| *** slaweq has joined #openstack-lbaas | 00:01 | |
| *** fnaval has quit IRC | 00:05 | |
| *** slaweq has quit IRC | 00:06 | |
| *** yamamoto has joined #openstack-lbaas | 00:12 | |
| *** fnaval has joined #openstack-lbaas | 00:17 | |
| *** yamamoto has quit IRC | 00:18 | |
| *** sshank has quit IRC | 00:25 | |
| *** yamamoto has joined #openstack-lbaas | 01:14 | |
| *** yamamoto has quit IRC | 01:20 | |
| *** harlowja has quit IRC | 01:23 | |
| *** slaweq has joined #openstack-lbaas | 02:02 | |
| *** slaweq has quit IRC | 02:06 | |
| *** yamamoto has joined #openstack-lbaas | 02:16 | |
| *** yamamoto has quit IRC | 02:18 | |
| *** yamamoto has joined #openstack-lbaas | 02:18 | |
| *** dlundquist has quit IRC | 02:58 | |
| *** dlundquist has joined #openstack-lbaas | 03:13 | |
| xgerman_ | ^^ fail every test but pep8 and py27; which tells me our unit tests for this patch need help! | 03:33 |
|---|---|---|
| *** slaweq has joined #openstack-lbaas | 04:03 | |
| *** slaweq has quit IRC | 04:08 | |
| *** dayou has quit IRC | 04:45 | |
| *** links has joined #openstack-lbaas | 05:04 | |
| *** dayou has joined #openstack-lbaas | 05:06 | |
| *** links has quit IRC | 05:08 | |
| *** links has joined #openstack-lbaas | 05:13 | |
| *** links has quit IRC | 05:13 | |
| *** fnaval has quit IRC | 05:44 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 05:46 | |
| *** Alexey_Abashkin has joined #openstack-lbaas | 05:47 | |
| *** AlexeyAbashkin has quit IRC | 05:51 | |
| *** Alexey_Abashkin is now known as AlexeyAbashkin | 05:51 | |
| *** AlexeyAbashkin has quit IRC | 05:55 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 05:55 | |
| *** AlexeyAbashkin has quit IRC | 06:02 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 06:03 | |
| *** pcaruana has joined #openstack-lbaas | 06:33 | |
| *** pcaruana has quit IRC | 06:50 | |
| *** slaweq has joined #openstack-lbaas | 06:52 | |
| *** pcaruana has joined #openstack-lbaas | 07:06 | |
| *** diablo_rojo has joined #openstack-lbaas | 07:08 | |
| *** rcernin has quit IRC | 07:17 | |
| *** tesseract has joined #openstack-lbaas | 07:27 | |
| *** tesseract has quit IRC | 07:53 | |
| *** tesseract has joined #openstack-lbaas | 07:56 | |
| *** pcaruana has quit IRC | 08:56 | |
| *** fnaval has joined #openstack-lbaas | 09:06 | |
| *** pcaruana has joined #openstack-lbaas | 09:07 | |
| *** fnaval has quit IRC | 09:11 | |
| *** annp has quit IRC | 09:44 | |
| *** diablo_rojo has quit IRC | 09:51 | |
| *** rpittau has quit IRC | 10:08 | |
| *** AlexeyAbashkin has quit IRC | 10:11 | |
| *** l4yerffej has quit IRC | 10:15 | |
| *** l4yerffej has joined #openstack-lbaas | 10:17 | |
| *** rpittau has joined #openstack-lbaas | 10:23 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 10:46 | |
| *** AlexeyAbashkin has quit IRC | 10:51 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 10:52 | |
| mugsie | can I upload a cert via octavia, and have it store it in what ever store it is using? | 10:53 |
| cgoncalves | mugsie, no. octavia doesn't offer an API to upload certs. you can either use barbican or castellan to store and pass in to octavia a href or use the local cert manager (highly not recommended) by placing the certs on the controller disk | 10:59 |
| mugsie | damn, thats what I thought | 10:59 |
| cgoncalves | I have never tested the local cert manager, though | 10:59 |
| mugsie | I saw https://github.com/openstack/octavia/blob/30bafb11d7eff23d826b39534e31acc249bfa1f3/octavia/certificates/manager/barbican.py#L48 and wondered if there was an API | 10:59 |
| cgoncalves | hehe me too a few days ago :) | 11:00 |
| mugsie | :) | 11:02 |
| *** fnaval has joined #openstack-lbaas | 11:06 | |
| *** fnaval has quit IRC | 11:11 | |
| *** atoth has quit IRC | 12:29 | |
| *** atoth has joined #openstack-lbaas | 12:29 | |
| *** yamamoto has quit IRC | 12:33 | |
| *** fnaval has joined #openstack-lbaas | 13:06 | |
| *** fnaval has quit IRC | 13:10 | |
| *** yamamoto has joined #openstack-lbaas | 13:14 | |
| *** yamamoto has quit IRC | 13:14 | |
| *** samccann has joined #openstack-lbaas | 13:18 | |
| *** yamamoto has joined #openstack-lbaas | 13:18 | |
| *** fnaval has joined #openstack-lbaas | 14:06 | |
| *** ivve has quit IRC | 14:08 | |
| *** fnaval has quit IRC | 14:11 | |
| johnsom | mugsie https://docs.openstack.org/octavia/latest/user/guides/basic-cookbook.html#deploy-a-tls-terminated-https-load-balancer | 14:19 |
| mugsie | yeah, but that relys on octavia using barbican right? | 14:20 |
| xgerman_ | or catellan which *could* go to vault | 14:20 |
| mugsie | so if octavia was setup to use castellan + vault, it wouldn't work? | 14:20 |
| johnsom | Or castellan yeah | 14:20 |
| xgerman_ | yep, that’s what they told us | 14:21 |
| johnsom | mugsie Yeah, castellan + vault works. | 14:21 |
| xgerman_ | +1 | 14:21 |
| mugsie | but not with those steps, becuase the upload would have been in barbican? | 14:21 |
| johnsom | Queens forward | 14:21 |
| xgerman_ | mugsie: our docs need help | 14:22 |
| mugsie | oh, is openstack secret store castellen now? | 14:22 |
| mugsie | xgerman_: don't they all :P ? | 14:22 |
| *** fnaval has joined #openstack-lbaas | 14:22 | |
| johnsom | Yeah, I think you are right. Those steps assume barbican. Not sure if the same CLI works for castellan or not | 14:22 |
| mugsie | cool - I was just trying to bring in let encrpyt stuff, so I can stick to barbican for now | 14:23 |
| johnsom | rm_work would know, but it's a bit early for him to be on yet | 14:23 |
| xgerman_ | I doubt anyone of us actually ran that end-to-end | 14:23 |
| mugsie | https://pypi.org/project/certbot-dns-openstack/ | 14:23 |
| xgerman_ | in a test | 14:23 |
| *** ivve has joined #openstack-lbaas | 14:23 | |
| johnsom | mugsie Nice docs | 14:23 |
| johnsom | lol | 14:24 |
| mugsie | so you can get a cert using designate to authenticate you own the domain, I wanted to have a way of pushing it to octavia to use in an LB | 14:24 |
| johnsom | Please fill here a long description which must be at least 3 lines wrapped on 80 cols, so that distribution package maintainers can use it in their packages. Note that this is a hard requirement. | 14:24 |
| mugsie | johnsom: sshhhh | 14:24 |
| mugsie | it is very new | 14:24 |
| xgerman_ | yeah, pushing it into Octavia would be slick | 14:24 |
| johnsom | Ah, neat | 14:24 |
| johnsom | We hope to have the ACL thing fixed soon-ish. We just need someone to do the API work to make that step go away | 14:25 |
| mugsie | :) | 14:25 |
| mugsie | "just need someone" the refrain of all PTLs ever | 14:25 |
| johnsom | Sigh, yes | 14:25 |
| johnsom | You going to the summit? | 14:25 |
| mugsie | yup, just booked flights | 14:26 |
| johnsom | Cool, see you there | 14:26 |
| mugsie | cool. I am heading down to SEA for a bit after as well | 14:27 |
| mugsie | xgerman_: you joining? | 14:27 |
| johnsom | cgoncalves was looking for a project, maybe he has cycles to fix the listener API to get rid of the ACL step.... | 14:27 |
| xgerman_ | yeah, I am goung to the summit | 14:27 |
| xgerman_ | alreday booked flights otherwise SEA would be fun ;-) | 14:28 |
| mugsie | going to drop by cannon again :D | 14:29 |
| johnsom | Ahh, cannon | 14:31 |
| *** yamamoto has quit IRC | 14:31 | |
| johnsom | Sometime we need to be in PDX and make a reservation at the Multnomah whiskey library | 14:32 |
| johnsom | And contrary to negative xgerman, I think our docs aren’t too bad. Still work to do of course, but not in too bad of shape. | 14:34 |
| xgerman_ | I was referring to the use castellan with vault and Octavia doc | 14:34 |
| xgerman_ | but nothing as bad as certbot-dns-openstack | 14:35 |
| mugsie | xgerman_: patches welcome :P | 14:35 |
| *** atoth has quit IRC | 14:36 | |
| *** AlexeyAbashkin has quit IRC | 14:38 | |
| *** yamamoto has joined #openstack-lbaas | 14:46 | |
| *** atoth has joined #openstack-lbaas | 14:51 | |
| *** rpittau has quit IRC | 14:52 | |
| *** yamamoto has quit IRC | 14:58 | |
| *** yamamoto has joined #openstack-lbaas | 15:07 | |
| *** pcaruana has quit IRC | 15:11 | |
| *** yamamoto has quit IRC | 15:12 | |
| *** yamamoto has joined #openstack-lbaas | 15:17 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 15:29 | |
| *** yamamoto has quit IRC | 15:35 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Creates provider driver base class and exceptions https://review.openstack.org/558013 | 16:19 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Create noop provider driver and data model https://review.openstack.org/558320 | 16:19 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Implement provider drivers https://review.openstack.org/563795 | 16:19 |
| johnsom | hmmm | 16:19 |
| johnsom | Ok, that looks like a good checkpoint / rebase. Everything in the load balancer API is switched over to using a driver except for single-call-create. Still have a bit of work to do there. Removed the reference to "handler" from the load balancer file, so that is progress. | 16:33 |
| *** yamamoto has joined #openstack-lbaas | 16:36 | |
| *** yamamoto has quit IRC | 16:42 | |
| *** aojea has joined #openstack-lbaas | 16:46 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Implement provider drivers https://review.openstack.org/563795 | 16:50 |
| johnsom | Somehow missed a flavor->flavor_id in the rebase | 16:51 |
| *** aojea has quit IRC | 16:54 | |
| *** tesseract has quit IRC | 17:09 | |
| openstackgerrit | German Eichberger proposed openstack/octavia master: Ignore a port not found when deleting an LB https://review.openstack.org/564848 | 17:28 |
| johnsom | xgerman_ You are telling me that a port we got from neutron seven lines of code above disappeared in those seven lines of code??????? | 17:31 |
| xgerman_ | yes, pretty sure my colleagues purge script is wacky | 17:31 |
| xgerman_ | but I don’t see a reason not to ignore that Exception | 17:31 |
| johnsom | It's just a bit bizarre as there is literally a check for that port existing or not a few lines before your change..... | 17:32 |
| johnsom | That was the whole point of that change. | 17:32 |
| xgerman_ | yes, what can I say — saw that in the wild… as I said either my colleagues are purging too aggressively or neutron’s wheels are spinning weirdly | 17:33 |
| johnsom | Well we *know* that environment has a broken neutron | 17:34 |
| xgerman_ | yep — | 17:34 |
| johnsom | yeah, so, here is the big question, why is something else deleting ports owned by Octavia? | 17:37 |
| *** yamamoto has joined #openstack-lbaas | 17:38 | |
| xgerman_ | neutron purge will do it | 17:38 |
| xgerman_ | the VIP port is in the user tenant’s network so they can mess with it | 17:39 |
| xgerman_ | should they? and should we be resilient to their messing? | 17:40 |
| *** AlexeyAbashkin has quit IRC | 17:43 | |
| *** yamamoto has quit IRC | 17:43 | |
| openstackgerrit | German Eichberger proposed openstack/octavia master: Ignore a port not found when deleting an LB https://review.openstack.org/564848 | 17:44 |
| openstackgerrit | German Eichberger proposed openstack/octavia master: Ignore a port not found when deleting an LB https://review.openstack.org/564848 | 17:45 |
| openstackgerrit | German Eichberger proposed openstack/octavia master: Allow fields filter on single object GETs https://review.openstack.org/559842 | 18:06 |
| *** yamamoto has joined #openstack-lbaas | 18:40 | |
| *** yamamoto has quit IRC | 18:45 | |
| *** imacdonn has quit IRC | 19:14 | |
| *** imacdonn has joined #openstack-lbaas | 19:15 | |
| rm_work | mugsie: the guide contains instructions for barbican, but if octavia is configured to use castellan+vault, then you can use whatever method you want to get them in vault, and then just pass the right vault id to octavia | 19:21 |
| rm_work | we don't document that because ... there's a ton of ways to get things in vault and i have no idea how people are doing it | 19:22 |
| rm_work | to be fair, when I say "it works", I mean "it SHOULD work", because i don't have a vault install to test it, i'm just relying on knowing that the castellan calls return properly | 19:23 |
| *** slaweq has quit IRC | 19:35 | |
| *** yamamoto has joined #openstack-lbaas | 19:42 | |
| *** yamamoto has quit IRC | 19:47 | |
| *** velizarx has joined #openstack-lbaas | 19:48 | |
| *** samccann has quit IRC | 19:49 | |
| *** samccann has joined #openstack-lbaas | 19:49 | |
| *** velizarx has quit IRC | 20:02 | |
| *** mlavalle has joined #openstack-lbaas | 20:07 | |
| mlavalle | johnsom: ping | 20:07 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Fix amp failover where failover already failed https://review.openstack.org/548989 | 20:29 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Experimental multi-az support https://review.openstack.org/558962 | 20:29 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: WIP: AZ Evacuation resource https://review.openstack.org/559873 | 20:29 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: WIP: Floating IP Network Driver (spans L3s) https://review.openstack.org/435612 | 20:29 |
| openstackgerrit | Thomas Morin proposed openstack/neutron-lbaas master: remove unused plugin.get_plugin_name() https://review.openstack.org/557665 | 20:30 |
| johnsom | mlavalle Pong! | 20:37 |
| mlavalle | johnsom: ready for the weekend | 20:37 |
| mlavalle | ? | 20:37 |
| johnsom | A wet weekend sadly. Beautiful week here, but pouring for the weekend. | 20:38 |
| johnsom | All good though. Probably can get out of yard work | 20:38 |
| johnsom | You? | 20:38 |
| mlavalle | it's going to be a beautiful weekend in this part of Texas. We are driving to Houston to attend a birthday party | 20:41 |
| mlavalle | so it should be a lot of fun | 20:41 |
| johnsom | Nice! | 20:41 |
| johnsom | What can I help you with? | 20:41 |
| mlavalle | I just want to see if we can get some Octavia eyeballs an this: https://review.openstack.org/#/q/status:open+project:openstack/octavia+branch:master+topic:udp-support | 20:42 |
| mlavalle | on this^^^ | 20:42 |
| mlavalle | no rush | 20:42 |
| mlavalle | but whenever you guys have some time | 20:42 |
| johnsom | Ah, yeah, I was wondering if that was good to go or not. Good to know it's ready for review. | 20:42 |
| johnsom | We have been pretty swamped recently. Gate issues, etc. | 20:43 |
| mlavalle | Thanks :-) | 20:43 |
| *** yamamoto has joined #openstack-lbaas | 20:43 | |
| johnsom | Sure, NP | 20:43 |
| mlavalle | enjoy the wet weekend. maybe go to the movies | 20:44 |
| johnsom | mlavalle ZhaoBo can also pop a message in IRC that he would like a review. We will see it even if we are sleeping | 20:44 |
| *** velizarx has joined #openstack-lbaas | 20:44 | |
| mlavalle | johnsom: that's true. I'll pass the message along | 20:45 |
| johnsom | +! | 20:45 |
| johnsom | +1 | 20:45 |
| *** samccann has quit IRC | 20:46 | |
| mlavalle | johnsom: I just left him a message in WeChat. He will get it when he wakes up | 20:49 |
| johnsom | Sounds good. I will try to load it up early next week | 20:49 |
| *** yamamoto has quit IRC | 20:49 | |
| mlavalle | that's good, Thanks | 20:49 |
| *** velizarx has quit IRC | 20:58 | |
| *** harlowja has joined #openstack-lbaas | 21:07 | |
| *** yamamoto has joined #openstack-lbaas | 21:46 | |
| *** yamamoto has quit IRC | 21:52 | |
| *** fnaval has quit IRC | 22:08 | |
| *** yamamoto has joined #openstack-lbaas | 22:47 | |
| *** yamamoto has quit IRC | 22:53 | |
| rm_work | johnsom: wait, it spins up two machines as webservers? I thought we just did one | 23:15 |
| rm_work | and two ports | 23:16 |
| johnsom | No, two webservers, can't test round robin without it | 23:16 |
| johnsom | They are cirros though | 23:16 |
| rm_work | err | 23:16 |
| rm_work | can't you? | 23:16 |
| rm_work | <_< | 23:16 |
| rm_work | also: cirros for you... centos for me and they take 6 min each to be ready | 23:16 |
| johnsom | I think all versions of that plugin had two | 23:16 |
| rm_work | no | 23:16 |
| rm_work | the last version that we reverted made one | 23:17 |
| rm_work | and used two ports | 23:17 |
| rm_work | which I SWEAR this one did too, or I wouldn't have +2'd it | 23:17 |
| johnsom | Pretty sure it create two each with two ports | 23:17 |
| rm_work | I remember having to tweak the SG to allow both ports, too | 23:17 |
| rm_work | eugh WHY | 23:17 |
| rm_work | you can add two members, and test round-robin with them, with one webserver IP and two ports | 23:18 |
| johnsom | Yeah, you just don't test plugging multiple member networks into the amps | 23:23 |
| rm_work | T_T k | 23:25 |
| rm_work | https://review.openstack.org/#/c/563737/ | 23:25 |
| rm_work | should I change the commit message? | 23:25 |
| rm_work | it became not WIP by virtue of those merging | 23:25 |
| rm_work | but we need to merge this | 23:25 |
| rm_work | johnsom / xgerman_ / nmagnezi / dayou | 23:25 |
| johnsom | commit message on what? | 23:26 |
| rm_work | ^^ | 23:26 |
| johnsom | Oh, yes please | 23:26 |
| openstackgerrit | Adam Harwell proposed openstack/octavia-tempest-plugin master: Fix tests for Octavia API compat changes https://review.openstack.org/563737 | 23:27 |
| johnsom | I'm trying to track down why my single call create LBs don't seem to have health monitors | 23:27 |
| rm_work | needs +2 again then | 23:27 |
| rm_work | johnsom: eugh ok well | 23:27 |
| rm_work | let me know if you need help with that | 23:27 |
| rm_work | I was just toying with adding single-create back to the top of my worklist | 23:27 |
| rm_work | i'm in testing / validating mode | 23:27 |
| johnsom | Nah, just working through it. I'm pretty sure It's not my code at this point. | 23:27 |
| rm_work | though ugh, i'm also running like 8 patches ahead of master | 23:27 |
| rm_work | my patch list is not short :( | 23:28 |
| rm_work | johnsom: right so you think that, upstream master, probably if you single-create with a HM it won't be created? | 23:33 |
| rm_work | it's possible | 23:34 |
| rm_work | i can do some testing | 23:34 |
| johnsom | Yeah, maybe, tracking it down now. LB with pool, with HM | 23:34 |
| rm_work | umm | 23:35 |
| rm_work | is it health_monitor or healthmonitor | 23:35 |
| rm_work | yeah ok | 23:36 |
| rm_work | johnsom: i found it | 23:36 |
| rm_work | bugbugbug | 23:36 |
| johnsom | ha | 23:36 |
| johnsom | So, good to know it's not just me | 23:36 |
| johnsom | Yeah, confirmed it is somewhere in this graph mess | 23:36 |
| johnsom | The db_pools that comes out drops the HMs | 23:37 |
| johnsom | If I piss the spousal unit off and work late, I might be able to finish LB today. All I have is some of single call create left to do. | 23:38 |
| rm_work | yes | 23:38 |
| rm_work | i'm fixing it now | 23:38 |
| rm_work | patch in a moment | 23:38 |
| johnsom | Cool | 23:38 |
| rm_work | oh wait no, wat | 23:38 |
| rm_work | OH | 23:39 |
| rm_work | lol sec | 23:39 |
| rm_work | i HATE that we alternate seemingly randomly on which bits use an underscore and which dont | 23:39 |
| johnsom | Yeah, agreed. I have to "fix" that in this provider patch. | 23:40 |
| johnsom | I'm in brute-force mode now, then we can come back and fix things. | 23:40 |
| johnsom | Pretty sure you advocated for no underscore which is why the provider spec doesn't have one | 23:41 |
| johnsom | In fact, once I'm done here, we will have a nice file that will highlight all of the stupid naming stuff we did.... lol | 23:41 |
| rm_work | yes | 23:42 |
| rm_work | i actually have started | 23:42 |
| rm_work | a couple of times | 23:42 |
| rm_work | to literally go through and fix EVERYTHING | 23:42 |
| rm_work | even in the DB | 23:42 |
| johnsom | Please don't now.... | 23:42 |
| rm_work | I was mostly done with the migration file | 23:42 |
| johnsom | Wait until after this merges | 23:42 |
| rm_work | lol yeah i'll wait | 23:42 |
| rm_work | was going for zero instances of health_monitor or load_balancer | 23:43 |
| rm_work | the DB is "fun" | 23:43 |
| rm_work | lots of temporarily removing / reinstating constraints | 23:43 |
| rm_work | around alter-tables | 23:43 |
| rm_work | but yeah i am seeing something wrong with HM creation still | 23:43 |
| rm_work | it just isn't as as I thought | 23:44 |
| rm_work | *as easy | 23:44 |
| rm_work | johnsom: it may actually be creating, but not *returning* right | 23:45 |
| rm_work | if you do a HM list | 23:45 |
| rm_work | it may *be there* | 23:45 |
| johnsom | Ah yeah, maybe. | 23:46 |
| johnsom | brb | 23:47 |
| *** yamamoto has joined #openstack-lbaas | 23:49 | |
| rm_work | uhhh wtf | 23:52 |
| rm_work | i'm debugging functional tests for this | 23:52 |
| rm_work | and if I query like... self.repositories.pool.get_all(session) | 23:53 |
| rm_work | the first time it returns the pool | 23:53 |
| rm_work | and then if i run it again, it returns an empty list | 23:53 |
| rm_work | are we using mocks in the functional tests for the db or something? lol | 23:53 |
| rm_work | i don't think so :P | 23:53 |
| *** yamamoto has quit IRC | 23:54 | |
| johnsom | Back. I think we do in some tests to inject failures | 23:55 |
| rm_work | hmmmm | 23:56 |
| rm_work | this is really wonky | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!