Thursday, 2015-10-15

*** doug-fish has quit IRC00:00
*** ducttape_ has quit IRC00:03
*** apuimedo has quit IRC00:14
*** yamamoto has quit IRC00:21
*** diogogmt has quit IRC00:21
*** yamamoto has joined #openstack-lbaas00:22
*** apuimedo has joined #openstack-lbaas00:26
rm_worksbalukoff: had to run earlier, injured myself doing a "quick" household task ... but i also am planning to basically spend the whole summit on architectural discussions for "moving forward", as not many of the talks actually interest me00:34
rm_worki'm primarily there for the planning and such for our stuff00:34
rm_workso if we can find rooms00:34
rm_worki'm down for a 4 day design summit...00:34
*** SumitNaiksatam has joined #openstack-lbaas00:40
*** SumitNaiksatam has quit IRC00:40
*** bharathm has quit IRC00:44
*** bharathm has joined #openstack-lbaas00:47
*** Alex_Stef has quit IRC00:51
*** minwang2 has quit IRC00:52
*** minwang2 has joined #openstack-lbaas00:53
*** fnaval has quit IRC00:56
*** minwang2 has quit IRC00:58
*** bharathm has quit IRC01:00
*** bharathm has joined #openstack-lbaas01:03
*** Alex_Stef has joined #openstack-lbaas01:06
*** Alex_Stef has quit IRC01:15
*** apuimedo has quit IRC01:16
*** mestery has quit IRC01:18
*** Brian_shang has quit IRC01:20
*** Brian_shang has joined #openstack-lbaas01:20
*** bharathm has quit IRC01:20
*** bana_k has quit IRC01:22
*** Alex_Stef has joined #openstack-lbaas01:30
*** Alex_Stef has quit IRC01:35
*** ducttape_ has joined #openstack-lbaas01:43
*** bana_k has joined #openstack-lbaas01:44
*** Brian_shang has quit IRC01:50
*** Brian_shang has joined #openstack-lbaas01:50
*** Alex_Stef has joined #openstack-lbaas01:52
*** diogogmt has joined #openstack-lbaas01:57
sbalukoff1rm_work: Awesome! If not finding a room, finding a quiet corner works for me, too.02:01
*** Alex_Stef has quit IRC02:02
*** Alex_Stef has joined #openstack-lbaas02:17
*** ducttape_ has quit IRC02:26
*** ducttape_ has joined #openstack-lbaas02:27
*** yamamoto has quit IRC02:31
*** Alex_Stef has quit IRC02:32
*** Alex_Stef has joined #openstack-lbaas02:47
*** vivek-ebay has quit IRC02:50
*** KunalGandhi has quit IRC02:50
*** Brian_shang has quit IRC02:51
*** Brian_shang has joined #openstack-lbaas02:51
*** Alex_Stef has quit IRC03:01
*** yamamoto has joined #openstack-lbaas03:15
*** Alex_Stef has joined #openstack-lbaas03:16
*** ducttape_ has quit IRC03:26
*** diogogmt has quit IRC03:27
*** Alex_Stef has quit IRC03:28
openstackgerritBrandon Logan proposed openstack/neutron-lbaas: Add ability for octavia to allocate vip
*** Alex_Stef has joined #openstack-lbaas03:43
*** bharathm has joined #openstack-lbaas04:21
*** bharathm has quit IRC04:25
*** minwang2 has joined #openstack-lbaas04:35
*** amotoki has quit IRC04:57
*** amotoki_ has joined #openstack-lbaas04:57
*** minwang2 has quit IRC05:22
*** minwang2 has joined #openstack-lbaas05:30
*** michchap has joined #openstack-lbaas05:44
michchapHi channel. I have recently upgraded from Juno to Kilo and have active lbaas v1 resources. If I swap the driver over to v2, will those resources continue to be managed? If not, is there a migration procedure?05:51
*** bana_k has quit IRC06:07
*** macjack has joined #openstack-lbaas06:12
*** apuimedo has joined #openstack-lbaas06:18
*** bana_k has joined #openstack-lbaas06:18
*** bana_k has quit IRC06:25
*** bana_k has joined #openstack-lbaas06:26
*** numans has joined #openstack-lbaas06:30
*** kiran-r has joined #openstack-lbaas06:30
*** amotoki has joined #openstack-lbaas06:32
*** amotoki has quit IRC06:32
*** apuimedo has quit IRC06:36
*** minwang2 has quit IRC06:38
*** armax has quit IRC06:44
*** bana_k has quit IRC07:01
openstackgerritBrandon Logan proposed openstack/neutron-lbaas: Add ability for octavia to allocate vip
*** kobis has joined #openstack-lbaas07:09
*** St3F_A13x has joined #openstack-lbaas07:14
*** Alex_Stef has quit IRC07:16
*** evgenyf has joined #openstack-lbaas07:22
*** mhickey has joined #openstack-lbaas07:27
*** St3F_A13x has quit IRC07:30
*** KunalGandhi has joined #openstack-lbaas07:40
*** KunalGan_ has joined #openstack-lbaas07:45
*** KunalGandhi has quit IRC07:46
*** St3F_A13x has joined #openstack-lbaas07:47
*** amotoki_ has quit IRC07:55
*** amotoki has joined #openstack-lbaas08:06
*** St3F_A13x has quit IRC08:06
*** jschwarz has joined #openstack-lbaas08:07
*** KunalGan_ has quit IRC08:11
*** St3F_A13x has joined #openstack-lbaas08:23
*** kobis has quit IRC08:33
*** St3F_A13x has quit IRC08:34
*** kobis has joined #openstack-lbaas08:42
*** amotoki has quit IRC08:43
*** St3F_A13x has joined #openstack-lbaas08:49
*** amotoki has joined #openstack-lbaas08:53
*** kiranr has joined #openstack-lbaas08:57
*** kiran-r has quit IRC08:57
*** amotoki has quit IRC09:03
*** evgenyf has quit IRC09:48
*** evgenyf has joined #openstack-lbaas10:34
*** numans has quit IRC10:53
*** kiranr has quit IRC10:59
*** numans has joined #openstack-lbaas11:06
*** ducttape_ has joined #openstack-lbaas11:38
*** Brian_shang has quit IRC11:47
*** ducttape_ has quit IRC11:54
*** numans has quit IRC12:01
*** yamamoto has quit IRC12:08
*** numans has joined #openstack-lbaas12:14
*** bradjones has joined #openstack-lbaas12:16
*** bradjones has quit IRC12:16
*** bradjones has joined #openstack-lbaas12:16
*** doug-fis_ has quit IRC12:18
*** doug-fish has joined #openstack-lbaas12:18
*** amotoki has joined #openstack-lbaas12:26
*** ducttape_ has joined #openstack-lbaas12:36
*** ducttape_ has quit IRC12:39
*** mestery has joined #openstack-lbaas12:45
*** Brian_shang has joined #openstack-lbaas12:48
*** Brian_shang has quit IRC12:52
*** yamamoto has joined #openstack-lbaas12:54
openstackgerritOpenStack Proposal Bot proposed openstack/neutron-lbaas: Merge tag '7.0.0'
*** amotoki has quit IRC13:00
*** mestery has quit IRC13:00
*** mestery has joined #openstack-lbaas13:01
*** Brian_shang has joined #openstack-lbaas13:03
*** Brian_shang has quit IRC13:04
*** Brian_shang has joined #openstack-lbaas13:05
*** Brian_shang has quit IRC13:07
*** amotoki has joined #openstack-lbaas13:12
*** Brian_shang has joined #openstack-lbaas13:12
*** St3F_A13x has quit IRC13:13
*** ducttape_ has joined #openstack-lbaas13:13
*** Brian_shang has quit IRC13:16
*** Brian_shang has joined #openstack-lbaas13:19
*** jschwarz has quit IRC13:25
*** jschwarz has joined #openstack-lbaas13:25
*** St3F_A13x has joined #openstack-lbaas13:27
*** ducttape_ has quit IRC13:28
*** St3F_A13x has quit IRC13:31
*** yamamoto has quit IRC13:34
*** macjack has quit IRC13:42
openstackgerritIhar Hrachyshka proposed openstack/neutron-lbaas: Removed new=True argument from create_connection
*** Brian_shang has quit IRC13:47
*** Brian_shang has joined #openstack-lbaas13:50
*** yamamoto has joined #openstack-lbaas13:51
*** Brian_shang has quit IRC13:57
*** Brian_shang has joined #openstack-lbaas13:57
*** ducttape_ has joined #openstack-lbaas14:07
openstackgerritIhar Hrachyshka proposed openstack/neutron-lbaas: Removed new=True argument from create_connection
*** Brian_shang has quit IRC14:20
*** TrevorV|Home has joined #openstack-lbaas14:25
*** armax has joined #openstack-lbaas14:54
*** Brian_shang has joined #openstack-lbaas14:56
*** Brian_shang has quit IRC15:01
*** amotoki has quit IRC15:03
*** Brian_shang has joined #openstack-lbaas15:05
*** kobis has quit IRC15:08
*** Brian_shang has quit IRC15:10
*** numans has quit IRC15:11
*** bharathm has joined #openstack-lbaas15:20
*** diogogmt has joined #openstack-lbaas15:22
*** openstackgerrit has quit IRC15:31
*** openstackgerrit has joined #openstack-lbaas15:32
*** evgenyf has quit IRC15:44
*** diogogmt has quit IRC15:48
*** diogogmt has joined #openstack-lbaas15:49
*** vivek-ebay has joined #openstack-lbaas15:51
*** bana_k has joined #openstack-lbaas15:55
*** TrevorV|Home has quit IRC15:56
*** TrevorV has joined #openstack-lbaas15:58
*** crc32 has joined #openstack-lbaas16:05
*** vivek-eb_ has joined #openstack-lbaas16:14
*** vivek-ebay has quit IRC16:14
*** KunalGandhi has joined #openstack-lbaas16:23
*** jschwarz has quit IRC16:30
*** Alex_Stef has joined #openstack-lbaas16:33
*** Alex_Stef has quit IRC16:38
*** vivek-eb_ has quit IRC16:51
*** blogan_ has joined #openstack-lbaas16:53
*** fnaval has joined #openstack-lbaas16:55
*** Alex_Stef has joined #openstack-lbaas16:56
*** blogan__ has joined #openstack-lbaas16:57
*** blogan_ has quit IRC17:00
*** Alex_Stef has quit IRC17:01
*** amotoki has joined #openstack-lbaas17:06
*** vivek-ebay has joined #openstack-lbaas17:13
*** bitblt has joined #openstack-lbaas17:13
*** numans has joined #openstack-lbaas17:16
*** yamamoto has quit IRC17:31
*** bana_k has quit IRC17:31
*** yamamoto has joined #openstack-lbaas17:32
*** yamamoto has quit IRC17:33
*** bharathm has quit IRC17:36
*** ducttape_ has quit IRC17:36
johnsomsbalukoff sbalukoff1 Here is how the spec renders for me:
*** fnaval has quit IRC17:52
*** ducttape_ has joined #openstack-lbaas17:59
*** numans has quit IRC17:59
*** bana_k has joined #openstack-lbaas18:01
*** bharathm has joined #openstack-lbaas18:01
*** SumitNaiksatam has joined #openstack-lbaas18:02
*** amotoki has quit IRC18:03
openstackgerritCarlos Garza proposed openstack/octavia: Implementing EventStreamer
*** fnaval has joined #openstack-lbaas18:11
*** alejandrito has joined #openstack-lbaas18:14
*** bitblt has quit IRC18:16
*** numans has joined #openstack-lbaas18:17
markvanHi, devstack question. Has there been any thought for add devstack-gate DEVSTACK_GATE_NEUTRON_LBAASV2?  This would allow other infra jobs to easily start making use of LBaas v2.   Or is there a plan to push for LBaas v2 as default in devstack at some point?18:28
*** ajmiller_ is now known as ajmiller18:29
ajmillermarkvan The plan is for V2 to be the default at some point in the (hopefully not-too-distant) future, and for V1 to be deprecated.  In fact the beginning of the M cycle seems like a good time to make that switch.18:30
*** yamamoto has joined #openstack-lbaas18:33
markvanajmiller: yup, I would agree.  So, if/when v2 becomes default, there will still be a way to enable v1 for existing testing written against that (like in Heat)?18:33
johnsomFYI, LBaaSv1 is now deprecated:
ajmillermarkvan, yes, there is a flag in devstack that will allow V1 to be specified, even if the default changes.18:36
markvanajmiller: What flag is that?18:39
markvanjohnsom: so v1 will be removed at some point?  N release?18:40
ajmillerLets make sure we are on the same page, I'm assuming your are talking about the devstack flags, right?  If you enable q-lbaas, right now that is treated as LBaaS V1.18:40
johnsomYes, the plan is to remove it.  I think we need two cycles in deprecated, so likely O release.18:40
*** woodster_ has joined #openstack-lbaas18:43
markvanjohnsom: thx, that sounds right.18:43
markvanajmiller: yup, q-lbaas gives just v1 right now.  Need to enable lbaas devstack plugin to get v2.18:44
ajmillermarkvan Right.  There are q-lbaas, q-lbaasv1, and q-lbaasv2 flags in the plugin, and right now q-lbaas is treated as q-lbaasv1.18:45
johnsommarkvan Right now the test nodes don't have nested virtualization, so using lbaasv2 with the Octavia driver can be a bit slow.18:45
johnsomEither the test nodes need to enable that in KVM or we get containers going before we really can switch everything over.18:46
johnsomOtherwise qemu takes too long to boot vms for us18:46
markvanajmiller: yeah those flags are only in the lbaas devstack plugin, not in the devstack tree.  I was referring to enabling it in the default devstack tree.18:47
ajmillermarkvan, Ahh,  the ultimate plan is to remove lbaas completely from the devstack tree.18:48
markvanajmiller: the issue I have is that the devstack-gate does not know about this, so it takes extra steps to enable it.  And that mean every gate job that wants to use it will be doing these extra steps which is ugly.18:48
markvanjohnsom: for the Heat integration tests, they are using the default devstack gate and spinning up a heat stack with LBaas18:49
markvanajmiller: ok, I can see removing it from the devstack tree, but gate jobs still need it, so that's why I was asking about support in the devstack-gate tree18:50
*** mhickey has quit IRC18:51
ajmillerRight.  We ran into this when we wrote the devstack plugin last spring.  Pulled the V1 setup from the devstack tree and broke everything, so we had to put some of that back.18:52
*** yamamoto has quit IRC18:53
markvanajmiller: yup still many folks touching it.  So, what do you think about a devstack-gate patch for something like DEVSTACK_GATE_NEUTRON_LBAAS to allow gate jobs to easily enable this devstack plugin?18:54
ajmillermarkvan It seems like that would be a step in the right direction!18:55
markvanajmiller: k, then I'll take a pass at something like that.  thx for the chat time.18:56
*** vivek-eb_ has joined #openstack-lbaas18:59
*** vivek-ebay has quit IRC19:03
*** armax has quit IRC19:05
*** SumitNaiksatam has left #openstack-lbaas19:05
*** crc32 has quit IRC19:09
*** fnaval has quit IRC19:16
bana_kHi, I was suggested to divide this commit into multiple commits,19:23
bana_kI need to add a new folder, how does that work?19:24
bana_kwhat I mean is lets say I add lbaas folder and in one commit, n I need to add in the same folder in another commit.19:25
bana_kso just add the same folder in second commit n add first commit as depends on?19:25
*** numans has quit IRC19:26
*** fnaval has joined #openstack-lbaas19:28
bloganbana_k: looks like they just want you to add other modules, not another directory19:30
bloganbana_k: so lbaas/ lbaas/ lbaas/ etc19:30
bana_kblogan : Please look at the last but one comment19:31
bana_kblogan: n comment by Sergey Kraynev just above that19:32
bana_kI thought they want me to divide the patch into multiple ones19:33
bloganbana_k: yeah, thats still what im understanidng that he wants19:33
bloganbana_k: oh yeah he does19:33
bana_kblogan: so I need multiple commits right ?19:33
bloganbana_k: there's a number of ways to do it19:33
bloganbana_k: yeah19:34
bloganbana_k: would the depend on the
bana_kblogan : yes19:34
bana_kblogan: n pool depends on listener19:35
bana_kblogan : chain continues19:35
bloganokay so then just make new commits on top of the loadbalancer one19:36
bana_koh ok. so I ll clone the loadbalancer change n working on it?19:37
bana_kblogan : thats what you mean ?19:37
bloganbana_k: i mean just pull out everything except the load balancer code out of, git commit --amend that first commit19:38
bloganbana_k: then add only the listener code you pulled out, and just do a git commit, which adds a new commit on top of the load balancer one19:38
bloganrepeat for hte rest19:38
bloganand then when you're done, do a git review and it will create reviews for all of the new commits, and add a PS to the load baalncer one19:39
*** numans has joined #openstack-lbaas19:39
bana_koh ok thanks blogan, have not done it before, will try it out.19:39
bloganbana_k: np, you're entering the world of dependency chains, where up is down, left is right, and you will go insane19:41
bloganbana_k: basically be careful19:41
*** armax has joined #openstack-lbaas19:41
bana_kblogan: he he he sure thanks, but I still dont understand why we need separate commits for each resources, they are kind of tightly coupled.19:42
bloganbana_k: easier to review one at a time19:42
bloganbana_k: versus 800 lines19:43
bloganin one19:43
bana_kblogan : hmm I see.19:43
bloganoh and also, its just better organized id say, i dont like a huge file of code if it can be separated out into individual modules either19:43
bloganbut thats opinion, but my opinion is that i'm right :)19:44
bana_kblogan: individual module I def agree,19:44
bana_kblogan: separate commits are the one I am not sure of :)19:44
bloganbana_k: yeah in that case, easier to review19:45
bloganbana_k: more psychological than anything, big wall of text is just demoralizing19:45
bana_kblogan: he he he agreed :).19:46
bana_kthanks a lot for all the input.19:46
*** mestery_ has joined #openstack-lbaas19:53
*** armax has quit IRC20:08
*** armax has joined #openstack-lbaas20:11
*** crc32 has joined #openstack-lbaas20:13
*** mestery_ has quit IRC20:19
*** crc32 has quit IRC20:19
openstackgerritmin wang proposed openstack/octavia: Adds scenario test
*** armax has quit IRC20:26
*** armax has joined #openstack-lbaas20:31
*** mestery_ has joined #openstack-lbaas20:32
*** mestery_ has quit IRC20:32
dougwigjohnsom: ping ?20:52
johnsomWhat is up dougwig?20:53
dougwigabout the review to have the neutron-lbaas/octavia minimal job use the octavia noop driver, ihar's point about reducing the number of moving parts that all of the rest of openstack incurs seems reasonable.  (also ping blogan rm_work )20:53
dougwigyou have a -1 on there, so i don't want to merge without folks here having a discussion, though.20:54
johnsomYeah.  My only concern is we won't catch those issues like nova breakage since we aren't running actual code20:54
dougwigcan a change in neutron break that?20:55
johnsomLet me look at the comments again, two minutes20:55
*** crc32 has joined #openstack-lbaas20:55
johnsomOh, ok, so my reason for -1 was if we make this change, we should undo the other changes we made to get the actual tests to run.  I.e. not leave the image download in, etc.20:56
johnsomI also worry that neutron changes will break the hot plugging20:58
dougwigwould that be covered by other neutron tests?  if that's coverage that is otherwise missing, i'll remove my +2.20:58
johnsomblogan might be best to answer that.20:59
johnsomI will remove my -1 if we undo the other changes in this patch.20:59
johnsomheck, I will update the patch to do that.20:59
*** fnaval has quit IRC21:03
bloganjohnsom, dougwig: give me a few, in a meeting21:05
rm_workdougwig: which CR is this?21:07
rm_workkk thx21:09
bloganso the scenario tests do test the actual real drivers21:10
blogannot to the full extent that we would all like, but we won't get that anyway21:10
blogani think its reasonable to have the minimal job use noop, we might even be able to get all the noop tests into one job, so we dont have as many jobs21:12
bloganactually, pretty sure we'd be able to21:13
johnsomYeah.  Ok, so I would like to see us undo those settings we put in to be able to boot vms.21:14
bloganso then why dont we put all the current split tests we have into one job, where its just an api test for lbs, listeners, pools, members and hms21:16
blogankeep the scenario tests21:16
rm_workman all this testing-musical-chairs21:16
rm_workis a little nuts21:16
bloganand then add a job that test a minimal lb,listener,pool,hm, and member with nova boots21:16
bloganyeah i hate it too but at this point, necessary evils, we're damned if we do damned if we dont21:17
rm_workinfra is prolly like "what are these guys doing"21:17
bloganoh well21:17
rm_workespecially since so many of our changes have come with us begging them to fast-track21:17
bloganthese guys are dealing with requirements people heave on us21:17
*** fnaval has joined #openstack-lbaas21:18
johnsomI would lean towards keeping the slots21:18
*** fnaval has quit IRC21:22
bloganand use the real drivers?21:23
rm_worki thought the system we use now is a decent compromise that actually gets all the testing done21:24
johnsomI'm willing to give in on the real drivers, but given the test run window keeps getting smaller, it seems like it would be valuable to have multiple jobs defined and available.  Even if we are just doing API.21:24
rm_workand not a huge fan of testing just with noop21:24
bloganif we use the noop drivers, there's no reason to keep the slots, if we use the real drivers and keep the slots then we can only enable a minimal set of tests, in which case the scenario tests already do that21:24
rm_worklike everyone is saying, as long as the real testing happens SOMEWHERE (even in 3rd-party CI) I am ok21:24
johnsomWe still need to get something going for the haproxy driver.  When containers come along, there might be value in the multiple slots21:25
bloganrm_work: no one is a fan of it, its just the best bad option we've had21:25
bloganjohnsom: haproxy driver?21:26
johnsomAlso, what about a slot for "single call"21:26
bloganyou mean magnum?21:26
bloganor something21:26
bloganjohnsom: i'd assume we'd have a happy path scenario test using that21:26
rm_workstill need to do single-call21:26
rm_workis that on our roadmap anywhere21:26
rm_workstill irks me that wasn't done already21:26
rm_workand i kinda want to just sit down and do it... but need to figure out the correct layer for it21:27
bloganive been talking with one of the intel guys that sits next to us, he's giving it a stab21:27
bloganif not i can do it whenever we the higher priority things done21:27
*** fnaval has joined #openstack-lbaas21:29
bloganjohnsom, dougwig, rm_work: would yall agree that the neutron cogate minimal job should run using hte noop drivers, then we just need to figure out our own jobs?21:29
dougwigblogan: seems reasonable to me, and i'd say we also need to run minimal in our repo for parity's sake.21:29
bloganshould = would be a better option21:29
blogandougwig: yeah agreed21:30
*** TrevorV has quit IRC21:30
johnsomI am fine moving minimal to noop.  I just want to take out the funk we put in go get VMs to boot when we make this change.  I don't want that stuff laying around if it's not needed21:31
johnsomIt makes my hack-y spidey senses light up.  (yes, I know I put them in)21:32
rm_workyeah i could care less what the neutron-co-gate tests21:32
rm_workas long as it doesn't break the api calls directly21:32
rm_workit's already its own slot21:32
rm_workno problems there21:32
bloganwell if you take out the image thing then that'll add anothe 5 minutes to all slots? or is it longer?21:33
*** vivek-eb_ has quit IRC21:33
rm_workall slots except the minimal stuff21:33
bloganits not bad if its 10 or less21:33
rm_workit's ... about 10 i think21:33
rm_workright johnsom21:33
johnsomYeah, it can be 1021:34
bloganstill puts us under an hour for each slot21:34
bloganusing noops21:34
*** fnaval has quit IRC21:34
blogani'd still advocate condensing all slots into one, and then adding additonal slots for high priority tests using real drivers21:35
bloganbut thats a lot of back and forth with infra so i wouldnt put that as a strong requirement21:36
johnsomRight, minimizing what they do and hanging onto them a bit doesn't hurt much21:37
johnsomSo dougwig, blogan, rm_work what do you think about me adding to this patchset, undoing the changes for the vm boot and removing the image download?21:42
johnsomI just worry about leaving an old image laying around that isn't really being used.21:43
dougwigjohnsom: i'd be ok with it.21:43
johnsomMaybe we can hack it to load the cirros image as the amp image to save build time.21:43
bloganthat'd be nice, but that sounds like a lot of hacking21:44
johnsomUndoing the changes is quick and clean.  The cirros image named amphora image, not sure21:45
blogani mean about getting cirros to work for us21:45
johnsomI didn't say it would work21:46
johnsomWe don't really need it for noop21:46
bloganoh i see what you mean21:46
*** fnaval has joined #openstack-lbaas21:46
johnsombuilding a working cirros image is painful.  I looked at that once21:46
bloganlol yeah thats what i would suspect21:46
bloganyou looked once and just regained your eyesight21:46
rm_workjohnsom: err, why would we even need *an* image for noop?21:48
rm_workcan't we just avoid building the amps entirely? isn't that the point of noop?21:49
rm_workmaybe I am misunderstanding noop21:49
johnsomBecause otherwise devstack builds one21:49
rm_worknoop to me means "no interaction with neutron or nova"21:49
rm_workwe don't need to hack cirros to work21:49
bloganrm_work: yeah its not octavia needing it21:49
rm_workwe just need to not build/add an image at all21:49
bloganhe's not saying to hack cirros to work21:49
rm_workbut like21:50
rm_workto load up21:50
johnsomSo add a flag to the devstack scripts that the gate test inserts that bypasses the image build?21:50
rm_workbut there isn't even a point in going that far is there?21:50
rm_workjohnsom: that would work21:50
johnsomSeems like we touch less by creating a ghost image.  I don't know21:50
rm_workI just feel like it's a bit dirty to do that kind of hack to clone the cirros image and add it again21:54
rm_workjust to avoid an if/else block or so in the gate script21:54
johnsomOk, so let's do two patches.  This one, cleaned up, remove the download, set a flag.   Second one to modify octavia to care about the flag21:56
rm_workthe octavia one would need to be first, right?21:57
johnsomNo, it was just build an image, so it doesn't really matter the order21:57
*** vivek-ebay has joined #openstack-lbaas21:57
rm_workmaybe i need to see what you're talking about21:58
openstackgerritMerged openstack/neutron-lbaas: Updated from global requirements
*** mestery has quit IRC22:08
openstackgerritMerged openstack/octavia: Add .eggs/ to .gitignore
openstackgerritMerged openstack/octavia: Minor restructuring of documentation tree
openstackgerritMichael Johnson proposed openstack/neutron-lbaas: Use noop octavia driver for minimal job too
johnsomrm_work blogan ^^ updated version that removes the amp boot stuff and sets a variable.22:29
johnsomI'm pretty sure that variable set will work.22:29
rm_workjohnsom: i don't think pools was supposed to go away?22:32
rm_workor did i miss something22:32
johnsomI didn't make that change22:32
johnsomIhar made that change, I didn't22:33
rm_workbase -> patch2, lost pools22:33
rm_workweird... uh ok22:33
rm_workis it... right? i don't see why we'd drop pools22:33
johnsomYeah, I only did 1->222:33
rm_worki'm a little confused then still, because we still want pools right?22:33
rm_workthat didn't go away while i was stuck on internal stuff, did it?22:34
johnsomI guess I'm not seeing where he removed pool22:34
johnsomWait, now I am confused22:34
johnsomHe added minimal to the apiv2 testenv22:35
johnsomrm_work I commented on the backing file comment.22:37
johnsomIt was added so there was enough disk allocated in devstack to allow an amp to boot22:37
johnsomI added this stuff here:
rm_workah i think i understand22:38
rm_workI was thinking we were leaving the others22:38
rm_workand JUST switching the minimal to use noop22:38
rm_workbut i see you said "too" in the commit message22:39
rm_workso NONE of these would use octavia anymore?22:39
johnsomI think that is what is being proposed here, yes.  All noop22:39
rm_workthat's a lot harder decision then22:40
rm_workI was willing to say without any hesitation that the minimal jobs should absolutely be noop22:40
johnsomI haven't traced back the setting Ihar made yet to fully understand that change22:40
johnsomThey already switched all of the others to noop22:40
rm_workyou do at least see where he removed "pool" right?22:40
rm_workI think removing pool is a mistake (like, a typo)22:41
johnsomThe other patchset was here:
rm_workhe probably meant to ADD minimal to that list22:42
rm_workoh these were split out differently again than when I set it up originally22:43
rm_workI wasn't using that key slot22:43
johnsomrm_work No, I don't see pool removed....22:43
johnsomDownload the file22:43
rm_workline 1322:43
johnsomif [ "$2" = "healthmonitor" ] || [ "$2" = "listener" ] || [ "$2" = "loadbalancer" ] || [ "$2" = "member" ] || [ "$2" = "minimal" ] || [ "$2" = "pool" ]; then22:43
rm_workAH yeah i just copy/pasted it out and i see it22:44
rm_workit's freaking gerrit22:44
rm_workwon't let me scroll that far to the right22:44
johnsomOh, strange22:44
rm_workit cuts it right after "minimal"22:44
johnsomMine line wraps22:44
rm_workyeah mine does not22:45
rm_worki think it's new vs. old theme22:45
rm_worknew theme doesn't linewrap22:45
rm_workbut it helps a shitton with comment tracking22:45
rm_workand looks a bit cleaner to me in general22:45
rm_worknever notice the linewrap thing because i'm always looking at PEP-8 Python ;P22:46
rm_workblah I keep trying to dig into min's review for cert rotation22:47
rm_workand keep getting sidetracked22:47
johnsomYeah, it's a long one.  I have been through the code, but still need to look at the tests22:47
rm_workI think I've been through it about 5 times now22:49
rm_workbut it is always so long and i don't know what changed so i have to start over...22:49
*** minwang2 has joined #openstack-lbaas22:50
rm_workerr was there a meeting today for lab related stuff?22:53
rm_workjohnsom / xgerman22:53
johnsomYesterday, you called in22:53
rm_workI was in the middle of a bunch of stuff yesterday and was barely able to hear or pay attention to the meeting22:53
rm_worki thought i heard something about scheduling a followup22:54
rm_workfor a different time22:54
rm_workbut maybe not?22:54
johnsomNext week.22:54
johnsom10/21 3:30 pacific22:55
johnsomYou should have got an invite on your gmail22:55
*** bharathm has quit IRC22:56
*** bharathm has joined #openstack-lbaas22:56
rm_workyeah cool, i see it22:57
xgermanyep, and I have you down for the trouble shooting section22:58
rm_worklol k awesome22:59
xgermanI made a very purple design ;-)22:59
rm_workso am I down for WRITING the troubleshooting section, or just presenting it? :P22:59
rm_worki was thinking when you said troubleshooting that i'd be in charging of running around the lab like a madman literally helping people troubleshoot23:00
rm_workLOL ok we'll see how this goes...23:00
xgermanyeah, I think it will be collaborative since we all have seen different problems…23:01
rm_workxgerman: can you add my rackspace email as an editor23:01
rm_worki don't sign in to my personal gmail on my work machine23:01
xgermanok, I guessed an e-mail — let me know if it wporked23:02
rm_workit did :P23:02
xgermanyou guys are almost as predictable as us23:02
rm_workit's a pretty tried and true format23:02
*** crc32 has quit IRC23:04
*** chlong has quit IRC23:16
*** rm_you|wtf has joined #openstack-lbaas23:25
*** Kiall has quit IRC23:30
*** _laco has quit IRC23:30
*** Kiall has joined #openstack-lbaas23:31
*** rm_you| has quit IRC23:32
openstackgerritOpenStack Proposal Bot proposed openstack/octavia: Updated from global requirements
*** markvan has quit IRC23:34
*** markvan has joined #openstack-lbaas23:35
*** _laco_ has joined #openstack-lbaas23:44
*** bharathm has quit IRC23:54
*** bradjones has quit IRC23:54
*** ajmiller has quit IRC23:54
*** madhu_ak has quit IRC23:54
*** blogan__ has quit IRC23:54
*** davidlenwell has quit IRC23:54
*** sbalukoff1 has quit IRC23:54
*** xgerman has quit IRC23:54
*** zigo has quit IRC23:54
*** michchap has quit IRC23:54
*** clev-away has quit IRC23:54
*** vivek-ebay has quit IRC23:54
*** telmich has quit IRC23:54
*** codekobe has quit IRC23:54
*** rm_you|wtf has quit IRC23:54
*** armax has quit IRC23:54
*** diogogmt has quit IRC23:54
*** doug-fish has quit IRC23:54
*** mugsie has quit IRC23:54
*** shakamunyi has quit IRC23:54
*** Kiall has quit IRC23:54
*** sbalukoff has quit IRC23:54
*** rm_work has quit IRC23:54
*** adam_g has quit IRC23:54
*** kbyrne has quit IRC23:54
*** markvan has quit IRC23:54
*** bana_k has quit IRC23:54
*** ducttape_ has quit IRC23:54
*** thomasem_ has quit IRC23:54
*** jorgem has quit IRC23:54
*** logan2 has quit IRC23:55
*** yuanying has quit IRC23:55
*** dougwig has quit IRC23:55
*** alejandrito has quit IRC23:55
*** openstackgerrit has quit IRC23:55
*** ctracey has quit IRC23:55
*** amit213 has quit IRC23:55
*** kfox1111 has quit IRC23:55
*** _laco_ has quit IRC23:55
*** numans has quit IRC23:55
*** blogan has quit IRC23:55
*** HenryG has quit IRC23:55
*** eezhova has quit IRC23:55
*** pckizer has quit IRC23:55
*** woodster_ has quit IRC23:55
*** sballe has quit IRC23:55
*** clduser has quit IRC23:55
*** sc68cal has quit IRC23:55
*** pc-pothole has quit IRC23:55
*** KunalGandhi has quit IRC23:55
*** minwang2 has quit IRC23:55
*** bedis has quit IRC23:55
*** harlowja has quit IRC23:55
*** redrobot has quit IRC23:55
*** Dave has quit IRC23:55
*** intr1nsic has quit IRC23:55
*** crc32|znc has quit IRC23:55
*** raginbajin has quit IRC23:55
*** johnsom has quit IRC23:55
*** divya has quit IRC23:55

Generated by 2.14.0 by Marius Gedminas - find it at!