| *** abaindur has quit IRC | 00:52 | |
| *** hongbin has joined #openstack-lbaas | 01:39 | |
| *** kiennt26 has joined #openstack-lbaas | 02:19 | |
| *** yamamoto has joined #openstack-lbaas | 02:42 | |
| *** sapd1 has quit IRC | 03:16 | |
| *** hongbin has quit IRC | 03:43 | |
| *** yamamoto has quit IRC | 03:46 | |
| *** yamamoto has joined #openstack-lbaas | 03:51 | |
| *** kiennt26 has quit IRC | 04:08 | |
| *** ramishra has joined #openstack-lbaas | 04:14 | |
| *** ramishra has quit IRC | 05:02 | |
| *** ramishra has joined #openstack-lbaas | 05:18 | |
| reedipb | cgoncalves: around | 05:31 |
|---|---|---|
| cgoncalves | reedipb, sort of | 05:46 |
| *** yboaron_ has joined #openstack-lbaas | 06:23 | |
| *** numans has quit IRC | 06:39 | |
| *** pcaruana has joined #openstack-lbaas | 06:39 | |
| *** abaindur has joined #openstack-lbaas | 06:47 | |
| *** yamamoto has quit IRC | 06:54 | |
| *** yamamoto has joined #openstack-lbaas | 06:58 | |
| *** velizarx has joined #openstack-lbaas | 07:02 | |
| *** rcernin has quit IRC | 07:02 | |
| *** dims has quit IRC | 07:08 | |
| *** numans_ has joined #openstack-lbaas | 07:09 | |
| *** dims has joined #openstack-lbaas | 07:10 | |
| *** sapd1 has joined #openstack-lbaas | 07:19 | |
| *** abaindur has quit IRC | 07:25 | |
| *** celebdor1 has joined #openstack-lbaas | 07:30 | |
| *** sapd1 has quit IRC | 07:32 | |
| *** sapd1 has joined #openstack-lbaas | 07:33 | |
| *** celebdor1 has quit IRC | 07:33 | |
| *** celebdor has joined #openstack-lbaas | 07:33 | |
| *** yboaron_ has quit IRC | 07:39 | |
| *** yboaron_ has joined #openstack-lbaas | 07:40 | |
| *** ktibi has joined #openstack-lbaas | 07:40 | |
| *** yamamoto has quit IRC | 08:19 | |
| *** yamamoto has joined #openstack-lbaas | 08:20 | |
| *** yboaron_ has quit IRC | 08:29 | |
| *** yboaron_ has joined #openstack-lbaas | 08:29 | |
| *** ktibi has quit IRC | 08:52 | |
| *** yamamoto_ has joined #openstack-lbaas | 08:58 | |
| *** velizarx has quit IRC | 08:58 | |
| *** rcernin has joined #openstack-lbaas | 08:59 | |
| *** yamamoto has quit IRC | 08:59 | |
| *** velizarx has joined #openstack-lbaas | 09:09 | |
| *** takamatsu has joined #openstack-lbaas | 09:15 | |
| sapd1 | How can I recovery a Loadbalancer from ERROR provison_status ? | 09:43 |
| sapd1 | :D | 09:43 |
| *** rcernin has quit IRC | 09:47 | |
| cgoncalves | there's not much to recover from a LB creation failure. delete and create again | 10:10 |
| *** Emine has joined #openstack-lbaas | 10:10 | |
| *** Emine has quit IRC | 10:14 | |
| *** Emine has joined #openstack-lbaas | 10:14 | |
| sapd1 | So, I think we have to check protocol of pool if add a new listener use default_pool_id. | 10:35 |
| *** velizarx has quit IRC | 10:47 | |
| *** sapd1 has quit IRC | 10:51 | |
| *** velizarx has joined #openstack-lbaas | 10:57 | |
| openstackgerrit | Yang JianFeng proposed openstack/octavia master: Add quota support to octavia's l7policy and l7rule https://review.openstack.org/590620 | 11:18 |
| openstackgerrit | Yang JianFeng proposed openstack/python-octaviaclient master: Add l7policy and l7rule to octavia quota https://review.openstack.org/591568 | 11:20 |
| openstackgerrit | Yang JianFeng proposed openstack/octavia master: Add quota support to octavia's l7policy and l7rule https://review.openstack.org/590620 | 11:22 |
| *** celebdor has quit IRC | 11:34 | |
| reedipb | cgoncalves: ping | 11:34 |
| *** yamamoto_ has quit IRC | 11:39 | |
| *** yamamoto has joined #openstack-lbaas | 11:39 | |
| *** velizarx has quit IRC | 11:52 | |
| *** celebdor has joined #openstack-lbaas | 11:53 | |
| reedipb | Hi, I have a question. | 11:55 |
| reedipb | I am confused by https://github.com/openstack/octavia-tempest-plugin/blob/master/octavia_tempest_plugin/tests/api/v2/test_member.py#L795 . I am not sure why there are only 2 members, and not 3. The member-1 is not deleted, so it should remain in the list | 11:56 |
| reedipb | can anyone explain it , please? | 11:56 |
| *** amuller has joined #openstack-lbaas | 12:08 | |
| *** yamamoto has quit IRC | 12:09 | |
| *** velizarx has joined #openstack-lbaas | 12:16 | |
| *** Emine has quit IRC | 12:17 | |
| *** Emine has joined #openstack-lbaas | 12:21 | |
| *** yamamoto has joined #openstack-lbaas | 12:36 | |
| *** yboaron_ has quit IRC | 13:09 | |
| *** yboaron_ has joined #openstack-lbaas | 13:09 | |
| *** Emine has quit IRC | 13:16 | |
| *** numans_ has quit IRC | 13:17 | |
| *** numans has joined #openstack-lbaas | 13:21 | |
| *** velizarx has quit IRC | 13:23 | |
| *** ktibi has joined #openstack-lbaas | 13:25 | |
| *** rtjure has quit IRC | 13:34 | |
| *** rtjure has joined #openstack-lbaas | 13:37 | |
| *** Emine has joined #openstack-lbaas | 13:40 | |
| *** sapd1 has joined #openstack-lbaas | 13:47 | |
| *** hongbin has joined #openstack-lbaas | 14:00 | |
| dayou | reedipb, looking at https://github.com/openstack/octavia-tempest-plugin/blob/master/octavia_tempest_plugin/tests/api/v2/test_member.py#L775-L776, the member list was updated to include only member 2 and member 3 | 14:01 |
| *** amotoki has quit IRC | 14:03 | |
| openstackgerrit | weizj proposed openstack/octavia-dashboard master: Drop python-cinderclient usage https://review.openstack.org/597136 | 14:09 |
| openstackgerrit | weizj proposed openstack/neutron-lbaas-dashboard master: Drop python-cinderclient usage https://review.openstack.org/597139 | 14:10 |
| *** yboaron_ has quit IRC | 14:10 | |
| *** yamamoto has quit IRC | 14:10 | |
| openstackgerrit | German Eichberger proposed openstack/octavia master: Delete zombie amphora when detected https://review.openstack.org/587505 | 14:12 |
| *** yboaron_ has joined #openstack-lbaas | 14:20 | |
| *** pcaruana has quit IRC | 14:29 | |
| *** pcaruana has joined #openstack-lbaas | 14:30 | |
| reedipb | hi dayou , but where does member1 go. I understand that create_member is called for Member 1 and Member 2. So when you batch update, ideally, Member 3 should just be created. How is member 3 updating Member 1? | 14:34 |
| reedipb | dayou : there is no metadata between member1 and member3 which is common, and thats what is confusing me | 14:34 |
| johnsom | reedipb Member 1 is still there, but the sort order has changed there. | 14:36 |
| reedipb | johnsom : hey hi... so https://github.com/openstack/octavia-tempest-plugin/blob/master/octavia_tempest_plugin/tests/api/v2/test_member.py#L785 should have 3 members, right? | 14:37 |
| johnsom | reedipb If you look at that code, it's just creating two, then updating one using the batch update | 14:39 |
| johnsom | reedipb Actually, no, the batch update is deleting one | 14:39 |
| reedipb | johnsom : aaaah .... ok. So it means that when we batch update, we should clear the currently created members, if they are not in the list? | 14:40 |
| reedipb | johnsom: I mean if we have Members m1, m2 , m3 and if we batch update [m3,m4,m5], then should m1 and m2 be deleted and m4 and m5 created? | 14:40 |
| johnsom | reedipb Right, batch update does it all, add, update, delete. https://developer.openstack.org/api-ref/load-balancer/v2/index.html#batch-update-members | 14:41 |
| johnsom | reedipb Yep. It's great for bulk updates of members | 14:41 |
| reedipb | johnsom, ok thanks.. this info was helpful :) | 14:41 |
| *** yamamoto has joined #openstack-lbaas | 14:42 | |
| reedipb | johnsom: I used this with OpenstackClient, this logic is actually useful :) | 14:42 |
| reedipb | johnsom: I wanted to know one more thing.... | 14:42 |
| johnsom | reedipb Big use case is for k8s updates | 14:43 |
| reedipb | johnsom: Most of the test cases in Octavia Tempest plugins are for HTTP protocol. But some provider drivers may not support HTTP ( eg. OVN ). In that case, most of the test cases would fail . | 14:44 |
| *** yamamoto has quit IRC | 14:45 | |
| reedipb | johnsom: Would a solution be useful , if protocol in Octavia Tempest plugin be Conf driven ( default HTTP ) , so that other drivers can also use the same tests to validate their drivers? | 14:45 |
| johnsom | reedipb Ah, yes, I can see that. | 14:45 |
| reedipb | johnsom: one more solution, as discussed earlier with cgoncalves would be to migrate to TCP/UDP tests. Or create test cases for HTTP/TCP/UDP and skip tests if one of the protocol is not supported | 14:46 |
| johnsom | reedipb This is interesting, as we want good coverage. Maybe we should setup tests to iterate over protocols and then have the list of protocols to test in the tempest.conf, defaulted to all. | 14:46 |
| johnsom | reedipb Yeah, similar to your option #2 | 14:47 |
| reedipb | johnsom, yeah, that actually sounds good for the coverage. | 14:48 |
| reedipb | johnsom: if some one is taking this up from the LBaaS team, I can help them out :) | 14:49 |
| johnsom | I think it is a good goal to have the tests configurable for the limits of the various drivers so folks can still run the test suite | 14:50 |
| johnsom | reedipb lol, well, we can ask. It's not something I can take on right now. I have a few other hot issues. | 14:50 |
| reedipb | johnsom: yeah, ok. I can also start the drive, and others can pitch in :) | 14:51 |
| *** pcaruana has quit IRC | 15:00 | |
| *** hongbin has quit IRC | 15:00 | |
| *** yamamoto has joined #openstack-lbaas | 15:15 | |
| *** celebdor has quit IRC | 15:28 | |
| *** ktibi has quit IRC | 15:28 | |
| *** sapd1 has quit IRC | 15:31 | |
| *** sapd1 has joined #openstack-lbaas | 15:43 | |
| *** Emine has quit IRC | 16:02 | |
| *** yboaron_ has quit IRC | 16:34 | |
| *** ramishra has quit IRC | 16:44 | |
| colin- | is there a general timeline for the rocky release somewhere online? feels like there must be but i'm not finding it | 16:47 |
| cgoncalves | colin-, https://releases.openstack.org/rocky/schedule.html | 16:54 |
| cgoncalves | this week is release week! :) | 16:54 |
| colin- | it's always so embarrassing when the URIs are so obvious | 16:54 |
| colin- | ty cgoncalves :) | 16:54 |
| *** reedipb has quit IRC | 17:23 | |
| *** sapd1 has quit IRC | 18:26 | |
| jiteka | thanks xgerman_ and johnsom for your answer about spare pool and amphora VM building time | 19:16 |
| jiteka | 3min is a bit too long yes that's why I'm trying to improve that | 19:16 |
| jiteka | I also read somewhere that there is some alternative base image for amphora smaller than Centos or Ubuntu but I can't recall the name of the distribution | 19:16 |
| xgerman_ | the latest version of Octavia uses ubuntu-minimal which is quite small. Work on using alpine or gentoo hasn't gotten very far through and there isn't nothing out there... | 19:17 |
| johnsom | jiteka Ubuntu is the smallest image we have at the moment at 570MB in qcow2 | 19:22 |
| jiteka | johnsom: that's probably not what I'm using then | 19:24 |
| jiteka | http://paste.openstack.org/show/729005/ | 19:24 |
| jiteka | I generated it using master branch about 12 days ago | 19:24 |
| johnsom | jiteka Maybe it is. There is rounding funnyness with glance. I was just looking here at the nightly build image size: http://tarballs.openstack.org/octavia/test-images/ | 19:25 |
| jiteka | without specific option : /opt/stack/octavia/diskimage-create/./diskimage-create.sh -o amphora-x64-haproxy-master-ubuntu | 19:25 |
| johnsom | I think we still have opportunity to make ubuntu smaller, it's just priorities. | 19:26 |
| johnsom | jiteka Maybe it was this page xgerman_ created but never finished the work: https://wiki.openstack.org/wiki/AmphoraImageOnOtherLinux | 19:29 |
| johnsom | Probably should remove that or put some kind of warning on it | 19:29 |
| xgerman_ | ah, yes, we should delete that | 19:30 |
| jiteka | that's exactly what I read ! | 19:30 |
| johnsom | xgerman_ I don't have permissions to delete it all together, do you? | 19:30 |
| jiteka | :D | 19:30 |
| xgerman_ | though the introduction doesn't really endorse it | 19:30 |
| jiteka | VoyageLinux | 19:31 |
| jiteka | that's the distro I was refering to | 19:32 |
| xgerman_ | ok, I replaced all of it with "/removed" | 19:33 |
| xgerman_ | no, never got that working | 19:33 |
| johnsom | Ok, I also removed the link from the main page | 19:33 |
| xgerman_ | +1 | 19:33 |
| *** amuller has quit IRC | 19:34 | |
| colin- | interested in trying with ubuntu-minimal is using that over the full version discussed in amphora docs? | 20:12 |
| johnsom | colin- No, not really. We just fully switched from ubuntu (full) to ubuntu minimal as the default in Rocky. | 20:31 |
| johnsom | https://github.com/openstack/octavia/commit/f935a51fe4ed5f68a11f2971e62de45691755b71 | 20:31 |
| johnsom | There were three drivers for that: No "full" images for bionic release, the full image kept getting bigger, and the pip conflict with pyyaml required a packaging layout change. | 20:35 |
| colin- | understood | 20:45 |
| nmagnezi | o/ | 20:45 |
| johnsom | o/ | 20:45 |
| *** abaindur has joined #openstack-lbaas | 20:46 | |
| nmagnezi | xgerman_, zombie hunter :D | 20:50 |
| xgerman_ | :-) | 20:50 |
| *** abaindur has quit IRC | 20:55 | |
| *** abaindur has joined #openstack-lbaas | 20:55 | |
| *** andy__ has joined #openstack-lbaas | 21:45 | |
| *** rcernin has joined #openstack-lbaas | 21:46 | |
| *** andy__ is now known as andy_ | 21:46 | |
| *** abaindur has quit IRC | 22:17 | |
| *** abaindur has joined #openstack-lbaas | 22:17 | |
| rm_work | hmm, i DO want to get gentoo working | 23:04 |
| rm_work | i just needed some help on the networking | 23:04 |
| rm_work | so i got stalled | 23:04 |
| johnsom | I don't think gentoo really gets you that better of an image... | 23:39 |
| openstackgerrit | Merged openstack/neutron-lbaas master: Gate API test for the lbaasv2-proxy plugin https://review.openstack.org/539350 | 23:45 |
| *** threestrands has joined #openstack-lbaas | 23:54 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!