Wednesday, 2018-06-13

*** SumitNaiksatam has quit IRC00:05
*** yamamoto has joined #openstack-lbaas00:06
*** Swami has quit IRC00:09
*** yamamoto has quit IRC00:10
*** longkb has joined #openstack-lbaas00:36
*** amuller has joined #openstack-lbaas00:39
*** amuller has quit IRC00:51
rm_workjohnsom: you ever consider making an amp that has varnish or squid in it too? :P could patch them right into the http flow ^_^01:06
rm_worki mean, it'd be horribly orthogonal to LBaaS, but hilariously simple01:06
johnsomVarnish is slow, squid has protocol compliance issues.  I like Apache traffic server myself.01:07
rm_workok well, that then :P01:07
rm_workwait varnish is *slow*? :P01:08
johnsomYes, I go back and forth on if it should be in or its own service/vm01:08
johnsomYou want more storage, more ram, and want to bill different for it01:08
johnsomEating dinner, catch you in a bit01:09
rm_workyeah, service chaining prolly01:18
rm_workwould be better01:18
rm_workbut like01:18
rm_workno one is going to make Cache as a Service01:19
*** hongbin has joined #openstack-lbaas01:23
johnsomWell, my intent with the octavia worker was to make other aaS easy to make.01:37
johnsomFor the cache I don’t think we would need to use sfc, I think we could do it without.01:38
johnsomBut the other option is to integrate it, which would be reasonably straight forward01:38
johnsomIt is on our roadmap btw01:39
johnsomSo maybe Stein or T01:39
openstackgerritMerged openstack/octavia-tempest-plugin master: Fix a potential IP address conflict
johnsomAnd yes, varnish is slow. Slower than nginx even01:40
*** SumitNaiksatam has joined #openstack-lbaas01:41
rm_workyeah, it's really not difficult i think01:46
*** kiennt26 has joined #openstack-lbaas01:47
*** kiennt26 has quit IRC02:31
*** annp has joined #openstack-lbaas02:40
*** SumitNaiksatam_ has joined #openstack-lbaas03:22
*** SumitNaiksatam has quit IRC03:25
*** SumitNaiksatam_ is now known as SumitNaiksatam03:25
*** hongbin has quit IRC04:13
*** links has joined #openstack-lbaas04:24
openstackgerritMerged openstack/octavia master: Allow using spaces for primary common name in SSL certificates
*** pcaruana has joined #openstack-lbaas04:50
*** yamamoto has joined #openstack-lbaas05:01
*** pcaruana has quit IRC05:18
*** nmanos has joined #openstack-lbaas05:26
*** nmanos has quit IRC05:32
*** nmanos has joined #openstack-lbaas05:51
*** pcaruana has joined #openstack-lbaas06:06
*** AlexeyAbashkin has joined #openstack-lbaas06:54
*** nmanos has quit IRC06:59
*** nmanos has joined #openstack-lbaas06:59
*** rcernin has quit IRC07:00
*** tesseract has joined #openstack-lbaas07:07
*** kobis has joined #openstack-lbaas07:12
*** kobis has quit IRC07:15
*** threestrands has quit IRC07:21
*** kobis has joined #openstack-lbaas07:35
*** kobis has quit IRC07:37
*** kobis has joined #openstack-lbaas07:40
openstackgerritsunqingliang proposed openstack/octavia master: replace overriden with overridden
*** AlexeyAbashkin has quit IRC08:31
*** AlexeyAbashkin has joined #openstack-lbaas08:40
openstackgerritShangXiao proposed openstack/octavia-tempest-plugin master: Update links in README.rst
*** links has quit IRC09:08
*** salmankhan has joined #openstack-lbaas09:22
*** links has joined #openstack-lbaas09:25
*** kobis has quit IRC09:35
*** kobis has joined #openstack-lbaas09:59
*** salmankhan has quit IRC10:06
*** AlexeyAbashkin has quit IRC10:06
*** AlexeyAbashkin has joined #openstack-lbaas10:06
*** salmankhan has joined #openstack-lbaas10:07
*** AlexeyAbashkin has quit IRC10:11
*** links has quit IRC10:12
*** yamamoto_ has joined #openstack-lbaas10:18
*** yamamoto has quit IRC10:22
*** links has joined #openstack-lbaas10:25
*** longkb has quit IRC10:37
*** AlexeyAbashkin has joined #openstack-lbaas11:05
*** atoth has joined #openstack-lbaas11:33
*** annp has quit IRC11:45
openstackgerritMerged openstack/neutron-lbaas master: Allow to create loadbalancer with chinese character
*** yamamoto_ has quit IRC12:22
*** Alexey_Abashkin has joined #openstack-lbaas12:24
*** AlexeyAbashkin has quit IRC12:26
*** AlexeyAbashkin has joined #openstack-lbaas12:27
*** Alexey_Abashkin has quit IRC12:29
*** amuller has joined #openstack-lbaas12:30
*** fnaval has joined #openstack-lbaas12:33
*** aojea_ has joined #openstack-lbaas12:36
*** AlexeyAbashkin has quit IRC12:54
*** AlexeyAbashkin has joined #openstack-lbaas13:04
*** yamamoto has joined #openstack-lbaas13:20
*** irenab has quit IRC13:22
*** irenab has joined #openstack-lbaas13:23
*** yamamoto_ has joined #openstack-lbaas13:24
*** yamamoto has quit IRC13:28
*** yamamoto_ has quit IRC13:30
*** yamamoto has joined #openstack-lbaas13:32
*** AlexeyAbashkin has quit IRC13:33
*** irenab has quit IRC13:40
*** irenab has joined #openstack-lbaas13:41
*** aojea_ has quit IRC13:48
*** aojea_ has joined #openstack-lbaas13:52
*** aojea_ has quit IRC13:52
*** aojea_ has joined #openstack-lbaas13:52
*** kobis has quit IRC14:08
*** aojea_ has quit IRC14:15
*** yamamoto_ has joined #openstack-lbaas14:53
*** yamamoto_ has quit IRC14:54
*** yamamoto has quit IRC14:57
*** yamamoto has joined #openstack-lbaas15:03
*** yamamoto has quit IRC15:08
*** yamamoto has joined #openstack-lbaas15:18
*** AlexeyAbashkin has joined #openstack-lbaas15:18
*** yamamoto has quit IRC15:23
*** kobis has joined #openstack-lbaas15:27
*** AlexeyAbashkin has quit IRC15:30
*** ktibi has joined #openstack-lbaas15:32
ktibiHi, Can I use multiple provider for LBAAS with same agent ? (v2 without octavia)15:32
ktibior I need to have one provider by agent ?15:32
johnsomktibi Hi, FYI LBAAS is not part of the neutron project any longer.15:33
johnsomYes, you can have multiple providers (drivers) enabled at the same time for both neutron-lbaas and octavia deployments.15:33
*** yamamoto has joined #openstack-lbaas15:33
johnsomThe neutron-lbaas agent is only used for the old haproxy/namespace driver. I think all of the other drivers do not use that agent.15:34
ktibijohnsom, ok I need to deploy one agent by provider or can I use one agent with configuration like : service_provider : haproxy,F5,...15:34
ktibijohnsom, I try to use F5 LBAAS but I want to use haproxy too15:34
johnsomktibi Yes, it is a comma separated list for the service_provider setting.15:35
johnsomThen on the API you specify a "provider" to select the driver.15:36
johnsomwhen you create  the LB15:36
ktibijohnsom, ok thx15:36
*** links has quit IRC15:36
johnsomktibi Oh, I might be wrong on the comma separated thing. It may be that you just add additional service_provider lines.15:37
ktibijohnsom, no I can't have two times line with the LBAAS attribut15:38
*** yamamoto has quit IRC15:38
ktibiwe tried too : service_provider = LOADBALANCERV2:Haproxy:neutron_lbaas.drivers.haproxy.plugin_driver.HaproxyOnHostPluginDriver:default,LOADBALANCERV2:F5Networks:neutron_lbaas.drivers.f5.driver_v2.F5LBaaSV2Driver15:38
johnsomOk. I haven't run neutron-lbaas in a long time15:38
ktibiI can't find an example of configuration with a thing like that :/15:39
johnsomAnd this form didn't work either?15:39
ktibijohnsom, the error with two lines : ERROR neutron ValueError: Multiple plugins for service LOADBALANCERV2 were configured15:40
johnsomktibi That error might be a problem with the neutron.conf "service_plugins" section and not the "service_providers" section15:41
ktibiI'll check thx15:42
*** AlexeyAbashkin has joined #openstack-lbaas15:42
johnsomYeah, can you paste use the neutron.conf {DEFAULT] service_plugins line15:42
johnsomMine looks like this:15:43
ktibijohnsom, ok yes we have lbaasv2 and
ktibiI guess it's the same thing15:44
johnsomktibi Yeah, bingo.  Only lbaasv2 should be there15:44
*** aojea has joined #openstack-lbaas15:51
*** ktibi_ has joined #openstack-lbaas15:58
*** ktibi has quit IRC16:01
*** yamamoto has joined #openstack-lbaas16:03
*** yamamoto has quit IRC16:08
*** SumitNaiksatam has quit IRC16:18
*** yamamoto has joined #openstack-lbaas16:18
*** yamamoto has quit IRC16:23
*** kobis has quit IRC16:29
*** tesseract has quit IRC16:33
*** yamamoto has joined #openstack-lbaas16:33
*** ianychoi has quit IRC16:34
*** salmankhan has quit IRC16:35
*** ianychoi has joined #openstack-lbaas16:35
*** yamamoto has quit IRC16:38
*** SumitNaiksatam has joined #openstack-lbaas16:45
*** yamamoto has joined #openstack-lbaas16:48
*** yamamoto has quit IRC16:53
*** Swami_ has joined #openstack-lbaas16:58
*** Swami has joined #openstack-lbaas16:58
*** Alexey_Abashkin has joined #openstack-lbaas16:58
*** AlexeyAbashkin has quit IRC16:58
*** Alexey_Abashkin is now known as AlexeyAbashkin16:58
*** aojea has quit IRC17:07
*** aojea has joined #openstack-lbaas17:07
*** AlexeyAbashkin has quit IRC17:10
*** yamamoto has joined #openstack-lbaas17:14
*** yamamoto has quit IRC17:14
*** AlexeyAbashkin has joined #openstack-lbaas17:28
*** AlexeyAbashkin has quit IRC17:36
*** yamamoto has joined #openstack-lbaas17:39
*** aojea has quit IRC17:42
*** yamamoto has quit IRC17:45
*** sshank has joined #openstack-lbaas17:47
*** openstackgerrit has quit IRC18:04
*** harlowja has joined #openstack-lbaas18:10
*** kobis has joined #openstack-lbaas18:33
*** amuller has quit IRC18:39
*** nmanos has quit IRC18:39
*** kobis has quit IRC18:40
*** kobis has joined #openstack-lbaas18:44
*** yamamoto has joined #openstack-lbaas18:46
*** kobis has quit IRC18:50
*** yamamoto has quit IRC18:52
*** sshank has quit IRC18:56
*** yamamoto has joined #openstack-lbaas19:48
*** yamamoto has quit IRC19:54
johnsom#startmeeting Octavia20:00
openstackMeeting started Wed Jun 13 20:00:47 2018 UTC and is due to finish in 60 minutes.  The chair is johnsom. Information about MeetBot at
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.20:00
*** openstack changes topic to " (Meeting topic: Octavia)"20:00
openstackThe meeting name has been set to 'octavia'20:00
johnsomHi folks!20:00
johnsomAnother fine day in Octavia land20:01
johnsom#topic Announcements20:01
*** aojea has joined #openstack-lbaas20:01
*** openstack changes topic to "Announcements (Meeting topic: Octavia)"20:01
johnsomI don't think I have any announcements this week.20:01
johnsomWe did release Rocky MS220:01
johnsomWe are now on the path towards MS3 the week of July 23rd20:02
johnsomWe need to start prioritizing what we want/need to get into Rocky.20:02
johnsomI plan to setup our "priority review" etherpad this week.20:03
johnsomDoes anyone have any other announcements today?20:03
johnsom#topic Brief progress reports / bugs needing review20:04
*** openstack changes topic to "Brief progress reports / bugs needing review (Meeting topic: Octavia)"20:04
johnsomI have been working on bug fixes, the driver support library, and some internal planning stuffs.20:04
xgerman_I got to know OSP 12…20:05
johnsomCurrently I am working on the amphora driver "updates" bug.  I had a patch and a plan yesterday but testing (+1 for octavia-tempest-plugin) showed that what I had was not a complete fix.20:05
johnsomI am now working on a new approach (well, the old one) to the problem.  Hopefully a patch will arrive today.20:05
johnsomThen I'm going to tackle the active/standby both amphora down bug. This is critical for Rocky (and backports) IMO20:06
nmagnezijohnsom, can you link that story please?20:07
* rm_work cheers for johnsom 20:07
johnsomSure, it might take me a minute to find it.20:07
rm_workif you want to discuss that, let me know20:07
johnsomYep, that one20:07
cgoncalveswe all can leave the room. johnsom is handling everything so nicely :)20:07
johnsomHappy to let someone take the dual down bug!!!!!!!20:08
rm_worki did have some thoughts, as i've been noodling it (and working in that area of the flows) for a bit20:08
johnsomStep right up!  No waiting....20:08
rm_workbut i don't have time to take it20:08
rm_workI think20:08
johnsomYeah, I am not sure if I will "make it good enough" or attack the "better way to do it".  It will just depend on the scope.  Whatever, it needs fixed sooner20:09
johnsomThat said, I have a number of Octavia patches open that could use some reviews20:10
johnsomAnybody else have updates?20:11
*** AlexeyAbashkin has joined #openstack-lbaas20:11
cgoncalvesgrenade patch is ready but awaits parent patch to merge20:12
cgoncalvesbarbical acl patch: I worked a little bit on it and got it to work, including revoke acls when not needed anymore20:12
johnsomcgoncalves Do grenade *need* that patch?  I think my zombie patch may replace rm_works20:13
cgoncalvesI've tested also
cgoncalvesjohnsom, I can rebase on top of yours instead, sure20:13
johnsomYeah, wondering if rm_work can review and decide if it's a complete duplicate or not20:13
johnsomI think our discussions led us to it was a dup20:14
johnsomIt would be great to get grenade merged so we can move forward on the updates stuff in Rocky20:14
johnsomcgoncalves How did you solve the multi-use of a cert when revoking?20:15
johnsomAlso, thank you for you work on all three of these.  Do you plan to vote on the DB speedup?20:15
*** AlexeyAbashkin has quit IRC20:16
cgoncalvesjohnsom, by querying if there's any other listener pointing to the same cert20:16
cgoncalvesif there is, do not revoke20:16
cgoncalvesjohnsom, I do plan but first I want to check with L7 rules. also health monitor listing still takes a long time20:17
johnsomHmm, ok, I will have to look at how that locks so we don't hit concurrency issues20:17
cgoncalvesI didn't want to discourage the author by asking to also work on the health monitor if other reviewers don't think it's required20:18
johnsomYeah, it is fine to incrementally make improvements.  As long as it doesn't make it worse...20:18
*** vegarl has quit IRC20:19
cgoncalvesso far it's been improving. the author had not touched listeners and I asked if he could otherwise I could submit since I had something drafted20:19
cgoncalvesanyway, I'll keep on reviewing it20:19
johnsomAny other updates?20:20
*** vegarl has joined #openstack-lbaas20:20
johnsom#topic Talk about API versioning/microversioning20:21
*** openstack changes topic to "Talk about API versioning/microversioning (Meeting topic: Octavia)"20:21
johnsomNow I know Adam is not able to participate in this discussion this week, so we will probably carry it over to next weeks meeting.20:21
johnsomHowever, I did look at this a bit.20:21
johnsomI'm not sure if others did any research on it.20:22
*** blake has joined #openstack-lbaas20:22
nmagnezijohnsom, I asked around about micro versioning in Neutron20:22
xgerman_well, they use extensions20:22
johnsomI think my biggest issue with microversions is the fact that if you don't ask for the newer versions with an HTTP header you get the oldest major20:22
johnsomnmagnezi What did you learn?20:23
nmagnezijohnsom, I was told to look at Nova... so, looks like Neutron doesn't really implement this. they indeed use api extentions20:23
johnsomYeah, that was my perception as well20:24
nmagnezias for Nova I didn't had the chance to reach out to anybody who work on it, but I started to read
johnsomAre two references I have been looking at.20:24
nmagnezijohnsom, question is: say we decide to use microversioning (or just bump minor versions) is that a thing we can still do in Rocky?20:25
johnsomTempest is basically just skipping tests if the API doesn't support the version20:25
xgerman_we can do version discovery20:26
johnsomnmagnezi I really think we should have a plan and advertise a version in Rocky.  Now a full on microversion support, maybe not.20:26
johnsomxgerman_ Not at the moment, the patch has not merged over the versioning discussion20:26
xgerman_+1 unless somebody has a lot of time20:26
nmagnezixgerman_, yeah I was suggested with that too (I presume you mean doing it in the tempest plugin). the question is how terrible will be the maintanance of it20:27
xgerman_yeah, that whole microversion stuff feels pre-mature + most clouds have a pretty recent version of Ovctavia20:27
xgerman_nova, on the other hand, has icehouse clouds to deal with20:28
johnsomRight.  I have some thoughts I need to put together, but at a high level I'm thinking we don't do microversions now. We do dot versions, 2.0, 2.1, 2.3 and only do additions to the API.  We then use a similar tempest setup that puts new features into versioned test classes. Or we cheat and just do skips.20:29
johnsomI only propose the skips because of the LB boot overhead the separate classes would bring.20:30
nmagnezijohnsom, skips by what? like in rm_work's patch?20:30
cgoncalvesso every single patch that touches the API should bump the dot version?20:30
johnsomUgh, another thing someone needs to work on is the multi-node gates20:30
johnsomcgoncalves right.20:30
johnsomnmagnezi I don't know, I haven't looked at his proposed patch in a few weeks20:31
nmagnezicgoncalves, yeah we might end up with version numbers that look like macOS :)20:31
xgerman_nothing wrong with that20:31
johnsomBetter than "all versions are 2.0"20:31
nmagnezijohnsom, +120:31
cgoncalvesok, wanted to make sure. it will be an important thing when reviewing20:31
johnsomThe nice thing with that approach is people will get the latest by default.20:32
johnsomDownside is it means we can't remove or break what we have there now.20:32
johnsomNot until a major bump20:32
nmagnezijohnsom, do you have anything in mind that you intended to refactor in the current API?20:33
johnsomWell, let's keep thinking about it and plan to make some decisions next week.20:33
johnsomUnless of course cgoncalves is out one yet another vacaction/holiday....  grin20:33
xgerman_he can vote by mail20:34
johnsom#topic Open Discussion20:34
*** openstack changes topic to "Open Discussion (Meeting topic: Octavia)"20:34
johnsomOther topics for today?20:34
nmagnezior by pidgin..20:34
xgerman_cgoncalves: nmagnezi wonder if you guys have a periodic Pike amp build somehwere at Redhat20:34
cgoncalvesjohnsom, I'm in Portugal since yesterday evening :) home-home office :)20:35
xgerman_(remember my OSP12 advantures…)20:35
nmagnezixgerman_, no sir. we have that starting Queens20:35
xgerman_mmh, where’s that at?20:35
nmagnezixgerman_, we didn't ship Octavia as part of OSP12.20:36
xgerman_yes, I know - hence my adventures20:36
nmagnezixgerman_, hmm. when OSP13 gets released.. but we do have something for RDO20:36
xgerman_ok, tell me more…20:36
cgoncalvesxgerman_, we do have periodic builds for master and queens in rdo20:37
cgoncalvesoh right, those links :)20:37
xgerman_sweet — might go with queens then20:37
johnsomAre those centos based?20:37
johnsomOh, yeah, I see it in the hostname20:38
nmagnezijohnsom, didn't check myself, but they should be :)20:38
cgoncalvessure they are centos-based :)20:38
johnsomYeah, just wasn't sure if it was a rhel that needed a key or something20:39
nmagneziyup. just remember that the user is centos (as opposed to ubuntu that is a habit for devstack workloads with default settings)20:40
johnsomAny other topics today?20:40
xgerman_will just switch off ssh20:41
johnsomOk, well, if there isn't anything else, have a good week folks!20:43
*** openstack changes topic to "Discussion of OpenStack Load Balancing (Octavia) |"20:44
openstackMeeting ended Wed Jun 13 20:44:11 2018 UTC.  Information about MeetBot at . (v 0.1.4)20:44
openstackMinutes (text):
*** yamamoto has joined #openstack-lbaas20:50
*** SumitNaiksatam has quit IRC20:54
*** yamamoto has quit IRC20:55
rm_worksad that the "make ssh user be consistent" thing died21:08
johnsomStill could be done...21:10
johnsomJust some DIB element work IMO21:12
rm_workhey nmagnezi / cgoncalves do you know anyone over there who does work on ansible?21:12
rm_workwould love to poke someone from redhat at ^_^21:13
rm_workjohnsom: for the record, even our *hardware in DCs* gets stuck on DHCP and has to sit for a minute or two waiting to time out21:15
rm_workon the default centOS image21:15
rm_workI'm not sure why you're so certain that DHCP isn't something that might need to be disabled in some cases :P21:15
johnsomrm_work DHCP happening is purely dependent on the neutron subnet configuration. It's on when the subnet has DHCP enabled, it's already disabled when the subnet is statically assigning the IP.21:17
rm_workis there a way I can check that with the neutron client?21:17
rm_workwill it show up in the show?21:17
johnsomenable_dhcp True21:18
johnsomThat is unless there is a bug in a cloud init version/implementation21:19
johnsomWhich maybe centos has21:19
johnsomIf I remember right, I went into the cloudinit code to confirm how I see it happen is what is coded. But, that whole discussion was a long time ago now21:20
rm_workyeah it's just this thing is very frustrating, and you've been blocking it for a loooong time21:21
johnsomI'm blocking bad/unnecessary code21:22
johnsomLooking back at that patch it assumes a interface name of "eth" which is not always the case21:24
rm_workjohnsom: enable_dhcp is False21:29
rm_workjohnsom: I believe the DHCP happens BEFORE the cloud-init stuff loads the network info21:30
*** SumitNaiksatam has joined #openstack-lbaas21:30
rm_workso that'd be a centos bug?21:30
johnsomMaybe, can you grab the nova console log from one of those booting?21:31
johnsomYeah, looks like I left it at "Please create a bug and provide the console log"21:33
rm_worklooking for it yeah21:34
rm_workwell it's hard for me21:34
rm_workbecause i already patched it out21:34
rm_workbut trying to get the log from like... one of our HVs21:35
rm_workah but they don't use the neutron networks :/21:35
rm_workhold on, i'll upload and boot one of the default images21:36
rm_workneed to find them again...21:37
rm_workah i'll grab one of carlos'21:37
*** fnaval has quit IRC21:42
*** yamamoto has joined #openstack-lbaas21:51
*** atoth has quit IRC21:54
*** yamamoto has quit IRC21:57
*** harlowja has quit IRC22:11
*** harlowja has joined #openstack-lbaas22:12
*** harlowja has quit IRC22:21
*** aojea has quit IRC22:21
rm_workbecause i am tired of being frustrated by this22:23
johnsomIt is lame you can't22:25
*** rcernin has joined #openstack-lbaas22:26
rm_workfor like ... so long22:26
rm_worki have been taking care of stuff that's frustrating me22:27
rm_workbut i've always just been annoyed and left it22:27
*** rcernin is now known as rcernin|brekkie22:27
*** rcernin|brekkie is now known as rcernin22:27
rm_workahh yeah i knew i forgot something22:30
*** threestrands has joined #openstack-lbaas22:47
rm_workfixed, added unit and functional22:49
rm_workwish i could run their functional locally but their functional is like... REAL CALLS. <_<22:49
*** yamamoto has joined #openstack-lbaas22:53
*** yamamoto has quit IRC22:59
*** blake has quit IRC23:22
*** blake has joined #openstack-lbaas23:23
*** blake has quit IRC23:30
*** blake has joined #openstack-lbaas23:31
*** yamamoto has joined #openstack-lbaas23:55

Generated by 2.15.3 by Marius Gedminas - find it at!