Friday, 2016-06-17

sbalukoffIt's probably going to be a while before n-lbaas goes away. How bad would it be for that information not bubbling up for, say... a long time.00:00
rm_workso i'd probably vote we fix the status issue00:00
sbalukoffYeah, feature parity is unlikely to make it this cycle, IMO.00:00
rm_workbecause yeah, i'd like that to not be a bug all the way into O-release00:00
TrevorV2sbalukoff well it ALREADY doesn't, so its "regular" right now?00:00
rm_worklol00:00
rm_workit's a "bug" now IMO00:00
TrevorV2Agreed, but no one has caught it yet, cept me, and no one is complaining about it, so idk?00:00
sbalukoffYeah... we should probably fix that in n-lbaas.00:00
sbalukoffOtherwise people will complain that this "feature" has changed.00:01
TrevorV2I mean, I'm not advocating to ignore it, don't get me wrong, just curious what comm's thoughts are00:01
rm_workand a pretty bad one actually, because without node status info, the user has no way to deal with fixing their nodes (and it's actually actively misleading)00:01
rm_work"oh god why are all my nodes offline"00:01
rm_workmakes it completely impossible for anyone to automate fixing their stuff00:01
TrevorV2rm_work to be clear, it actually records that all nodes are "online" when they might not necessarily be00:01
rm_workrofl00:02
rm_workok well, that's ... maybe slightly better, but not much00:02
sbalukoffIt's not better.00:03
sbalukoff(It's worse, IMO, to report everything OK when it's not....  at least a false alarm means production services aren't actually affected.)00:03
*** SumitNaiksatam has quit IRC00:05
*** piet has quit IRC00:06
rm_workeither way it's bad00:07
TrevorV2Alright, well then we need to fix octavia's reporting mechanism00:08
rm_worki think it's just that ... it was never added00:08
sbalukoffSounds like it.00:08
rm_worknot that it broke00:08
TrevorV2And make sure neutron lbaas is going to update the operating status with that same mechanism00:08
TrevorV2agreed rm_work00:08
rm_workthere's a framework for sending events to the queue for n-lbaas to pick up00:08
TrevorV2Alrighty, I'll take look at that then00:08
rm_work(part of what carlos wrote)00:08
rm_workyou just have to add the event in the right spot00:08
rm_workI guess the hm?00:08
rm_workwhatever detects nodes going online/offline00:09
rm_workand it should send it up the queue00:09
rm_workthen need something on the n-lbaas side to read them and update DB accordingly00:09
rm_workit's uh...00:09
johnsomSomeone rang00:09
rm_workhttps://github.com/openstack/octavia/blob/master/octavia/controller/queue/event_queue.py00:10
rm_workjohnsom: was wondering if we're still blocking https://review.openstack.org/#/c/298424/ for some reason or another00:11
rm_workand why exactly00:11
rm_workTrevorV2: ^^ event_queue.py00:11
johnsomLooking00:11
rm_workI think you inherit that and call the method for it00:11
rm_workor ... uhh something like that00:11
rm_workdamn, let me try to find the example00:12
TrevorV2rm_work I know what "listens" on neutron lbaas00:12
TrevorV2But I've just never dug into the bit that sends it to the queue in octavia00:12
rm_workhmm k good because I don't00:12
rm_workyeah i only know the octavia side00:12
rm_workbecause I helped design it with carlos00:12
rm_workok yeah not inherit, sorry00:13
rm_workhttps://github.com/openstack/octavia/blob/master/octavia/controller/healthmanager/update_db.py#L3800:13
rm_workyou make a thing using stevedore00:13
rm_workand then call .emit() on it00:13
rm_workwith your message...00:13
rm_workwhich, I think we had specific classes for them? :/00:13
rm_workit's been a while, you can figure it out, but everything you need should already exist00:14
rm_workjust need to actually emit the events in the right place00:14
johnsomrm_work I think my issue was around fixing the issue in the upstart script as well as the init.d.  Upstart is what the decstack, etc uses00:15
rm_workyeah though not for tooooo long00:15
rm_workupstart is dead00:15
johnsomRight.  Volunteering to do systemd?00:15
rm_workI mean00:15
rm_workwhat is this... sysvinit?00:16
rm_workI don't even know00:17
*** amotoki has quit IRC00:17
rm_workthe init landscape in linux is a clusterf@$%00:17
johnsomYeah, the patch is against the sysvinit scripts.  Agreed00:17
rm_workcan we just fix this and then focus on other things in other patches? >_>00:17
johnsomI haven't gone too deep one the systemd stuff yet, but it seems that is the way folks are going00:18
rm_workyeah, and great! let's add that00:18
rm_workthat has nothing to do with fixing sysvinit >_>00:18
rm_workit should be a different CR00:18
rm_workI'm just looking to get some acknowledgement from the various -1s on there right now00:19
rm_workthat blocking this because it doesn't ALSO fix some other unrelated init system is a bit silly00:19
rm_workif there was another concern (like, we should try to fix the root cause of the werkzeug restart issue) then I'm open to hearing that00:20
rm_workbut this is >_>00:20
johnsomSystemd is definitely a different patch.  Fixing the upstart at the same time as the sysvinit, I need to think about.  Let me read the patch again.  On my phone screen at the moment00:20
rm_workah lol00:20
rm_worknp00:20
sbalukoffrm_work: Is there a launchpad bug on this?00:22
rm_workhmmm, I don't know, let me check00:22
sbalukoffI think I'd be OK merging this as is, so long as it's listed as a fix for a bug. ...and also a second bug report for dealing with the systemd changes that need to happen. (Or that same bug, if this one is referenced as a partial fix for it.)00:23
johnsomYes, for werkz-whatever bug there is.  It is blocking another patch too if I remeber00:23
sbalukoffAdam: Can you update the commit message to actually reference the bug, then? (I think you said you would in a previous comment.)00:24
johnsomOk, so I remember this now, I IRC chatted with you about it.  My ask was to just add the process kill to the pre-stop upstart block in the upstart script.  If this fixes the werkz issue, that should solve it for everyone00:25
rm_workYeah, I remember that00:25
rm_workI also remember that I don't know how upstart works, and it's not used in any OS I touch anymore :P00:26
sbalukoffHeh!00:26
johnsomI think I gave a pointer to the upstart docs for that block too.  Not sure if I can paste the url via my phone00:27
rm_workhmm, failing at finding the bug too >_>00:28
johnsomhttp://upstart.ubuntu.com/cookbook/#pre-stop00:29
johnsomHa, it does work00:29
johnsomOk, just a sec and I will find the bug00:29
rm_workdo you remember what the bug was called?00:29
rm_worki just read through all of them and none of them SEEM related00:29
rm_work*scanned the titles00:30
johnsomYeah, I don't see it either.  I wonder if it got closed somehow.  Still looking.  It was assigned to xgerman if I remember right00:35
rm_workhmm00:35
rm_workadding in all the types00:35
TrevorV2rm_work I'm confused about something00:36
rm_workhttps://bugs.launchpad.net/octavia/+bug/149662800:36
openstackLaunchpad bug 1496628 in octavia "Amphora agent reload fails with socket in use" [High,Fix released] - Assigned to German Eichberger (german-eichberger)00:36
rm_workthis maybe00:36
TrevorV2in "update_db.py" line 6400:36
johnsomhttps://bugs.launchpad.net/bugs/149662800:37
johnsomI looks like it got closed.  I am not convinced that fixed it though00:38
rm_workme either00:38
johnsomFyi, I have been off the grid for the most part this week.  Mailed my gear back to HP and my personal laptop order got delayed, so, yeah.  It ended up a vacation week.00:45
rm_workheh00:45
rm_workyeah00:45
rm_workwhat about your new work laptop order? :P00:45
johnsomI don't know.  I gave Cody my selections.  I hope it will be there when I arrive next week.00:46
rm_workyeah, hopefully00:46
somerville32I believe it is.00:46
johnsomI take it you aren't heading down for next week?00:46
*** Purandar has quit IRC00:47
*** diogogmt has joined #openstack-lbaas00:52
*** rudrajit_ has joined #openstack-lbaas00:56
*** rudrajit has quit IRC00:59
rm_workI am not T_T00:59
*** diogogmt has quit IRC01:06
*** ducncvn has quit IRC01:25
konghi, guys, can we remove the steps of https://github.com/openstack/octavia/blob/master/devstack/plugin.sh#L141 here. and do the similar thing inside octavia-healthmonitor process? just like the way neutron-dhcp-agent does?01:29
kongour operators feel scared when they saw the a little bit 'hacking' thing in the script.01:30
kongor, do you have other suggestions to do that?01:30
rm_workhacking thing?01:30
rm_workThis is devstack plugin only...01:30
rm_workmost operators should never be running this :P01:31
*** bana_k has quit IRC01:33
kongrm_work: hi, but if not, how they could do the similar thing? Our operators just take that plugin for reference.01:37
kongs/plugin/plugin.sh01:37
*** kevo has quit IRC01:38
kongthey need to do similar things on the host o-healthmonitor is running, anyway01:39
kongI took a look at the neutron-dhcp-agent code, it also need to connect various tenant networks01:40
kongand it just creates the necessary port/interface by itself01:41
kongthings operators need to do, is just start dhcp-agent process01:41
kongwhat I mean is, why not we do the similar thing in octavia-healthmonitor process/01:42
johnsomkong Maybe there is some confusion.   This code in the plugin does not plug tenant networks.01:56
kongjohnsom: hi, I mean, the network that between the amphora vms and health monitor01:56
kongmanagement network01:57
johnsomThis just creates a management network for the amps in devstack.  Most operators manually create the network, typically as a provider network.  At least that is how we do it in Helion01:58
*** TrevorV2 has quit IRC02:00
kongjohnsom: is there any document for me to learn about how Helion does this?02:01
johnsomWell, docs.hpcloud.net has the documentation02:02
*** SumitNaiksatam has joined #openstack-lbaas02:04
*** SumitNaiksatam has quit IRC02:12
*** piet has joined #openstack-lbaas02:22
*** Purandar has joined #openstack-lbaas02:27
*** Purandar has quit IRC02:32
*** piet has quit IRC02:35
*** rudrajit_ has quit IRC02:38
*** somerville32 has quit IRC02:42
*** Purandar has joined #openstack-lbaas02:43
*** Purandar has quit IRC02:48
kongjohnsom: one question, Can tenant-network and provider-network exist together at one deployment?03:01
kongI'm not very farmiliar with neutron03:01
johnsomYes03:02
kongjohnsom: thanks!03:02
*** amotoki has joined #openstack-lbaas03:07
*** amotoki has quit IRC03:09
*** anilvenkata has joined #openstack-lbaas03:40
*** Purandar has joined #openstack-lbaas03:49
*** amotoki has joined #openstack-lbaas04:00
openstackgerritMerged openstack/neutron-lbaas: Use constants from neutron-lib  https://review.openstack.org/33048904:03
openstackgerritMerged openstack/neutron-lbaas: Add in missing translation for exception  https://review.openstack.org/33032204:10
openstackgerritMerged openstack/octavia: Set device owner for health manager port  https://review.openstack.org/32564904:11
*** amotoki has quit IRC04:18
*** links has joined #openstack-lbaas04:20
*** atul has quit IRC04:28
*** Purandar has quit IRC04:38
*** yuanying has quit IRC04:41
*** amotoki has joined #openstack-lbaas04:45
*** yamamoto has quit IRC04:46
*** Purandar has joined #openstack-lbaas04:46
*** fnaval has quit IRC04:48
*** fnaval has joined #openstack-lbaas04:59
*** fnaval has quit IRC04:59
*** fnaval has joined #openstack-lbaas04:59
*** yamamoto has joined #openstack-lbaas05:26
*** links has quit IRC05:27
*** links has joined #openstack-lbaas05:28
*** amotoki has quit IRC05:56
*** Purandar has quit IRC06:01
*** amotoki has joined #openstack-lbaas06:02
*** yuanying has joined #openstack-lbaas06:02
*** bana_k has joined #openstack-lbaas06:15
*** kevo has joined #openstack-lbaas06:22
*** duvarenkov__ has quit IRC06:22
*** rcernin has joined #openstack-lbaas06:23
*** pcaruana has joined #openstack-lbaas06:29
*** kobis has joined #openstack-lbaas06:31
*** duvarenkov has joined #openstack-lbaas06:32
*** mixos has quit IRC06:59
*** yuanying has quit IRC07:02
*** ihrachys has joined #openstack-lbaas07:03
*** kobis has quit IRC07:03
*** amotoki has quit IRC07:03
*** bana_k has quit IRC07:05
*** atul has joined #openstack-lbaas07:07
*** ihrachys has quit IRC07:08
*** tesseract has joined #openstack-lbaas07:10
*** dmk0202 has joined #openstack-lbaas07:32
*** dmk0202 has quit IRC07:33
*** kobis has joined #openstack-lbaas07:33
*** kobis has quit IRC07:39
*** rcernin has quit IRC07:45
*** yamamoto has quit IRC08:01
*** yuanying has joined #openstack-lbaas08:05
*** eezhova has joined #openstack-lbaas08:06
*** noshankus has joined #openstack-lbaas08:09
*** rcernin has joined #openstack-lbaas08:10
*** redrobot has quit IRC08:21
*** redrobot has joined #openstack-lbaas08:24
*** redrobot is now known as Guest3155308:24
*** dmk0202 has joined #openstack-lbaas08:27
*** mestery_ has joined #openstack-lbaas08:36
*** jorgem| has quit IRC08:38
*** jorgem has joined #openstack-lbaas08:39
*** yamamoto has joined #openstack-lbaas08:39
*** mbanszex has quit IRC08:39
*** mbanszex has joined #openstack-lbaas08:40
*** harlowja_ has quit IRC08:41
*** clduser has quit IRC08:42
*** mestery has quit IRC08:43
*** mdavidson has quit IRC08:43
*** mestery_ is now known as mestery08:43
*** clduser has joined #openstack-lbaas08:43
*** ihrachys has joined #openstack-lbaas08:44
*** mdavidson has joined #openstack-lbaas08:44
*** kobis has joined #openstack-lbaas08:54
openstackgerritSirushti Murugesan proposed openstack/neutron-lbaas: Add python 3 classifiers  https://review.openstack.org/33096709:11
*** openstackgerrit has quit IRC09:18
*** openstackgerrit has joined #openstack-lbaas09:19
*** eezhova has quit IRC09:19
*** eezhova has joined #openstack-lbaas09:24
*** kobis has quit IRC09:38
*** kobis1 has joined #openstack-lbaas09:38
*** eezhova has quit IRC10:01
*** openstackgerrit has quit IRC10:03
*** openstackgerrit has joined #openstack-lbaas10:04
*** rcernin has quit IRC10:05
*** eezhova has joined #openstack-lbaas10:06
*** kobis1 has quit IRC10:12
*** numans has joined #openstack-lbaas10:15
*** kobis has joined #openstack-lbaas10:20
openstackgerritElena Ezhova proposed openstack/neutron-lbaas: Set up hooks for the functional job and add test_migrations test  https://review.openstack.org/32099910:46
*** permalac has joined #openstack-lbaas10:47
*** rtheis has joined #openstack-lbaas10:52
*** eezhova has quit IRC11:03
*** anilvenkata has quit IRC11:29
*** rcernin has joined #openstack-lbaas11:47
*** eezhova has joined #openstack-lbaas11:52
*** dmk0202 has quit IRC12:05
*** numans has quit IRC12:17
*** piet has joined #openstack-lbaas12:37
*** TrevorV_ has joined #openstack-lbaas12:50
*** matt-borland has joined #openstack-lbaas12:53
openstackgerritElena Ezhova proposed openstack/neutron-lbaas: Set up hooks for the functional job and add test_migrations test  https://review.openstack.org/32099913:20
*** mixos has joined #openstack-lbaas13:41
*** piet has quit IRC13:45
*** links has quit IRC13:46
*** piet has joined #openstack-lbaas13:48
*** piet has quit IRC13:49
*** armax has joined #openstack-lbaas13:54
*** eezhova has quit IRC14:07
*** eezhova has joined #openstack-lbaas14:14
*** Frito has joined #openstack-lbaas14:19
Fritoo/14:19
Fritobouncer is having trouble connecting so I'll just have to direct connect :-)14:20
*** ihrachys has quit IRC14:45
*** duvarenkov has quit IRC14:46
*** ajmiller has quit IRC14:53
TrevorV_You sound like you're complaining, but I don't think you are14:53
*** ajmiller has joined #openstack-lbaas14:53
*** duvarenkov has joined #openstack-lbaas14:53
*** pcaruana has quit IRC14:54
*** daneyon has joined #openstack-lbaas14:55
*** fnaval has quit IRC14:55
*** kobis has quit IRC14:56
*** kobis has joined #openstack-lbaas14:56
Fritolol14:57
Fritosmidget of complaint there but more apathy than anything14:58
diltrammorning15:02
*** diogogmt has joined #openstack-lbaas15:02
xgermanmorning15:02
*** diogogmt has quit IRC15:04
*** tesseract has quit IRC15:05
*** rcernin has quit IRC15:07
*** amotoki has joined #openstack-lbaas15:09
*** fnaval has joined #openstack-lbaas15:12
*** mixos has quit IRC15:27
*** Purandar has joined #openstack-lbaas15:40
*** ajmiller_ has joined #openstack-lbaas15:40
matt-borlandhere's a patch to LBaaS UI...it wasn't taking OPENSTACK_ENDPOINT_TYPE into account for its Barbican calls: https://review.openstack.org/#/c/330697/15:43
*** ajmiller has quit IRC15:43
*** Guest31553 is now known as redrobot15:50
*** daneyon has quit IRC15:56
*** permalac has quit IRC16:03
*** noshankus has quit IRC16:08
*** dmk0202 has joined #openstack-lbaas16:09
*** rudrajit has joined #openstack-lbaas16:09
*** eezhova has quit IRC16:10
*** mdavidson has quit IRC16:11
*** rudrajit has quit IRC16:19
*** rudrajit has joined #openstack-lbaas16:20
*** mixos has joined #openstack-lbaas16:27
*** rudrajit has quit IRC16:29
diltrammatt-borland: reviewsed16:35
matt-borlanddiltram, thank you!  :)16:35
diltrammatt-borland: you're welcome16:36
*** TrevorV_ has quit IRC16:49
*** TrevorV_ has joined #openstack-lbaas16:51
*** harlowja has joined #openstack-lbaas16:53
*** kevo has quit IRC16:54
*** bana_k has joined #openstack-lbaas16:56
*** fnaval_ has joined #openstack-lbaas17:00
*** fnaval has quit IRC17:00
*** manjrem has joined #openstack-lbaas17:08
*** yamamoto has quit IRC17:10
*** amotoki has quit IRC17:13
*** daneyon has joined #openstack-lbaas17:16
*** daneyon_ has joined #openstack-lbaas17:19
*** daneyon has quit IRC17:22
*** piet has joined #openstack-lbaas17:31
*** ajmiller has joined #openstack-lbaas17:40
*** ajmiller_ has quit IRC17:43
*** rudrajit has joined #openstack-lbaas17:46
*** kobis has quit IRC17:49
*** manjrem has quit IRC17:57
*** cody-somerville has joined #openstack-lbaas17:59
*** harlowja has quit IRC18:04
*** kevo has joined #openstack-lbaas18:07
*** yamamoto has joined #openstack-lbaas18:10
*** kobis has joined #openstack-lbaas18:14
*** cody-somerville has quit IRC18:17
*** Trevor_V_ has joined #openstack-lbaas18:17
*** Trevor_V_ has quit IRC18:17
*** Trevor_V_ has joined #openstack-lbaas18:18
*** TrevorV_ has quit IRC18:21
*** yamamoto has quit IRC18:24
*** Trevor_V_ has quit IRC18:25
*** ajmiller has quit IRC18:29
*** ajmiller_ has joined #openstack-lbaas18:29
*** Frito has quit IRC18:35
*** Frito has joined #openstack-lbaas18:35
*** Frito has quit IRC18:47
*** cody-somerville has joined #openstack-lbaas18:52
*** TrevorV_ has joined #openstack-lbaas19:01
*** Frito has joined #openstack-lbaas19:04
*** TrevorV_ has quit IRC19:06
*** shakamunyi has joined #openstack-lbaas19:13
*** piet has quit IRC19:25
*** piet has joined #openstack-lbaas19:26
*** harlowja has joined #openstack-lbaas19:32
*** TrevorV_ has joined #openstack-lbaas19:38
*** piet has quit IRC20:00
openstackgerritMerged openstack/neutron-lbaas: Fix subunit trace help  https://review.openstack.org/33115420:01
*** daneyon_ has quit IRC20:06
*** kobis has quit IRC20:08
*** daneyon has joined #openstack-lbaas20:10
*** piet has joined #openstack-lbaas20:13
*** ihrachys has joined #openstack-lbaas20:22
openstackgerritMerged openstack/neutron-lbaas: Updated from global requirements  https://review.openstack.org/33117320:32
openstackgerritMerged openstack/octavia: Updated from global requirements  https://review.openstack.org/33117520:32
*** ihrachys has quit IRC20:38
*** TrevorV_ has quit IRC20:47
*** Frito has quit IRC20:47
*** davidlenwell has quit IRC21:02
*** davidlenwell has joined #openstack-lbaas21:13
*** piet has quit IRC21:15
*** cody-somerville has quit IRC21:20
*** rtheis has quit IRC21:22
*** dmk0202 has quit IRC21:33
*** dmk0202 has joined #openstack-lbaas21:34
*** dmk0202 has quit IRC21:50
*** matt-borland has quit IRC22:18
*** mixos has quit IRC22:35
openstackgerritAdam Harwell proposed openstack/octavia: Updates to amphora-agent init.d script for Debian  https://review.openstack.org/29842422:36
rm_workjust rebasing ^^22:36
*** bana_k has quit IRC23:37

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!