Monday, 2014-10-27

*** mestery has quit IRC00:02
*** HenryG has quit IRC00:41
*** HenryG has joined #openstack-lbaas00:46
*** HenryG has quit IRC01:24
*** mestery has joined #openstack-lbaas01:28
*** mestery has quit IRC01:28
*** mestery has joined #openstack-lbaas01:29
*** woodster_ has joined #openstack-lbaas01:30
*** HenryG has joined #openstack-lbaas01:32
*** mestery has quit IRC01:40
*** mestery has joined #openstack-lbaas01:41
*** vivek-ebay has joined #openstack-lbaas02:52
*** mestery has quit IRC03:08
*** mestery has joined #openstack-lbaas03:09
*** vivek-ebay has quit IRC03:11
*** vivek-ebay has joined #openstack-lbaas03:12
*** vivek-ebay has quit IRC03:17
*** vivek-ebay has joined #openstack-lbaas03:27
*** vivek-ebay has quit IRC03:42
*** vivek-ebay has joined #openstack-lbaas03:43
*** vivek-ebay has quit IRC04:06
*** rm_you has quit IRC05:05
*** rm_you has joined #openstack-lbaas05:06
*** vivek-ebay has joined #openstack-lbaas05:30
*** vivek-ebay has quit IRC05:34
*** vivek-ebay has joined #openstack-lbaas05:54
*** ksamoray has joined #openstack-lbaas06:04
*** enikanorov_ has joined #openstack-lbaas06:10
*** amotoki has joined #openstack-lbaas06:28
*** jschwarz has joined #openstack-lbaas06:29
*** jschwarz has quit IRC06:38
*** vivek-ebay has quit IRC06:54
*** woodster_ has quit IRC07:00
*** vivek-ebay has joined #openstack-lbaas07:55
*** vivek-ebay has quit IRC07:59
*** jschwarz has joined #openstack-lbaas08:15
*** jschwarz has quit IRC08:16
*** jschwarz has joined #openstack-lbaas08:18
*** ksamoray has left #openstack-lbaas08:27
*** ksamoray has joined #openstack-lbaas08:28
*** amotoki has quit IRC09:01
*** ksamoray has quit IRC09:17
*** kobis has joined #openstack-lbaas09:18
*** itzikb has joined #openstack-lbaas11:31
itzikbHI, regarding it seems like the patch doesn't solve the issue. Still happens.11:59
itzikbAny idea why the IP address field for the VIP is mandatory?11:59
*** openstackgerrit has quit IRC12:19
*** openstackgerrit has joined #openstack-lbaas12:20
*** openstackgerrit has quit IRC12:34
*** enikanorov has quit IRC12:35
*** openstackgerrit has joined #openstack-lbaas12:35
*** enikanorov has joined #openstack-lbaas12:35
*** vivek-ebay has joined #openstack-lbaas12:38
*** vivek-ebay has quit IRC12:42
*** woodster_ has joined #openstack-lbaas12:55
*** enikanorov has quit IRC13:05
*** enikanorov has joined #openstack-lbaas13:06
*** mestery_ has joined #openstack-lbaas13:34
*** mestery has quit IRC13:37
*** ptoohill_ has joined #openstack-lbaas14:23
*** mestery_ is now known as mestery14:29
ptoohillitzikb, that bug is showing that its fixed and it seems it was just a Horizon error? On that note, im unsure we have any Horizon specialist in this channel as we have been highly focused on current development of LBaaS v214:43
ptoohillnvm (lack of coffee), not sure why it isnt fixed nor why the vip address is required. You may have better luck in the horizon channel regarding this patch.14:51
itzikbptoohill: Thanks14:58
*** xgerman has joined #openstack-lbaas15:01
*** TrevorV_ has joined #openstack-lbaas15:30
*** ptoohill_ has quit IRC15:34
*** ptoohill_ has joined #openstack-lbaas15:36
*** TrevorV_ has quit IRC15:42
*** itzikb has quit IRC15:50
*** mlavalle has joined #openstack-lbaas15:53
*** jschwarz has quit IRC15:58
*** jschwarz has joined #openstack-lbaas15:58
*** jorgem has joined #openstack-lbaas16:02
*** fnaval has joined #openstack-lbaas16:13
*** vivek-ebay has joined #openstack-lbaas16:38
*** mikedillion has joined #openstack-lbaas16:41
*** mikedillion has quit IRC16:41
*** kobis has quit IRC16:49
*** kobis has joined #openstack-lbaas16:49
*** kobis has quit IRC16:53
*** sbfox has joined #openstack-lbaas16:57
*** sbfox1 has joined #openstack-lbaas17:23
*** sbfox has quit IRC17:23
*** sbfox has joined #openstack-lbaas17:25
*** sbfox1 has quit IRC17:25
*** sbfox has quit IRC17:28
*** markmcclain has joined #openstack-lbaas18:14
*** sbfox has joined #openstack-lbaas18:21
*** jschwarz has quit IRC18:30
rm_workdougwig / sbalukoff: any idea how to convince tox to let me include ,diag files in the spec CRs?18:39
rm_workerr, .diag18:39
rm_work is failing py26/27 because the test for the filenames fails18:40
rm_workI can have it do filename.endswith(".rst") or filename.endswith(".diag")18:40
rm_workand do self._check_titles(titles) conditionally on .rst18:40
rm_workbut if there is a proper way...18:40
sbalukoffrm_work: I seem to recall having to do something similar to get .dot files to work. Let me revisit that...18:42
rm_worki see that seqdiag stuff is registered (at the top of the same test file)18:45
sbalukoffFor .dot files, I simply excluded them from testing because there wasn't a good syntax checker for the same. This was done in specs-tests/tests/ toward the end of the file.18:45
rm_workbut that test is specifically checking filenames18:45
rm_workyeah that's the one...18:45
rm_workah yes18:45
rm_workfiles = files - set(glob.glob('specs/*/*.dot'))18:45
rm_worki like that syntax, lol18:46
rm_workdidn't know you could just... subtract a set from a set18:46
rm_workmakes sense18:46
rm_worki guess really it's that I never use set()18:46
sbalukoffSeemed to be a good use of set() in this case. :)18:47
sbalukoff(Speaking as someone who comes from the perl world and would have done this using hashes functioning as sets. It's nice that python forgoes that nonsense and just allows you to use set operations.)18:47
bloganpython >>>>>> perl18:49
bloganthen again anything > perl18:49
dougwigruby > python18:49
* dougwig ducks18:49
* ptoohill swings and misses18:50
blogani will be the bigger man and ignore that comment18:50
bloganand i am the bigger man unfortunately :(18:52
blogandougwig what time is the neutron meeting today?19:05
bloganwait it alternates between monday and tuesday so it msut be at 2100 utc then19:05
*** sbfox has quit IRC19:10
dougwigIt's Monday this week.  2pm your time.19:15
dougwigNo, 4pm your time19:16
*** sbfox has joined #openstack-lbaas19:22
*** woodster_ has quit IRC19:43
*** woodster_ has joined #openstack-lbaas19:45
*** vivek-ebay has quit IRC19:54
*** ptoohill_ has quit IRC20:04
*** ptoohill_ has joined #openstack-lbaas20:07
openstackgerritTrevor Vardeman proposed a change to stackforge/octavia: Nova virtual machine driver spec
*** vivek-ebay has joined #openstack-lbaas20:21
*** ctracey has quit IRC20:26
*** ctracey has joined #openstack-lbaas20:27
TrevorVxgerman sballe have you guys heard from al miller?20:43
xgermanI see him every day20:44
TrevorVxgerman could you have him jump in IRC for a bit?20:44
*** mwang2 has joined #openstack-lbaas20:46
*** ajmiller has joined #openstack-lbaas20:46
xgermanI had his office mates poke him :-20:46
TrevorVsounds great, thanks!20:46
ajmillerTrevorV, I hear I'm being pinged20:47
ajmillerwhat's up?20:47
mwang2this is Min, what can we do for you TrevorV?20:47
TrevorVHey mwang2, its specifically about, which is ajmiller, but thank you!20:48
bloganxgerman's summoning powers are amazing20:48
TrevorVI was curious if you had some progress on the next patch set here, ajmiller20:48
TrevorV2 for 1 special in his repotoire20:48
johnsom_blogan +120:49
ajmillerYes, I worked on that last week, but didn't get the patch posted.20:49
ajmillerI will get that done ASAP, this afternoon.20:49
TrevorVAlright, I have a WIP review dependent on it, and then a WIP review on THAT one, and then etc etc ha ha, so just curious if you had something in the works.  Thanks for the update!  I look forward to more reviewing!20:49
bloganwe should get some eyes on this review, its pretty simple
bloganjust need another +220:51
*** TheFutonEng has joined #openstack-lbaas20:52
TrevorVWell, I have no +2 powers, otherwise I'd jump right on this :P20:52
blogandoesn't prevent you from looking at it and giving +120:53
bloganor -120:53
bloganmight find something i missed20:53
TrevorVYeah, I was about to do that20:54
bloganits not like csgo where i don't miss20:55
TrevorVcept you do20:55
blogannot that i recollect20:55
TrevorVI mean, you block out your own errors, which is understandable, but you'll never grow at that rate20:55
bloganhard to block out what didn't happen20:56
TrevorVCase in point my friend.20:56
bloganlink or it didnt happen20:57
TrevorVI'll record the next time you suck.20:57
TrevorVI promise :)20:57
dougwigblogan: approved20:58
bloganthanks dougwig20:58
bloganfigured it'd be nice to get a review in from someone who looks like they are just trying to contribute20:58
*** dboik has joined #openstack-lbaas20:59
dougwigneutron meeting starting now, and talking about adv. services spinout might be on the agenda!20:59
*** jorgem has quit IRC21:00
*** barclaac has joined #openstack-lbaas21:06
*** vivek-ebay has quit IRC21:13
openstackgerritChad Lung proposed a change to stackforge/octavia: PEP8 fixes, spelling fixes
*** vivek-ebay has joined #openstack-lbaas21:16
sballeI need the +1's back : :-) Thanks21:20
openstackgerritAl Miller proposed a change to stackforge/octavia: Initial draft of compute driver spec.
ajmillerTrevorV ^^21:30
dougwigneutron lightning talks (5 minutes, just to introduce an idea):
dougwigdoes someone want to do a 5 minute intro of octavia as the new reference lbaas?21:31
rm_workin the neutron meeting?21:32
rm_workblogan hopefully21:32
dougwigno, at the summit.21:33
rm_workthough I didn't realize until I read his email today (the one he sent a few days ago) that he meant to actually split the API out into another project again21:33
rm_workah, blogan hopefully :P21:33
blogansplitting the api is an option, as is keeping it with octavia21:34
bloganas is just spinning out with advanced services21:34
dougwigorthogonal issue to just presenting octavia to the openstack networking community21:34
blogananyway i think sbalukoff would be best for a lightning talk on octavia21:34
blogani would mumble and wet myself and embarass everyone21:34
dougwigas long as one of us hooks him off stage at the 5 minute mark, it'll be good.21:35
bloganheckle him with incubator chants21:35
blogani was going to say introducing octavia might be a bit too early, but then again its a lightning talk and thats a good use of it21:36
rm_worki was going to do a lightning talk :(21:37
openstackgerritA change was merged to stackforge/octavia: PEP8 fixes, spelling fixes
bloganyou were but then rax stole your thunder21:38
blogansee what i did there21:38
dougwigi just signed up sbalukoff.21:38
blogandid you give alot him 10 mins?21:39
bloganim not sure any of the spin out talk is going to happen with how long the current discussion is going21:39
xgermanwe can always vote with our feet IHMO21:42
*** dboik has quit IRC22:00
sballeblogan: +10022:00
bloganhopefully that was just the attention grabber, and we can get some real decisions made in paris22:01
bloganim pretty sure its going to require a lot of side talks with various people22:01
sballethat's waht we are good at22:01
bloganmestery: who is in charge of creating an agenda for the spin out meeting? or is it just going to be an open discussion based on the etherpad?22:02
dougwigwe should go in with a plan.  v2 in feature branch by date X, prototype a split by date Y, standalone and deprecated by date Z.22:02
blogandougwig: we sitll need to decide if we want to split with advanced services (if they split too)22:02
dougwigsimultaneously talk with the guys and see about leverage.22:02
mesteryblogan: Mark McClain is moderating that one, but PLEASE add things to the etherpad22:03
dougwigblogan: true, but that just means we go in with multiple plans.22:03
mesteryIt's a community effort, the leader is just amo derator22:03
bloganokay thats what i was looking for22:03
blogansorry to hear you wont be there, but its for a great reason22:03
dougwigi'd be happy just coding a split in K to find the missing pieces/interfaces.22:03
blogansame here22:03
bloganbc i think that is the only real way to find them22:04
dougwigmaybe a few of us even having a mini-split-meetup, add red bull, find some answers.22:04
bloganohter than meticulously going through code and diagramming and even then not fully confident you foudn them all22:04
sballecould you give me the link to the etherpad?22:04
bloganim going to add the octavia api to that then22:05
bloganunless there are objections22:05
xgermanblogan also send that to jpipes if younhaven;t done so22:05
dougwigblogan: quick objection22:05
bloganxgerman: send what?22:05
blogandougwig: objection overruled22:05
blogandont want to start that talk?22:05
xgermanblogan, you alreday gave Jay the API link. Scratch that22:06
dougwigthe community spent a lot of time on lbaas v2 consensus.  octavia deviates from that API.  are we confident that enough of the community that was involved in v2 was part of that changing interface view that it won't cause problems?  (and if so, shouldn't we iterate the neutron feature branch to more closely resemble octavia) ?22:06
bloganxgerman: haha np22:06
xgermandougwig +1 -- yeah, iterating good22:07
bloganneutron extensions are incapable of having that deep of nested resources22:07
bloganthat can be fixed, but they're going to be overhauling the extensions soon anyway22:07
dougwigi meant more the model associations, required root objects, etc.22:07
bloganso that in itself causes problems22:07
dougwigwouldn't getting rid of the notion of "everything is root object" actually help the neutron code?  we do a lot of gyrations for that.22:07
bloganyes i think it would help, but extensions just won't allow more than a 2nd level resource22:08
dougwigeh, that's syntax.  there are ways to represent all these objects without nesting, in a more 1:1 fashion than now.22:08
bloganyeah there is but it just doesn't feel right22:09
dougwigfeels more wrong than everything being a root object?22:09
blogandont play the lesser of two evils card!22:09
dougwigand by root, i mean *real* root, not just appears to be so at the URI level.22:09
bloganwell the real root is the load balancer, but the listener is what ties everything together22:10
ptoohillwe could expand on the resources to fit Octavia's model a bit better but within the Neutron constraints and have yet another layer that converts this for us?22:11
dougwigi'm actually arguing to tweak the neutron interface to make the translation to octavia *easier* here, which strengthens the 'octavia as user api' argument.22:12
bloganwell the models are already similar, its just some associations are different and the URIs are different22:12
sballedougwig: that make sense to me22:12
dougwigi also wonder if the networking umbrella should share a db, or everyone does their own.22:12
blogani think that should be as optional as possible22:13
ptoohillWould be nice to not be coupled to other services, was this brought up at all as part of these talks?22:14
ptoohillfor adv-svc22:15
blogani don't see the octavia API being drastically different, other than the constraints the nested URIs entail22:15
bloganptoohill what do you mean?22:15
ptoohillhaving separate db's22:16
bloganoh okay22:16
ptoohilli can see both sides to that argument22:16
dougwigi have to run to the airport (denver tonight!).  i'll check in in a bit.22:16
blogani gotta go for a bit too22:16
bloganwe should continue this22:16
sbalukoffFWIW, I don't like the idea of sharing DBs. Not unless you're sharing code, too. (Which I don't think we should.)22:31
dougwigI don't either.  But I'm also not too fond of having 100 wsgi servers each on their own port.22:33
sballesbalukoff: dougwig +1 on not sharing22:36
rm_workyeah don't think a shared DB is a good route >_>22:47
rm_workdougwig: well, in a production deployment, those 100 wsgi servers won't be on the same box -- pretty much each service will have their own infrastructure. I'd see that as a PLUS actually, we don't have to be stuck on the neutron infra22:48
dougwigI knew someone would come back with that counter.  :)22:49
rm_workdo you disagree with it?22:49
rm_workwell then :)22:50
*** TheFutonEng has quit IRC22:54
rm_workhey dougwig sbalukoff blogan could use a +2 on the first change here22:55
openstackgerritAdam Harwell proposed a change to stackforge/octavia: TLS Data Security Overview
openstackgerritAdam Harwell proposed a change to stackforge/octavia: Allow .diag file extensions in spec reviews
rm_workerr, second change22:55
rm_workthe diag extension change22:55
rm_workI mean, I guess you can wait for jenkins :P22:55
*** markmcclain has quit IRC23:13
*** xgerman has quit IRC23:20
*** vivek-ebay has quit IRC23:39
*** vivek-ebay has joined #openstack-lbaas23:39
*** vivek-ebay has quit IRC23:39
*** vivek-ebay has joined #openstack-lbaas23:40
*** ajmiller has quit IRC23:41

Generated by 2.14.0 by Marius Gedminas - find it at!