Tuesday, 2015-03-31

openstackgerritDoug Wiegley proposed stackforge/octavia: Sync with oslo-incubator, tweak as needed  https://review.openstack.org/16911600:06
openstackgerritmin wang proposed openstack/neutron-lbaas: Introduced tempest API tests for openstack/neutron-lbaas  https://review.openstack.org/16843900:09
*** banix has joined #openstack-lbaas00:10
openstackgerritmin wang proposed openstack/neutron-lbaas: Introduced tempest API tests for openstack/neutron-lbaas  https://review.openstack.org/16843900:11
*** Aish has quit IRC00:17
*** madhu_ak has quit IRC00:18
*** xgerman_ has quit IRC00:20
*** bharath has quit IRC00:23
openstackgerritAdam Harwell proposed stackforge/octavia: Implemented Queue Consumer  https://review.openstack.org/14978900:28
*** banix has quit IRC00:28
*** sbfox has joined #openstack-lbaas00:30
crc32no lines changed.00:33
*** blogan_ has joined #openstack-lbaas00:40
openstackgerritBrandon Logan proposed stackforge/octavia: Added neutron allowed address pairs network driver  https://review.openstack.org/16744100:52
*** Varun_Lodaya has quit IRC00:52
rm_workcrc32: just a rebase00:53
rm_workcrc32: johnsom fixed the pep8 issue00:53
rm_workcrc32: or did you mean, you are getting that error?00:53
crc32neither. I was looking for code changes.00:54
rm_workyou try to push up a review yet?00:55
*** jamiem has quit IRC00:56
crc32no still screwing around with mock00:56
*** blogan_ has quit IRC00:58
*** sballe has quit IRC01:08
*** amotoki has joined #openstack-lbaas01:12
*** amotoki has quit IRC01:13
*** haigang has joined #openstack-lbaas01:13
*** sbfox has quit IRC01:17
*** banix has joined #openstack-lbaas01:18
*** blogan_ has joined #openstack-lbaas01:25
*** blogan_ has quit IRC01:26
*** amotoki has joined #openstack-lbaas01:42
*** amotoki has quit IRC01:47
*** amotoki has joined #openstack-lbaas01:47
*** amotoki has quit IRC01:52
*** banix has quit IRC02:14
*** SumitNaiksatam has quit IRC02:14
*** amotoki has joined #openstack-lbaas02:28
*** banix has joined #openstack-lbaas02:34
*** Varun_Lodaya has joined #openstack-lbaas02:40
*** sbfox has joined #openstack-lbaas02:42
*** amotoki has quit IRC02:43
*** crc32 has quit IRC02:44
*** haigang has quit IRC02:57
*** amotoki has joined #openstack-lbaas03:10
*** banix has quit IRC03:11
*** mlavalle has joined #openstack-lbaas03:11
*** mlavalle has quit IRC03:12
*** mlavalle has joined #openstack-lbaas03:13
*** amotoki has quit IRC03:16
*** haigang has joined #openstack-lbaas03:22
*** bharath has joined #openstack-lbaas03:23
*** xgerman_ has joined #openstack-lbaas03:24
*** haigang has quit IRC03:27
*** bharath has quit IRC03:28
*** sbalukoff has quit IRC03:32
*** ajmiller__ has joined #openstack-lbaas03:34
*** ajmiller_ has quit IRC03:37
*** SumitNaiksatam has joined #openstack-lbaas03:38
*** amotoki has joined #openstack-lbaas03:43
*** Varun_Lodaya has quit IRC03:53
*** haigang has joined #openstack-lbaas03:55
*** Varun_Lodaya has joined #openstack-lbaas03:57
*** Varun_Lodaya has joined #openstack-lbaas03:57
*** sbfox has quit IRC03:59
*** BrianShang_ has quit IRC04:13
*** BrianShang has joined #openstack-lbaas04:15
*** xgerman_ has quit IRC04:32
*** Varun_Lodaya has quit IRC04:40
*** mlavalle has quit IRC04:55
*** sbalukoff has joined #openstack-lbaas04:58
*** Varun_Lodaya has joined #openstack-lbaas05:03
*** Varun_Lodaya has quit IRC05:26
*** apuimedo has joined #openstack-lbaas05:45
dougwigSantosh_ns_: hiya05:48
* rm_work waves at dougwig 05:50
dougwigrm_work: hiya05:50
rm_workdougwig: just noticed in infra that you were asking about the same pypi bug i was just researching T_T05:50
rm_workand that they've got it handled, hopefully05:51
dougwigi noticed they were chugging on it, so i paused my CI and will try again in the morning.  :)05:51
rm_workalways great timing on these things05:51
dougwigrm_work: if you're bored, this is merge conflict central if it sits too long: https://review.openstack.org/#/c/169116/05:51
rm_workbut blogan already +2'd05:51
rm_workyeah, jeeze, this is messy05:52
dougwigyeah, it was no fun to create, either.  :)05:53
rm_worki especially like cache.py05:55
rm_work"code from octavia-incubator"05:55
rm_workSOMEONE did a find-replace a while back :P05:55
rm_workerr sorry05:55
rm_workdougwig: +2 … dunno about +A as it's just RAX right now05:58
dougwigrm_work: everything in that dir is edited by a script, yeah.05:59
dougwighmm, i'm not from rax.  :)05:59
rm_workok, you convinced me05:59
rm_workman, our gate runs quick06:04
openstackgerritMerged stackforge/octavia: Sync with oslo-incubator, tweak as needed  https://review.openstack.org/16911606:05
openstackgerritAdam Harwell proposed stackforge/octavia: Implements Octavia Controller Worker  https://review.openstack.org/15149606:07
*** jschwarz has joined #openstack-lbaas06:09
rm_workman, yeah, lots of conflicts now on other patchsets T_T06:10
*** sbfox has joined #openstack-lbaas06:17
dougwigYeah it was a beast06:22
openstackgerritAdam Harwell proposed stackforge/octavia: Implemented Queue Consumer  https://review.openstack.org/14978906:23
bloganwhat yall doing up06:26
rm_workfixing stuff06:26
rm_workwas about to rebase two of your patchsets actually blogan06:26
rm_workblogan: you could do it instead if you're up :P06:26
blogandoing it now06:26
bloganway ahead of ya sparky06:27
bloganstupid dougwig breaking shit06:27
*** haigang has quit IRC06:27
rm_worki was about to `git review` https://review.openstack.org/#/c/168736/06:27
rm_workso not THAT far ahead06:27
*** haigang has joined #openstack-lbaas06:27
*** amotoki has quit IRC06:27
bloganthats the one i just did, running tox as we speak06:27
bloganbut rebuilding environments too06:28
bloganyou can review that one too06:28
rm_workdetox == winning06:29
* rm_work has been raving about "detox" today06:29
bloganyou and the new hotness06:29
bloganit just runs the environments in parallel?06:31
dougwigWhat is detox?06:31
dougwigblogan: fine, you can do the next Oslo update.  :)06:31
*** haigang has quit IRC06:32
openstackgerritBrandon Logan proposed stackforge/octavia: DRY'ed keystone session creation and retrieval  https://review.openstack.org/16873606:32
blogandougwig: in the words of emo darth vader...06:33
*** BrianShang has quit IRC06:35
*** BrianShang has joined #openstack-lbaas06:36
openstackgerritBrandon Logan proposed stackforge/octavia: Added neutron allowed address pairs network driver  https://review.openstack.org/16744106:39
rm_workblogan: yeah, just runs everything in parallel06:39
rm_workwhich is super nice when installing deps takes like 90% of the time for the test runs, and you need to do it like 4 times06:39
bloganthey should have called it paratox06:39
rm_workdistributed tox :)06:39
bloganbutrificated tox = butox06:40
blogani would have so tried to make buttox06:40
*** BrianShang_ has joined #openstack-lbaas06:41
openstackgerritMadhusudhan Kandadai proposed openstack/neutron-lbaas: Introduced tempest API tests for openstack/neutron-lbaas  https://review.openstack.org/16562706:43
openstackgerritMadhusudhan Kandadai proposed openstack/neutron-lbaas: Introduced tempest API tests for openstack/neutron-lbaas  https://review.openstack.org/16562706:44
*** BrianShang has quit IRC06:44
*** amotoki has joined #openstack-lbaas06:53
*** evgenyf has joined #openstack-lbaas06:58
*** BrianShang_ has quit IRC07:00
*** amotoki has quit IRC07:01
*** amotoki has joined #openstack-lbaas07:02
*** BrianShang has joined #openstack-lbaas07:02
bloganptoohill: ping07:04
*** haigang has joined #openstack-lbaas07:12
*** _kiran_ has joined #openstack-lbaas07:16
*** _kiran_ is now known as kiran-r07:16
ptoohillpong: blogan07:17
bloganptoohill: what did you mean by verifying the version?07:18
ptoohillyour just checking for v2 then everything else uses the v3 client07:18
ptoohillshouldnt be a problem07:18
ptoohillmore of a nit07:18
bloganyeah v3 works for v3-vN07:19
bloganand v107:19
bloganand v007:19
ptoohillfair enough, i just assumed there woudl be something specific that may change, idk, it shouldnt be a prob. didnt mins 1 for that07:19
blogando you know why project_id is needed in barbican and nova for v3 keystone?07:20
blogani was joking, you're right i should have checked for it07:20
ptoohillwhere is it needed in barbican?07:20
bloganthats what the original code had in octavia07:21
bloganrm_work: do you know why project_id was needed?07:22
rm_workthe old method07:22
rm_workthe new method doesn't need it07:22
rm_worksee the updates we did in neutron-lbaas07:22
bloganok i shall follow that bc wiht that the arguments are the same07:22
bloganand it can be simplified07:22
bloganthough i wonder if nova needs that07:23
rm_workthey should all match07:23
rm_worki would have backported the change already if i hadn't been stuck on CLB107:23
rm_workhey, that _SESSION catch was MY catch :P07:23
bloganptoohill will forever has his name on that07:27
bloganhe has been credited and will get the reward07:27
rm_workdamn, all those OpenStack Credits, stolen out from under me!07:28
rm_workHow will I ever afford the 10,000,000 Credit stuffed OpenStack Bear? T_T07:28
bloganwith the 100,000,000 Credit reward for a bug ptoohill found07:29
rm_workugh, i feel a sore throat coming on07:32
rm_workthe rest of this week is going to suck07:32
rm_workptoohill / blogan: can we all just vote to sleep through the huddle that's in 7.5 hours? <_<07:33
*** jschwarz has quit IRC07:33
openstackgerritBrandon Logan proposed stackforge/octavia: DRY'ed keystone session creation and retrieval  https://review.openstack.org/16873607:34
rm_workblogan: so is there another patchset for this coming soon, that I should be waiting for? :P07:34
rm_workthere it is07:34
rm_workgreat timing07:34
rm_workso, does it really just use 2/3 for auth version?07:35
rm_worknot 2.0/3 ?07:35
rm_work"if cfg.CONF.keystone_authtoken.auth_version == '2':"07:35
rm_workI can never remember what uses 2 and what uses 2.007:36
bloganfrom the exmaples ive seen07:36
blogannova uses minor versions now07:36
bloganthough you wouldn't know it form their client07:36
*** eugene_bag has joined #openstack-lbaas07:36
bloganversioning is quite inconsistent07:36
openstackgerritBrandon Logan proposed stackforge/octavia: Added neutron allowed address pairs network driver  https://review.openstack.org/16744107:37
rm_workthis is keystone tho07:37
rm_workshould probably wait on jenkins for the +2, but whatev07:38
bloganhold off on the +207:38
blogani want to verify that its not 2.0 and jsut 207:38
bloganbc if it is 2.0, neutron_lbaas is wrong too :(07:38
rm_worktoo late07:39
rm_workbut i can always retract it07:39
*** eugene_bag1 has joined #openstack-lbaas07:40
*** woodster_ has quit IRC07:40
*** eugene_bag has quit IRC07:41
bloganglad their help message is useful07:41
blogan'API version of the admin Identity API endpoint.'07:41
*** evgenyf has quit IRC07:42
*** evgenyf has joined #openstack-lbaas07:42
bloganargh i see examples using 2, v2, 2.007:42
bloganand now v2.007:43
rm_workyeah i don't even07:44
bloganim just going to keep it as is, if its wrong it can be fixed later07:45
bloganalright bed times07:45
bloganhave a good night07:45
rm_workoctavia suffers from the tox pyc bug07:46
rm_workgotta submit a fix07:46
*** pck has quit IRC07:47
ptoohillwhy only 2/3?07:47
rm_workptoohill: ?07:48
*** kobis has joined #openstack-lbaas07:48
ptoohillhes comparing the version as '2' or '3'07:48
ptoohillguess theres no hard rule on what it should be ive just seen every where else use v2/v307:48
ptoohillfor the auth_version07:48
rm_workwe were just discussing that07:50
rm_workit's WTF07:50
ptoohilloh, i just started reading back :P07:50
ptoohillkk, never mind then :)07:51
openstackgerritAdam Harwell proposed stackforge/octavia: Implements Octavia Controller Worker  https://review.openstack.org/15149607:51
*** evgenyf has quit IRC07:52
openstackgerritPhillip Toohill proposed openstack/neutron-lbaas: Fixes passphrase encoding error  https://review.openstack.org/16920607:54
rm_workugh going to need ANOTHER patchset on that07:56
*** pck has joined #openstack-lbaas07:56
rm_workpy27 issue07:57
rm_workdougwig: you still up?07:57
ptoohillon mine?07:57
rm_workerr, johnsom's, really07:58
ptoohillah, alrighty, im calling it el niteo07:58
rm_worki was trying to be nice and fix it since the patch we merged an hour or two ago broke a ton of stuff07:58
rm_worki wonder if i'll sleep through my alarm07:58
rm_workif i don't show up for the huddle… uhh… say i'm working on … stuff07:59
ptoohillsounds good to me ;)07:59
ptoohillTake care, g'night07:59
openstackgerritAdam Harwell proposed stackforge/octavia: Implemented Queue Consumer  https://review.openstack.org/14978908:12
openstackgerritAdam Harwell proposed stackforge/octavia: Implements Octavia Controller Worker  https://review.openstack.org/15149608:12
*** sbfox has quit IRC08:17
*** haigang has quit IRC08:38
*** jschwarz has joined #openstack-lbaas08:42
*** kiran-r has quit IRC08:45
*** _kiran_ has joined #openstack-lbaas08:54
*** _kiran_ is now known as kiran-r08:56
*** Miouge has joined #openstack-lbaas09:09
*** kiran-r has quit IRC09:18
*** amotoki_ has joined #openstack-lbaas09:20
*** ajmiller_ has joined #openstack-lbaas09:23
*** ajmiller__ has quit IRC09:26
*** eugene_bag has joined #openstack-lbaas09:32
*** eugene_bag1 has quit IRC09:34
smcgough1is there anyone with jenkins 3rd party CI experience that can tell me why when I "Test Connection" against the gerrit server it succeeds and yet, the status is a red ball?10:23
*** eugene_bag has quit IRC10:31
*** eugene_bag has joined #openstack-lbaas10:40
*** eugene_bag1 has joined #openstack-lbaas10:47
*** eugene_bag has quit IRC10:47
Santosh_ns_Hey smcgough1 i use pygerrit https://pypi.python.org/pypi/pygerrit/0.2.211:17
Santosh_ns_for CI setup11:17
*** eugene_bag1 has quit IRC11:20
*** woodster_ has joined #openstack-lbaas12:01
*** amotoki_ has quit IRC12:20
*** banix has joined #openstack-lbaas12:33
*** amotoki_ has joined #openstack-lbaas12:34
*** eugene_bag has joined #openstack-lbaas12:34
*** amotoki_ has quit IRC12:34
*** eugene_bag1 has joined #openstack-lbaas12:39
*** eugene_bag has quit IRC12:41
*** eugene_bag1 has quit IRC12:55
*** eugene_bag has joined #openstack-lbaas12:56
*** eugene_bag1 has joined #openstack-lbaas13:05
*** eugene_bag has quit IRC13:05
*** banix has quit IRC13:11
*** logan2 has quit IRC13:16
*** logan2 has joined #openstack-lbaas13:19
*** Miouge_ has joined #openstack-lbaas13:20
*** Miouge has quit IRC13:22
*** Miouge_ is now known as Miouge13:22
*** banix has joined #openstack-lbaas13:30
smcgough1Santosh_ns: Thank you!13:49
smcgough1ill give it a go13:50
*** amotoki_ has joined #openstack-lbaas13:51
*** banix has quit IRC13:53
*** logan2 has quit IRC13:53
*** logan2 has joined #openstack-lbaas13:56
*** eugene_bag1 has quit IRC13:59
*** banix has joined #openstack-lbaas14:21
*** enikanorov has quit IRC14:23
*** banix has quit IRC14:26
*** enikanorov has joined #openstack-lbaas14:27
*** kobis has quit IRC14:28
*** banix has joined #openstack-lbaas14:40
*** amotoki_ has quit IRC14:49
*** jorgem has joined #openstack-lbaas14:49
*** eugene_bag has joined #openstack-lbaas14:50
*** jorgem1 has joined #openstack-lbaas14:52
*** Miouge has quit IRC14:52
*** Miouge has joined #openstack-lbaas14:52
*** jorgem has quit IRC14:55
*** eugene_bag1 has joined #openstack-lbaas14:57
*** eugene_bag has quit IRC15:00
*** BrianShang_ has joined #openstack-lbaas15:06
*** BrianShang has quit IRC15:08
*** Varun_Lodaya has joined #openstack-lbaas15:11
*** eugene_bag has joined #openstack-lbaas15:11
*** eugene_bag1 has quit IRC15:11
*** jorgem1 has quit IRC15:11
*** ajmiller has joined #openstack-lbaas15:13
*** kobis has joined #openstack-lbaas15:14
*** jorgem has joined #openstack-lbaas15:14
*** ajmiller_ has quit IRC15:15
*** kobis has quit IRC15:22
*** kobis has joined #openstack-lbaas15:22
*** kobis has quit IRC15:29
*** banix has quit IRC15:34
*** BrianShang has joined #openstack-lbaas15:36
*** BrianShang_ has quit IRC15:37
*** eugene_bag has quit IRC15:40
*** jamiem has joined #openstack-lbaas15:45
*** mestery has quit IRC15:45
*** xgerman has joined #openstack-lbaas15:46
*** Aish has joined #openstack-lbaas15:47
*** bharath has joined #openstack-lbaas16:01
*** mestery has joined #openstack-lbaas16:03
dougwigsmcgough1: what are you using to connect?16:12
dougwigstandard jenkins/zuul, or something else?16:12
TrevorVxgerman Hey!  I'm a little lost about the upstart thing.16:17
openstackgerritAdam Harwell proposed stackforge/octavia: Fix stale pyc issue for unit tests  https://review.openstack.org/16941116:26
*** jorgem has quit IRC16:26
rm_work^^ for anyone who was getting weird results with unit tests after last night's merge16:27
rm_workthat is the issue16:27
rm_workthe in-tree oslo-incubator modules went away, but your pyc files were still there, so they were still loadable16:28
*** Varun_Lodaya has quit IRC16:32
rm_workxgerman: https://review.openstack.org/#/c/140575/ is what you need16:37
rm_workxgerman: but it looks like it needs someone to finish it up16:37
rm_workI might be able to do that16:37
*** sbfox has joined #openstack-lbaas16:43
*** xgerman has quit IRC16:44
*** SumitNaiksatam has quit IRC16:46
*** sbfox has quit IRC16:58
*** sbfox has joined #openstack-lbaas16:59
*** jorgem has joined #openstack-lbaas16:59
*** TrevorV_ has joined #openstack-lbaas17:06
*** SumitNaiksatam has joined #openstack-lbaas17:16
*** eugene_bag has joined #openstack-lbaas17:17
*** bharath has quit IRC17:27
*** bharath has joined #openstack-lbaas17:27
*** bharath has quit IRC17:37
*** bharath has joined #openstack-lbaas17:38
johnsomrm_work You are a working machine man, 2:30?  ugh.  Anyway, thanks for updating the controller worker code17:44
rm_workyeah np17:44
rm_workthat merge caused so many conflicts17:45
johnsomI see that.  I am happy to get that chuck of openstack code out of our repo though.17:45
rm_workblogan: is crc32 in today?17:51
rm_workhe's going to need to fix a number of imports possibly in his code, after that merge last night17:52
openstackgerritMerged stackforge/octavia: DRY'ed keystone session creation and retrieval  https://review.openstack.org/16873617:53
ptoohillwhos that talking now, i cant see names17:53
ptoohilli need to redo this box, or get one of those fancy dells :P my comp starts blinking/freaking out while running vidyo18:00
openstackgerritMichael Johnson proposed stackforge/octavia: Implements Octavia Controller Worker  https://review.openstack.org/15149618:02
rm_workLinux Vidyo?18:03
rm_workor OSX?18:03
rm_workyeah i have heard it has some issues18:03
rm_workvideo chat in linux has never been great T_T18:03
ptoohillgraphics drivers on this box already dont work with ubuntu18:03
johnsomhi ho hi ho it's off to re-basing we go18:03
openstackgerritAdam Harwell proposed stackforge/octavia: Implemented Queue Consumer  https://review.openstack.org/14978918:07
*** jorgem has quit IRC18:07
*** jschwarz has quit IRC18:14
*** TrevorV_ has quit IRC18:25
openstackgerritOpenStack Proposal Bot proposed stackforge/octavia: Updated from global requirements  https://review.openstack.org/16945618:40
johnsomOnce that goes through the gate I will push it through18:48
openstackgerritPhillip Toohill proposed openstack/neutron-lbaas: WIP: Neutron LBaaS: TLS Barbican Scenario Test  https://review.openstack.org/16482818:57
*** TrevorV_ has joined #openstack-lbaas19:04
*** TrevorV_ has quit IRC19:04
bloganrm_work: he is now19:08
rm_workk, you going to help him fix imports?19:09
bloganmy linux vidyo worked fine19:09
bloganother than the fact my camera doesn't work but thats no vidyo's fault19:09
bloganrm_work: he's about to leave19:09
rm_workblogan: yeah I meant "linux video" not "linux vidyo" :P19:09
rm_worki mean, one has issues because of the other19:10
rm_workvideochat in general in linux has issues due to bad camera support in most cases T_T19:10
openstackgerritmin wang proposed openstack/neutron-lbaas: Introduced tempest API tests for openstack/neutron-lbaas  https://review.openstack.org/16843919:10
*** Aish has quit IRC19:15
johnsomHmm, so the py27 test passes for me local, will have to look at that gate issue after lunch.19:18
bloganbut yes i will when he gets to it19:18
*** sbfox has quit IRC19:19
openstackgerritMerged stackforge/octavia: Updated from global requirements  https://review.openstack.org/16945619:26
*** sbfox has joined #openstack-lbaas19:26
*** SumitNaiksatam has quit IRC19:33
*** SumitNaiksatam has joined #openstack-lbaas19:44
rm_workjohnsom: https://review.openstack.org/16941119:59
rm_workjohnsom: try that19:59
*** Aish has joined #openstack-lbaas20:01
*** xgerman has joined #openstack-lbaas20:06
bloganany of yall remember the use case for us to allow load balancing across multiple networks?20:06
xgermanso you are talking members on multiple networks?20:11
bloganxgerman: yes20:12
blogani have a feeling none of us have a real use case because it was requested at the summit20:13
bloganmaybe dougwig does since he's a vendor20:13
bloganadn vendor's like to do all the thinsg20:13
xgermanwell, I think bluebox wanted it, too20:13
bloganwell if sbalukoff would actaully not be rude and say hello to us instead of idling in channel20:13
sbalukoffI really am pretty rude, aren't I?20:14
bloganyou hsould change your name to ghost_of_sbalukoff20:14
bloganand yes you are!20:14
sbalukoffAnd actually, that's not a feature we needed.20:14
dougwigblogan: yes, very common use case.20:14
blogandougwig: coudl you explain the use case in practical terms20:14
sbalukoffI think it was one that came from one of the feature request spreadsheets that was put together like a year ago...  maybe ebay needed it?20:15
dougwigin a meeting.  i can later, yes20:15
blogandougwig: priorities20:15
blogansbalukoff: yeah i remember it beign requested, i just can't think of a use case, and our engineers are looking for use cases20:15
sbalukoffI think it would be a pretty unusual use case... but I also seem to recall trying to design things to support it.20:16
sbalukoffI mean...20:16
blogani think its unusual for a cloud load balancer, dedicated sure20:16
bloganbut dougwig will have to enlighten all of us20:16
dougwigyou don't want to let your customers use a single LB for vm's in multiple of your data centers?20:17
sbalukoffThe use case that I can see is if you have members of a pool split up geographically, but want all of them in a single pool in case a whole site goes down or something.20:17
sbalukoffSo yeah, what dougwig said.20:17
dougwigor what if i have legacy VMs and just want to add an LB?  force me to move?20:17
sbalukoff(BTW, I *think* I will be in Vancouver...  the person who does travel plans was asking me stuff last week about it, which is usually a good indication.)20:18
bloganwhere would the LB reside if it can load balance across regions?20:18
blogansbalukoff: its just up the road, why couldnt you do it20:18
sbalukoffblogan: Stupid deadlines.20:19
sbalukoffblogan: Also a big part of the reason I haven't been active here at all.20:19
blogansbalukoff: i figured you just hated us20:19
sbalukoffWell, there's that, too.20:19
sbalukoffI especially can't stand dougwig.20:19
sbalukoffI don't know why you haven't kicked him out yet.20:20
dougwigi haven't figured that out yet either.20:20
rm_workholyshit it's sbalukoff20:20
blogansbalukoff: its all part of the grand scheme, he's feeling safe right now...which makes his fall all the better to witness20:21
sbalukoffSince I'm here:  One other thing to bring up--  the US PTO has given us a few months to submit paperwork proving that we're using the Octavia name in some important way. I have no idea what this entails just yet, but we've got until mid July to submit this paperwork.  So if y'all want to keep that name, eventually we'll have to figure that out. :)20:22
sbalukoffblogan: Oh, good! I can totally approve of something that Machiavellian.20:22
sbalukoff(Just wanted to make y'all aware of the trademark thing. Carry on, eh!)20:23
sbalukoffOk, I've gotta run off again. See ya!20:23
xgermanok, that shouldn't be too difficult just point them to the repos and the talk ;-)20:24
sbalukoffxgerman: I suspect it's going to be about that easy.20:24
*** BrianShang_ has joined #openstack-lbaas20:24
xgermanwell, can't your lawyers talk to their lawyers20:24
*** BrianShang has quit IRC20:27
sbalukoffxgerman: I have had no time to pursue this at all.20:29
sbalukoff(And the deadline isn't close yet. XD )20:29
blogansbalukoff: i definitely dont want to go through the pain of renaming20:32
blogansbalukoff: but if we had to, i could use it as good reason to rename amphora...20:32
* blogan waits for the mob20:32
* rm_work pulls out a torch20:33
openstackgerritMerged stackforge/octavia: Fix stale pyc issue for unit tests  https://review.openstack.org/16941120:35
rm_workjohnsom: ok, now that it merged, i think your issue should go away20:36
rm_work99% of "works local, not in devstack" are caused by stale pyc files20:37
johnsomrm_work my issue is the way the new keystone config stuff was done.  oslo_fixture is not setting the keystone version for some reason even though I have added20:38
johnsomconf.config(group="keystone_authtoken", auth_version=AUTH_VERSION)20:38
rm_workah, hmm weird20:38
rm_workbut it passes locally?20:38
rm_workagain, i would have assumed "because the old pyc files"20:38
johnsomSeems to be test order dependent20:38
johnsomI did a tox -e py27 -- --until-failure to reproduce20:39
johnsomtestr is kind enough to randomize the test execution order us, so sometimes if the tests have an issue they pop up randomly20:50
openstackgerritGerman Eichberger proposed stackforge/octavia: Implements the haproxy amphora agent api server  https://review.openstack.org/16003421:03
bloganjohnsom: your tests aren't able to read auth_version from the config? is taht the issue?21:03
johnsomYeah, the override for the value.  I'm digging.  I am not very familiar with the import method you used so I think the override above is wrong21:05
blogancant you just do cfg.CONF.set_override?21:06
johnsomI think that would not "undo" after the test method like the oslo_fixture stuff does.21:09
dougwigfyi, l7 and flavors specs are in the gate, approved for liberty.21:12
xgermanyeah!! that will make jamiem happy :-)21:14
jamiemgood to know.21:14
xgermanrm_work: how did you salve the error involving log.log_opts ?21:14
rm_workwhich error21:15
xgermanyou should have run into that with the queue worker21:15
rm_worki think i know, sec21:16
xgermanthat's why I am asking ;-)21:16
*** sbfox has quit IRC21:16
rm_workslow computer21:17
xgermanno kidding.. mine has 8 GB and usually a loas of 421:19
xgermanload of 421:19
rm_worklooks like21:22
rm_worklog.log_opts -> log._options.log_opts21:23
rm_workbut that seems wonky since _options is "private" kinda21:23
*** sbfox has joined #openstack-lbaas21:26
*** SumitNaiksatam has quit IRC21:26
*** sbfox has quit IRC21:38
ptoohillbedis: you around?21:39
*** sbfox has joined #openstack-lbaas21:42
*** eugene_bag has quit IRC21:58
ptoohillWould it be best to ensure the user created an un-encrypted private key container and simply load that configuration, or allow the user to create an encrypted private-key/passphrase container and have a mechinism to 'unlock' it before we save the configuration?22:00
rm_workptoohill: i assumed we'd unlock it22:00
ptoohillwould this util be part of the cert_parser ?22:01
rm_workptoohill: which you can do by loading and then dumping the key again (without a passphrase)22:01
rm_workor something22:01
rm_workwe'd want to do it right before sending it over22:01
rm_workwell, i guess before the driver layer22:02
ptoohillyou mean before saving it?22:02
ptoohillthe data doesnt get sent over any q's or anything, so for example, the data is loaded in the driver in neutron lbaas, then saved22:03
ptoohilland, in octavia, the driver also loads the certificate data, i would assume we would make an additional call to unlock it before it is sent to host/saved?22:04
ptoohillthis is currently not being done in neutron-lbaas/octavia, something i wasnt too focused on and overlooked. Ill need to make patches this evening.22:06
rm_workwait yeah, are we talking about octavia or n-lbaas?22:06
ptoohillboth, im mostly concerned with ensuring everything was done right for TLS in n-lbaas atm22:06
rm_workptoohill: so yeah, for n-lbaas, were we actually loading it up into objects or were we just grabbing the strings from barbican?22:07
rm_workptoohill: the verification code loads into objects i think by design, so around that point I guess we could strip the passphrase22:07
ptoohillbut, since a lot of the logic is the same between the two i was assuming once i had this solid octavia would be a quick follow22:07
ptoohillyou mean during validation in plugin?22:08
ptoohillbecause, its not till the driver/jinja template that the actual data is used/saved22:08
rm_worki guess i am saying22:09
rm_workit doesn't matter much22:09
rm_workthe operation is super simple22:09
rm_workdo it wherever it makes sense or seems easiest22:09
*** SumitNaiksatam has joined #openstack-lbaas22:14
bloganjohnsom: you could always undo it with a cleanup method22:15
bloganjohnsom: but dont fully understand the issue so i could be wrong22:16
*** eugene_bag has joined #openstack-lbaas22:26
johnsomblogan: I have been using: conf = oslo_fixture.Config(cfg.CONF) and conf.config(group="controller_worker", amp_flavor_id=AMP_FLAVOR_ID) for the tests, but for some reason the auth_version isn't working.22:38
johnsomIf I go too long I will use your method and just stomp on it (which it why the test sometimes works if your test runs first)22:38
bloganso conf.config(group='keystone_authtoken', auth_version='2') wouldn't work?22:39
johnsomYeah, I have conf.config(group="keystone_authtoken", auth_version=AUTH_VERSION) with AUTH_VERSION = '2' above22:40
blogandoes it say unrecognizeable group or option?22:40
johnsom Exception('Unknown keystone version!')22:40
johnsomNope, runs just fine.22:40
bloganprobably bc auth_version is None22:40
blogancan you debug and see what the auth_version is?22:41
johnsomI'm fairly new to this oslo_fixtures stuff.  I'm sure it's something simple I have messed up22:43
bloganhonestly ive never used the oslo_fixture stuff, i've always just used set_override22:44
blogani can't even find any documentation on it22:44
blogani see oslo_config.fixture.COnfig22:44
bloganoh yeah tahts the one i found22:45
bloganbut its not oslo_fixture22:45
johnsomBut, yeah, it changed to oslo_config last night22:45
blogando you need to register the options first?22:46
johnsomThat is the only difference I see, the other configs have cfg.CONF.register_opts(controller_worker_opts, group='controller_worker') in config.py but you are using cfg.CONF.import_group22:49
bloganyeah so you mean have to import keystonemiddleware directly then and get a hold of those options22:50
*** devlaps has joined #openstack-lbaas22:50
bloganyeah that fixes it22:52
bloganfrom keystonemiddleware import auth_token22:52
bloganconf.config(group='keystone_authtoken', auth_version='2')22:52
bloganOut[13]: '2'22:52
rm_workugh that's weird22:52
bloganonly problem is the import auth_token will be unused, which is hwy i used import_group, but you should be able to just comment with # noqa after the import so pep8 ignores it22:53
rm_workyeah # noqa22:53
rm_workbut bleh22:53
*** devlaps has quit IRC22:54
bloganhmm i was expecting ot have to call conf.register_opts though22:54
johnsomIs there a reason to do it this way rather than how we are defining all of the other options in config.py?22:56
bloganyou mean import_group?22:56
bloganor registering options defined in keystonemiddleware?22:57
johnsomusing import_group vs reg_ipts22:58
bloganwell without import_group we'd just have to import the auth_token module to get those options registered, but it would then be an unused import bc thats all we needed22:59
bloganredefining the keystonemiddleware options is not a good idea either23:00
*** eugene_bag has quit IRC23:00
*** sbfox has quit IRC23:11
*** jamiem has quit IRC23:14
*** Aish has quit IRC23:14
*** xgerman has quit IRC23:15
*** bharath has quit IRC23:33

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