Thursday, 2017-02-09

*** ducttape_ has joined #openstack-lbaas00:01
rm_workyeah, the real issue was just it getting the wrong interface because it only checked for network_id00:03
rm_worktechnically maybe we should actually check the subnet_id instead of the network_id? >_>00:04
rm_workjohnsom: is that a +1 until you test it?00:05
johnsom+1 until there are release notes00:06
openstackgerritMichael Johnson proposed openstack/octavia master: Fix the amphora failover flow docs diagram  https://review.openstack.org/43123200:06
rm_workah k00:11
openstackgerritMichael Johnson proposed openstack/octavia master: Fix the amphora failover flow docs diagram  https://review.openstack.org/43123200:21
johnsomderp00:23
rm_workok making a change to the name-length one that you should like00:24
*** cody-somerville has quit IRC00:26
openstackgerritAdam Harwell proposed openstack/octavia master: Allow a configuration option for random Amphora name  https://review.openstack.org/43118600:27
rm_workhow do I make a new release note?00:28
rm_workIs there a command?00:28
rm_workI've seen the files in the past, they look uuid-like00:28
johnsom$ reno new slug-goes-here00:28
johnsomCreated new notes file in releasenotes/notes/slug-goes-here-95915aaedd3c48d8.yaml00:28
johnsomhttp://docs.openstack.org/developer/reno/usage.html00:29
rm_workkk00:29
rm_workdoes a bugfix need a release note? :/00:30
rm_workah I guess "fixes" k00:30
johnsomOnly if it was a significant bug or has some impact on someone using/deploying00:31
johnsomIn this case it changes behavior/adds functionality so it would be nice to let people know.00:32
openstackgerritAdam Harwell proposed openstack/octavia master: Allow to create vip in lb-mgmt-net  https://review.openstack.org/43117900:35
rm_workalso possibly pre-caught what could maybe have been a weird bug00:36
rm_workI think00:36
rm_workline 317/31800:36
rm_worki think that's required now?00:36
johnsomyes, good catch00:39
openstackgerritAdam Harwell proposed openstack/octavia master: Add network_id to VIP to match neutron-lbaas  https://review.openstack.org/42992600:42
openstackgerritMichael Johnson proposed openstack/octavia master: Fix Octavia v1 API project_id for POST  https://review.openstack.org/43120700:43
johnsomOk, I think that is good for review00:43
* rm_work looks00:44
johnsomhttp://logs.openstack.org/79/431179/2/check/gate-octavia-pep8-ubuntu-xenial/7793524/console.html#_2017-02-09_00_38_54_28766800:47
rm_worksrsly01:08
rm_workFML01:08
rm_workOH GOD it's this mess01:11
rm_workpycharm, which normally chooses the correct indentation, chooses the wrong one for these because the "correct" indentation is also wrong, because it's the same as the next line01:11
rm_workI don't remember how to resolve this >_<01:12
johnsomJust bump it in more01:12
rm_workbut then it's at the same indentation as the next logical line01:13
rm_workwhich is also an error01:13
rm_work./octavia/network/drivers/neutron/allowed_address_pairs.py:318:5: E129 visually indented line with same indent as next logical line01:13
rm_work                    fixed_ip.ip_address != amphora.lb_network_ip):01:13
johnsomRight. bump it in more than the next line01:13
rm_workuhh01:13
rm_workbut then isn't it wrong again?01:13
* rm_work checks01:13
rm_workah I guess not01:14
rm_workbut i'm fixing it a different way anyway01:14
johnsomOk.  Stepping away for a bit.  I will likely be back around later01:15
rm_workkk01:16
*** cpuga has quit IRC01:17
*** ducttape_ has quit IRC01:22
rm_workwow, the test_plug_vip unit test is just... wrong01:26
rm_worklol01:26
rm_workwhelp01:27
*** beardedeagle has joined #openstack-lbaas01:57
armaxjohnsom: I did this02:10
armaxhttps://review.openstack.org/#/c/431251/02:10
armaxno bot cheered02:10
johnsomIt must be slacking off02:10
johnsomOr it is not setup for stable somehow02:11
armaxoh02:11
armaxthat might be it02:11
armaxactually it is it02:12
armaxhttps://github.com/openstack-infra/project-config/blob/master/gerritbot/channels.yaml#L111202:12
armaxI never noticed that02:13
*** ianychoi has quit IRC02:13
*** ianychoi has joined #openstack-lbaas02:13
johnsomI will take a look at the patch after dinner02:16
rm_workugh02:17
rm_workjohnsom: so the unit testing for plug_vip in test_allowed_address_pairs is totally useless02:18
rm_workI am starting to rewrite it so I can actually have a test for plugging on same-net02:18
rm_workbut02:18
rm_workfff i don't have the mental energy for this right now02:18
rm_workblogan: I am crying on the inside right now https://github.com/openstack/octavia/blame/master/octavia/tests/unit/network/drivers/neutron/test_allowed_address_pairs.py#L250-L28402:20
rm_worka bunch of the stuff that sets up isn't even called, and we don't test for what should be T_T02:20
rm_workI wonder if people changed it out from under the test and didn't update it because it didn't fail02:20
*** kberger has quit IRC02:33
*** kberger has joined #openstack-lbaas02:33
*** ducttape_ has joined #openstack-lbaas02:38
*** cody-somerville_ has quit IRC02:39
*** gongysh has quit IRC02:40
*** gongysh has joined #openstack-lbaas02:40
*** gongysh has quit IRC02:45
*** cody-somerville_ has joined #openstack-lbaas02:52
johnsomI have made some patches to the AAP driver so maybe I'm guilty02:55
*** openstackstatus has quit IRC02:58
*** coreycb has quit IRC02:59
*** openstackstatus has joined #openstack-lbaas03:01
*** ChanServ sets mode: +v openstackstatus03:01
*** coreycb has joined #openstack-lbaas03:05
*** yuanying has quit IRC03:05
*** yuanying has joined #openstack-lbaas03:07
*** beardedeagle has quit IRC03:08
*** yuanying has quit IRC03:11
*** armax has quit IRC03:31
*** links has joined #openstack-lbaas03:33
*** dileepr has quit IRC03:39
*** ducttape_ has quit IRC03:44
*** beardedeagle has joined #openstack-lbaas03:44
*** ducttape_ has joined #openstack-lbaas03:44
*** ducttape_ has quit IRC03:56
*** fnaval has joined #openstack-lbaas04:34
*** beardedeagle has quit IRC04:34
*** faizy has joined #openstack-lbaas04:35
*** amotoki has joined #openstack-lbaas04:46
*** amotoki_ has joined #openstack-lbaas04:47
*** amotoki has quit IRC04:51
*** fnaval has quit IRC04:51
*** fnaval has joined #openstack-lbaas04:52
*** ducttape_ has joined #openstack-lbaas04:55
*** cody-somerville has joined #openstack-lbaas05:27
*** cody-somerville has quit IRC05:28
openstackgerritMichael Johnson proposed openstack/octavia master: Fix a bug where ports may not be deleted  https://review.openstack.org/43132405:34
*** amotoki has joined #openstack-lbaas05:51
*** amotoki has quit IRC05:52
*** amotoki has joined #openstack-lbaas05:52
*** amotoki_ has quit IRC05:54
*** amotoki_ has joined #openstack-lbaas05:56
*** amotoki has quit IRC06:00
*** ducttape_ has quit IRC06:02
*** yuanying has joined #openstack-lbaas06:07
*** kberger has quit IRC06:32
*** cody-somerville has joined #openstack-lbaas06:35
*** csomerville has joined #openstack-lbaas06:42
*** cody-somerville has quit IRC06:45
*** eezhova has joined #openstack-lbaas06:54
*** kobis has joined #openstack-lbaas06:55
*** rcernin has joined #openstack-lbaas07:00
*** ducttape_ has joined #openstack-lbaas07:03
*** amotoki_ has quit IRC07:05
*** amotoki has joined #openstack-lbaas07:05
*** tesseract has joined #openstack-lbaas07:11
*** yamamoto has quit IRC07:17
*** gongysh has joined #openstack-lbaas07:29
*** pcaruana has joined #openstack-lbaas07:36
*** eezhova has quit IRC07:46
*** yamamoto has joined #openstack-lbaas07:53
*** yamamoto_ has joined #openstack-lbaas07:54
*** hjensas has joined #openstack-lbaas07:56
*** hjensas has quit IRC07:56
*** hjensas has joined #openstack-lbaas07:56
*** yamamoto has quit IRC07:57
*** yamamoto_ has quit IRC08:09
*** faizy has quit IRC08:11
*** faizy has joined #openstack-lbaas08:12
*** eezhova has joined #openstack-lbaas08:13
*** ducttape_ has quit IRC08:14
*** rcernin has quit IRC08:50
*** openstackgerrit has quit IRC09:03
*** rcernin has joined #openstack-lbaas09:06
*** amotoki has quit IRC09:11
*** jsheeren has joined #openstack-lbaas09:12
*** jsheeren has quit IRC09:13
*** jsheeren has joined #openstack-lbaas09:13
*** Alex_Staf has joined #openstack-lbaas09:20
*** openstackgerrit has joined #openstack-lbaas09:44
openstackgerritBernard Cafarelli proposed openstack/octavia master: Use cryptsetup/LUKS for encrypted ramfs  https://review.openstack.org/40381709:44
*** dosaboy has joined #openstack-lbaas09:58
*** dosaboy has quit IRC09:58
*** dosaboy has joined #openstack-lbaas10:00
*** anilvenkata has joined #openstack-lbaas10:01
*** ducttape_ has joined #openstack-lbaas10:11
*** faizy has quit IRC10:13
*** faizy has joined #openstack-lbaas10:14
*** ducttape_ has quit IRC10:17
*** kobis has quit IRC10:18
*** Alex_Staf has quit IRC10:25
*** links has quit IRC10:27
*** pcaruana has quit IRC10:32
*** pcaruana has joined #openstack-lbaas10:33
*** yamamoto has joined #openstack-lbaas10:35
*** yuanying has quit IRC10:36
*** itsuugo has joined #openstack-lbaas10:39
*** kobis has joined #openstack-lbaas10:39
*** links has joined #openstack-lbaas10:40
*** Alex_Staf has joined #openstack-lbaas11:13
*** Alex_Staf has quit IRC11:35
*** yamamoto has quit IRC11:36
*** yamamoto has joined #openstack-lbaas11:39
*** gongysh has quit IRC11:41
*** links has quit IRC11:55
*** faizy has quit IRC11:59
*** Alex_Staf has joined #openstack-lbaas12:00
*** links has joined #openstack-lbaas12:07
*** ducttape_ has joined #openstack-lbaas12:13
*** ducttape_ has quit IRC12:18
*** Alex_Staf has quit IRC12:34
*** yamamoto has quit IRC12:39
*** links has quit IRC12:50
*** Alex_Staf has joined #openstack-lbaas12:55
*** ducttape_ has joined #openstack-lbaas13:02
*** faizy has joined #openstack-lbaas13:13
*** ducttape_ has quit IRC13:36
*** kberger has joined #openstack-lbaas13:45
*** gongysh has joined #openstack-lbaas13:52
*** ducttape_ has joined #openstack-lbaas14:04
*** matt-borland has joined #openstack-lbaas14:06
openstackgerritNir Magnezi proposed openstack/octavia master: Auto detect haproxy user_group  https://review.openstack.org/42939814:07
openstackgerritNir Magnezi proposed openstack/octavia master: Auto detect haproxy user_group  https://review.openstack.org/42939814:09
*** catintheroof has joined #openstack-lbaas14:31
openstackgerritNir Magnezi proposed openstack/octavia master: Adds devstack plugin support for all amp base OS  https://review.openstack.org/43157614:40
bcafarelhe I want to try that one ^14:44
*** fnaval has quit IRC14:49
*** ipsecguy has quit IRC14:52
*** anilvenkata has quit IRC15:01
*** fnaval has joined #openstack-lbaas15:05
*** fnaval has quit IRC15:06
*** fnaval has joined #openstack-lbaas15:06
*** jsheeren has quit IRC15:10
*** armax has joined #openstack-lbaas16:02
*** gongysh has quit IRC16:13
johnsomI think we already have an environment variable for that, but I will test it out to make sure it passes down all the way.16:14
*** Alex_Staf has quit IRC16:29
*** rcernin has quit IRC16:32
*** eezhova has quit IRC16:43
*** ankur-gu_ has joined #openstack-lbaas16:54
*** itsuugo has quit IRC17:00
*** catintheroof has quit IRC17:06
*** catintheroof has joined #openstack-lbaas17:07
nmagnezio/17:10
*** catintheroof has quit IRC17:11
*** ducttape_ has quit IRC17:12
nmagnezijohnsom, ack. I saw this only in the diskimage-create script, do I was not aware of any other env variable. but sure if we can reuse i totally agree17:12
johnsomYeah, all of those should work by being set in localrc17:13
johnsomI'm loading up a devstack right now to test17:13
*** kobis has quit IRC17:13
nmagnezijohnsom, thanks for spending cycles on this :)17:15
nmagnezijohnsom, btw, i updated my other patch to use DISTRO_NAME https://review.openstack.org/#/c/429398/6/elements/haproxy-octavia/post-install.d/20-haproxy-user-group-config17:16
nmagnezibut I still need to test it17:16
johnsomNice17:16
nmagnezibuilding a new image locally now17:16
nmagnezialso i need to updates some existing tests17:16
nmagnezito keep CI happy17:16
johnsom+1 and keep the cores happy17:16
nmagnezithat is always a good idea.17:18
*** tesseract has quit IRC17:18
johnsomHmm, ok, doesn't look like it is working.  I will investigate more and if the existing env doesn't work we will move forward with your change.17:21
*** eezhova has joined #openstack-lbaas17:24
*** pcaruana has quit IRC17:30
*** Alex_Staf has joined #openstack-lbaas17:30
*** Alex_Staf has quit IRC17:46
johnsomnmagnezi Still around?17:46
nmagnezijohnsom, yes17:47
nmagnezijohnsom, just about to spin a devstack to review those api patches17:47
nmagnezi:)17:47
johnsomCool17:47
johnsomSo, you are right with your patch for devstack17:47
nmagnezijohnsom, nice to know :)17:47
johnsomI was wondering, should we do the same for the release ID?  I.e. xenial, 7, etc.17:47
nmagnezijohnsom, I can add that if you wish17:48
johnsomI think that would be good.  Easier for use to transition to centos 8...  grin17:48
johnsom'-d' distribution release id (default on ubuntu: xenial)"17:48
nmagnezijohnsom, it's actually a good idea, we would just need to keep it tidy so we won't feed the script ubuntu version when ppl choose centos or rhel for example17:49
johnsomThen I think that patch is good17:49
nmagnezijohnsom, so I need to submit another patchset, also with the condition i just mentioned. so I'll do my best to get to that today.17:51
johnsom+117:52
nmagnezii just don't want to feed the script with a ubuntu version when not needed. even if it works it might changed in the future17:53
johnsomYeah, we can do a condition to only add it if it is present17:54
*** SumitNaiksatam has joined #openstack-lbaas18:02
*** ducttape_ has joined #openstack-lbaas18:04
*** eezhova has quit IRC18:07
*** catintheroof has joined #openstack-lbaas18:26
*** Alex_Staf has joined #openstack-lbaas18:26
*** csomerville has quit IRC18:27
openstackgerritKen'ichi Ohmichi proposed openstack/octavia master: Switch to use test_utils.call_until_true  https://review.openstack.org/43170618:36
rm_worknmagnezi / johnsom: personally I feel like "even if it works it might changed in the future" is exactly why we SHOULD have a default, so it doesn't get changed out from under us and break all of our stuff like it did before18:43
*** ducttape_ has quit IRC18:52
*** eezhova has joined #openstack-lbaas18:55
johnsomYeah, I kind of lean the same way.  That is why I added those options in the first place18:55
nmagnezirm_work, i'm not sure i'm following, Adam. for the os flavor we keep the default (ubuntu) with my patch. and I'll just do the same for distribution (xenail if ubuntu, otherwise unset)18:56
rm_workah so *per OS* you would still pass a default version?18:57
rm_workso for ubuntu we would leave the default as "xenial", not unset?18:57
rm_workthe way i read that first was that we would unset all defaults for distro-version18:57
*** csomerville has joined #openstack-lbaas18:59
rm_work"I think we should also expose the release parameter here,18:59
rm_work'-d' distribution release id (default on ubuntu: xenial)"18:59
rm_workInclude it only if it is defined vs. setting a default."18:59
rm_worknmagnezi: ^^19:00
*** cody-somerville has joined #openstack-lbaas19:02
*** cody-somerville has quit IRC19:02
*** cody-somerville has joined #openstack-lbaas19:02
*** itsuugo has joined #openstack-lbaas19:04
*** SumitNaiksatam has quit IRC19:05
*** csomerville has quit IRC19:05
*** Alex_Staf has quit IRC19:16
nmagnezirm_work, oh, got it now. will do that in the next ps19:16
rm_worknmagnezi: do what? :P19:17
rm_workWhat you said, or what I said? They are opposite I think 6-619:18
rm_workerr.... s/6-6/T_T/19:18
*** itsuugo has quit IRC19:19
*** itsuugo has joined #openstack-lbaas19:20
*** itsuugo has quit IRC19:24
*** kobis has joined #openstack-lbaas19:24
rm_workjohnsom: does my patch make more sense when I change it to match the random-name bent?19:25
johnsomI will take a look again.  I don't remember where we left that yesterday...19:27
*** ducttape_ has joined #openstack-lbaas19:34
rm_workupdated https://review.openstack.org/#/c/431186/ to be more directly applicable, since you were understandably dubious about the conflation of the two issues19:34
rm_workchanged the variable name and the help messages to better reflect the purpose19:35
rm_workand reformatted the actual code portion to be simpler19:37
*** faizy has quit IRC19:39
openstackgerritAdam Harwell proposed openstack/octavia master: Add option to choose an availability_zone for amps  https://review.openstack.org/42481619:40
rm_work^ just a rebasw19:40
rm_work*rebase19:40
*** eezhova has quit IRC19:41
*** beardedeagle has joined #openstack-lbaas19:43
*** eezhova has joined #openstack-lbaas19:57
*** itsuugo has joined #openstack-lbaas19:57
openstackgerritKen'ichi Ohmichi proposed openstack/octavia master: Switch to use test_utils.call_until_true  https://review.openstack.org/43170620:11
-openstackstatus- NOTICE: Restarting gerrit due to performance problems20:20
rm_workjohnsom: https://review.openstack.org/#/c/428999/20:27
*** eezhova has quit IRC20:31
*** ankur-gu_ has quit IRC20:31
*** ankur-gu_ has joined #openstack-lbaas20:33
*** harlowja_ has joined #openstack-lbaas20:35
rm_workjohnsom: yep looks like I'll be editing the VRRP templates a bit and making the HM listen to TCP as well, and completing failover event requests <_<20:36
rm_workI wonder if that'll be possible to get upstream20:36
*** harlowja has quit IRC20:37
*** kobis has quit IRC20:47
*** ducttape_ has quit IRC20:50
*** ducttape_ has joined #openstack-lbaas20:55
nmagnezirm_work, hi, sorry I was afk. so re "do what" I think that adding a distribution flag plus "Include it only if it is defined vs. setting a default." covers it, no?21:04
nmagnezijohnsom, ^^21:04
*** fishbone has quit IRC21:22
rm_workI'm saying I *want* to set a default per OS21:29
rm_workif someone says ubuntu but doesn't set a specific version, I WANT to have a default of "xenial"21:30
rm_workand similar for centos21:30
*** eezhova has joined #openstack-lbaas21:30
johnsomI think what Nir is saying is with these options we can specify specifically which to use for the gates.  The default is in our diskimage-create already.  Though currently we are only listing defaults for ubuntu/xenial.  centos isn't really default anywhere at this point21:38
*** cody-somerville_ has quit IRC21:38
*** ducttape_ has quit IRC21:40
*** beardedeagle has quit IRC21:45
*** beardedeagle has joined #openstack-lbaas21:46
rm_workkk21:49
johnsomYou do make a good point about having defaults in multiple locations.  I have added a comment to the patch about that.21:52
*** harlowja_ has quit IRC21:58
*** eezhova has quit IRC22:00
*** harlowja has joined #openstack-lbaas22:05
*** ducttape_ has joined #openstack-lbaas22:10
*** matt-borland has quit IRC22:25
*** yamamoto has joined #openstack-lbaas22:56
*** itsuugo has quit IRC23:11
*** itsuugo has joined #openstack-lbaas23:11
*** yuanying has joined #openstack-lbaas23:12
*** itsuugo has quit IRC23:16
rm_workjohnsom: https://review.openstack.org/#/c/428999/23:41
rm_workjohnsom: https://review.openstack.org/#/c/431186/23:42
johnsomWill have a look23:42
openstackgerritAdam Harwell proposed openstack/octavia master: Allow to create vip in lb-mgmt-net  https://review.openstack.org/43117923:46
rm_workfixed the pep8 bug, still working on tests, it's kinda ridiculous23:46
rm_workhaving to figure out how all this works23:46
rm_work... which is probably a good thing23:46
*** ipsecguy has joined #openstack-lbaas23:52
*** ankur-gu_ has quit IRC23:57

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