in both amphorav1 and amphorav2 | 16:17 |
| gthiemonge | #link http://paste.openstack.org/show/804828/ | 16:17 |
| gthiemonge | https://codesearch.opendev.org/?q=UpdateAmphoraeVIPData&i=nope&files=&excludeFiles=&repos= | 16:17 |
| gthiemonge | https://codesearch.opendev.org/?q=MarkListenerDeletedInDB&i=nope&files=&excludeFiles=&repos= | 16:17 |
| gthiemonge | any reason for that? can we remove them? | 16:17 |
| johnsom | I am sure there are some | 16:18 |
| cgoncalves | are you by chance suggesting we leave cruft code behind? tss-tss... | 16:18 |
| johnsom | Great expectations that never materialized | 16:18 |
| gthiemonge | perhaps there's a plan to use them ;-) | 16:18 |
| johnsom | Well, some of these, there *was* a plan, like four years ago.... | 16:19 |
| johnsom | Can we move that to etherpad? That way I can comment on each one? | 16:19 |
| gthiemonge | the good thing is that we have tests for those classes :D | 16:20 |
| gthiemonge | johnsom: Yeah I will create an etherpad | 16:20 |
| gthiemonge | #link https://etherpad.opendev.org/p/octavia-unused-controller-classes | 16:21 |
| johnsom | Some of this is leftover from changes I made in the failover flow. I had to leave them in the first patches as v2 was still using them. So, yeah, many of these can go. Some were wishful thinking back when we started the project. | 16:21 |
| johnsom | Others make me think we have a bug somewhere.... sigh | 16:22 |
| gthiemonge | haha | 16:22 |
| johnsom | I will add my comments after the meeting | 16:22 |
| gthiemonge | thanks johnsom | 16:23 |
| gthiemonge | any other topics today? | 16:23 |
| gthiemonge | ok | 16:25 |
| gthiemonge | Thanks everyone! | 16:25 |
| gthiemonge | #endmeeting | 16:25 |
| *** openstack changes topic to "Discussions for OpenStack Octavia | Priority bug review list: https://etherpad.openstack.org/p/octavia-priority-reviews" | 16:25 | |
| openstack | Meeting ended Wed Apr 28 16:25:29 2021 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:25 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/octavia/2021/octavia.2021-04-28-16.01.html | 16:25 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/octavia/2021/octavia.2021-04-28-16.01.txt | 16:25 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/octavia/2021/octavia.2021-04-28-16.01.log.html | 16:25 |
| johnsom | Thanks gthiemonge | 16:25 |
| *** QG has quit IRC | 16:31 | |
| johnsom | gthiemonge Is this something we could have as a periodic job? | 16:34 |
| gthiemonge | johnsom: what do you mean? checking unused code? | 16:35 |
| johnsom | Yeah, checking for unused tasks | 16:36 |
| johnsom | Or even just a tox env maybe? | 16:36 |
| gthiemonge | hm, I don't think it's worth it to setup a job | 16:37 |
| gthiemonge | tox env, yeah maybe | 16:37 |
| *** rcernin has joined #openstack-lbaas | 16:38 | |
| *** rpittau is now known as rpittau|afk | 16:39 | |
| *** rcernin has quit IRC | 16:43 | |
| *** rcernin has joined #openstack-lbaas | 16:57 | |
| *** rcernin has quit IRC | 17:02 | |
| johnsom | gthiemonge I have added my comments to the etherpad about the unused tasks. | 17:02 |
| gthiemonge | johnsom: thanks! | 17:30 |
| openstackgerrit | Merged openstack/octavia stable/wallaby: Cap hacking and pylint https://review.opendev.org/c/openstack/octavia/+/788433 | 17:48 |
| openstackgerrit | Merged openstack/octavia stable/ussuri: Cap hacking https://review.opendev.org/c/openstack/octavia/+/788464 | 18:42 |
| openstackgerrit | Merged openstack/octavia stable/victoria: Cap hacking https://review.opendev.org/c/openstack/octavia/+/788458 | 18:42 |
| *** vishalmanchanda has quit IRC | 19:40 | |
| *** njohnston has joined #openstack-lbaas | 20:53 | |
| *** gcheresh has quit IRC | 21:38 | |
| *** rcernin has joined #openstack-lbaas | 21:40 | |
| *** gregraka has quit IRC | 21:59 | |
| *** rcernin has quit IRC | 22:06 | |
| *** rcernin has joined #openstack-lbaas | 22:07 | |
| *** luksky has quit IRC | 22:56 | |
| *** bcafarel has quit IRC | 23:11 | |
| *** bcafarel has joined #openstack-lbaas | 23:11 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!