Thursday, 2018-01-18

*** _k0da has joined #openstack-powervm00:00
*** esberglu has quit IRC00:24
*** edmondsw has joined #openstack-powervm00:35
*** _k0da has quit IRC00:35
*** edmondsw has quit IRC00:40
*** openstackstatus has quit IRC01:55
*** toanster has quit IRC01:55
*** toan has joined #openstack-powervm01:56
*** openstackstatus has joined #openstack-powervm01:56
*** ChanServ sets mode: +v openstackstatus01:56
*** edmondsw has joined #openstack-powervm02:24
*** edmondsw has quit IRC02:28
*** csky has joined #openstack-powervm02:41
*** AlexeyAbashkin has joined #openstack-powervm02:46
*** csky has quit IRC02:46
*** AlexeyAbashkin has quit IRC02:50
openstackgerritOpenStack Proposal Bot proposed openstack/networking-powervm master: Updated from global requirements  https://review.openstack.org/53502403:07
openstackgerritOpenStack Proposal Bot proposed openstack/nova-powervm master: Updated from global requirements  https://review.openstack.org/53503303:12
*** AndyWojo has quit IRC03:30
*** AndyWojo has joined #openstack-powervm03:31
*** AlexeyAbashkin has joined #openstack-powervm03:46
*** AlexeyAbashkin has quit IRC03:51
*** csky has joined #openstack-powervm04:28
*** _k0da has joined #openstack-powervm06:18
*** csky has quit IRC07:01
*** edmondsw has joined #openstack-powervm07:06
*** _k0da has quit IRC07:06
*** edmondsw has quit IRC07:11
*** _k0da has joined #openstack-powervm08:03
*** _k0da has quit IRC08:19
*** AlexeyAbashkin has joined #openstack-powervm08:26
*** edmondsw has joined #openstack-powervm08:54
*** edmondsw has quit IRC08:59
*** csky has joined #openstack-powervm09:01
*** csky has quit IRC09:06
*** _k0da has joined #openstack-powervm09:31
*** edmondsw has joined #openstack-powervm10:42
*** edmondsw has quit IRC10:47
*** csky has joined #openstack-powervm12:01
*** csky has quit IRC12:06
*** edmondsw has joined #openstack-powervm12:30
*** edmondsw has quit IRC12:35
openstackgerritMerged openstack/networking-powervm master: Updated from global requirements  https://review.openstack.org/53502413:19
*** edmondsw has joined #openstack-powervm13:23
*** apearson has joined #openstack-powervm13:57
*** apearson has quit IRC14:00
*** csky has joined #openstack-powervm14:06
*** svenkat has joined #openstack-powervm14:08
*** csky has quit IRC14:09
svenkatefried: edmondsw: I want to start a conversation on max_capacity for VNIC back devs for SR-IOV in pypowervm and nova_powervm.14:09
edmondswsvenkat sure, though efried may be pretty busy today (nova feature freeze)14:10
*** apearson has joined #openstack-powervm14:10
svenkati will propose a change to pypowervm : to add max_capacity parameter to set_vnic_back_devs method in tasks/sriov.py. and in that method will pass it to card.VNICBackDev.bld method along with existing capacity…14:10
svenkatedmondsw : i will propose my changes and we can continue later on…14:11
edmondswsounds good14:11
svenkatand as VNICBackDev already supports max_capacity, no more pypowervm changes are needed.. and i will also propose a change to nova_powervm - sriov plug method  - will pass max_capacity to set_vnic_back_devs  method call in it.14:12
svenkatthanks, i will propose these changes and proceed from there.14:12
efriedsvenkat Yup, sounds like a good plan.14:20
efriedsvenkat I suspect some interesting stuff will shake out in the implementation.14:21
svenkatefried : yes.. definetely ! i will also propose a change to sriov mechanism driver in networking_powervm as well. (gather max capacity from qos attached to port and pass it onto to vif object to nova-powervm…). more details later…14:25
efriedsvenkat *that* is where I think the trickiness will come in.14:26
*** esberglu has joined #openstack-powervm14:26
svenkatefreid: yes.14:26
*** esberglu has quit IRC14:30
*** esberglu has joined #openstack-powervm14:33
esbergluedmondsw: efried: Can one of you hit 6238 when you get a chance? Remove ovs from the skip list14:33
edmondswesberglu ack14:33
*** tjakobs has joined #openstack-powervm14:33
efriedDone14:34
esberglutx14:34
esbergluThe patching logic still isn't quite perfect, it doesn't handle the patches if they are merged14:34
efriedBecause we're only backtracing to master14:35
edmondswefried is just too quick14:35
edmondsw:)14:35
esbergluI ended up restacking the undercloud yesterday afternoon. Run times are back to normal14:37
esbergluThe compute nodes were having trouble communicating with the rabbitmq server on the control14:37
esbergluI thought it was network related, but every once in a while it seems rabbitmq just goes haywire14:37
esbergluAnd restacking the control is the only way I've found that resets it14:38
esbergluStill not sure what the underlying issue is14:38
*** tjakobs has quit IRC14:40
*** tjakobs has joined #openstack-powervm14:46
*** csky has joined #openstack-powervm14:49
*** csky has quit IRC14:49
*** csky has joined #openstack-powervm14:50
edmondswesberglu I saw some pok network issues unrelated to CI yesterday, and some others did as well, so I do think we can probably blame the network14:57
edmondswthough maybe it caused rabbitmq to go haywire14:57
*** AlexeyAbashkin has quit IRC16:13
esbergluedmondsw: Network issues seem to be persisting. Pretty much every CI run is failing before stacking today16:37
esbergluFailing to access github16:37
edmondsw:(16:37
efriedesberglu What'd you do here?  https://review.openstack.org/#/c/524658/316:57
esbergluefried: Just pulled the latest patch off the logserver and uploaded16:57
esbergluI'm trying to figure out the best way to carry that forward, a logserver redeploy will wipe it16:58
efriedesberglu Oh, yeah, so that ain't right.16:58
efriedGotcha - appreciate the effort.  But that stuff was for a different change: https://review.openstack.org/#/c/524744/16:58
efried...which I think should match what's on the log server as is.16:59
efriedIf you could confirm that, and revert the other patch set...16:59
efriedI'll get back to these eventually :)16:59
esbergluefried: We're never gonna be able to merge those correct?17:00
efriedI think the color themes have a good chance.17:01
efriedJust needs polish, and pestering reviewers.17:01
efriedI've been frying bigger fish lately tho.17:01
esbergluefried: Reverted, sorry bout that17:04
esbergluYeah looks like the logserver actually has17:06
esbergluhttps://review.openstack.org/#/c/524731/17:07
esbergluand https://review.openstack.org/#/c/524744/17:07
efriedmm17:19
*** _k0da has quit IRC17:24
openstackgerritEric Berglund proposed openstack/nova-powervm master: Autospeccing: Tasks  https://review.openstack.org/53542017:35
*** esberglu has quit IRC17:41
*** csky_ has joined #openstack-powervm17:54
*** esberglu has joined #openstack-powervm17:56
*** csky has quit IRC17:57
*** AlexeyAbashkin has joined #openstack-powervm17:58
*** AlexeyAbashkin has quit IRC18:03
*** AlexeyAbashkin has joined #openstack-powervm18:14
*** AlexeyAbashkin has quit IRC18:19
*** AlexeyAbashkin has joined #openstack-powervm18:22
*** AlexeyAbashkin has quit IRC18:26
openstackgerritEric Berglund proposed openstack/nova-powervm master: Autospeccing: NVRAM  https://review.openstack.org/53544218:27
openstackgerritEric Berglund proposed openstack/nova-powervm master: Remove unsupported nova-networking  https://review.openstack.org/53544818:55
openstackgerritMerged openstack/nova-powervm master: Remove unsupported nova-networking  https://review.openstack.org/53544819:25
edmondswesberglu why the change to assertEqual(0, ...call_count) instead of assert_not_called?19:39
esbergluedmondsw: I was getting attribute errors19:41
esbergluAttributeError: 'function' object has no attribute 'assert_not_called'19:41
edmondswoh, because that's not part of the autospec19:42
edmondswI thought mock must handle those special, but I guess not19:42
edmondswesberglu what about https://review.openstack.org/#/c/535420/1/nova_powervm/tests/virt/powervm/tasks/test_vm.py@46 ?19:43
esbergluAnother attribute error, the mock stg_ftsk needs a name attribute19:44
esbergluhttps://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/tasks/vm.py#L11519:45
esbergluOtherwise it fails there19:45
*** AlexeyAbashkin has joined #openstack-powervm20:45
*** AlexeyAbashkin has quit IRC20:49
*** _k0da has joined #openstack-powervm21:04
openstackgerritEric Berglund proposed openstack/nova-powervm master: Autospeccing: Tasks  https://review.openstack.org/53542021:11
openstackgerritEric Berglund proposed openstack/nova-powervm master: Autospeccing: NVRAM  https://review.openstack.org/53544221:11
edmondswesberglu I would have +2d PS2 on 535420 if you hadn't -2d... do you want me to remove my +2 on 535442 then?21:29
edmondswand I should be clear, I'm not saying we have to autospec everything we could possibly autospec. Was just pointing out that we *could*21:30
esbergluedmondsw: I was just thinking it would be easier to do in 1 pass than 221:30
esbergluHowever it is really easy to just roll through the files hitting the @mock.patch, finding everything else not as much21:31
edmondswyeah, that's why I thought you might want to just do what you've done as a first pass (which may end up being the only pass)21:32
edmondswand not try to bite it all off at once21:32
edmondswbut I'd be interested in efried's thoughts on this21:32
esbergluedmondsw: Yeah lets just move forward, I'll pull my -2. I don't really want to spend a ton of time on this, would rather knock out the easy ones21:33
*** openstackgerrit has quit IRC21:33
edmondsw++21:33
edmondswas long as efried agrees, I think that's a good plan21:33
efriedFar as I'm concerned, any patch that imposes any number of autospec improvements and passes tests can be merged.  We don't have to be anal about making sure we cover entire modules at once, or whatever.21:33
edmondsw++21:34
edmondswesberglu +2, just please hold off merging until CI passes21:39
esbergluedmondsw: Yep. CI still being plagued with git pull failures21:40
edmondswesberglu actually, this is just UT changes, I don't think we care about CI... just zuul21:41
*** AlexeyAbashkin has joined #openstack-powervm21:45
*** AlexeyAbashkin has quit IRC21:50
esbergluedmondsw: Yeah just UT, I'll go ahead and merge21:50
*** openstackgerrit has joined #openstack-powervm22:08
openstackgerritMerged openstack/nova-powervm master: Autospeccing: Tasks  https://review.openstack.org/53542022:08
openstackgerritMerged openstack/nova-powervm master: Autospeccing: NVRAM  https://review.openstack.org/53544222:11
*** apearson has quit IRC22:15
*** edmondsw has quit IRC22:24
*** csky_ has quit IRC22:48
efriedesberglu For these, I'm good with a fast approve; but it's not good form to +2 your own patch22:48
*** csky has joined #openstack-powervm22:49
*** esberglu has quit IRC22:49
*** csky has quit IRC22:54
*** svenkat has quit IRC23:00
*** csky has joined #openstack-powervm23:12
*** csky has quit IRC23:17
*** tjakobs has quit IRC23:23
*** edmondsw has joined #openstack-powervm23:23
*** edmondsw has quit IRC23:29

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!