Tuesday, 2018-03-27

*** k0da has quit IRC00:07
openstackgerritMerged openstack/networking-powervm master: add lower-constraints job  https://review.openstack.org/55593600:07
*** k0da has joined #openstack-powervm00:27
*** k0da has quit IRC00:45
*** arunman has joined #openstack-powervm03:37
*** arunman has quit IRC03:38
*** AlexeyAbashkin has joined #openstack-powervm04:33
*** AlexeyAbashkin has quit IRC04:37
*** chhagarw has joined #openstack-powervm05:03
*** arunman has joined #openstack-powervm05:25
*** AlexeyAbashkin has joined #openstack-powervm07:51
*** k0da has joined #openstack-powervm08:08
*** k0da has quit IRC09:18
*** k0da has joined #openstack-powervm09:30
*** k0da has quit IRC09:44
*** chhagarw has quit IRC10:51
*** AlexeyAbashkin has quit IRC11:05
*** AlexeyAbashkin has joined #openstack-powervm11:07
*** arunman has quit IRC11:28
*** openstackgerrit has quit IRC11:33
*** edmondsw has joined #openstack-powervm12:13
*** arunman has joined #openstack-powervm12:46
*** arunman has quit IRC12:51
*** arunman has joined #openstack-powervm12:52
*** chhagarw has joined #openstack-powervm13:21
*** k0da has joined #openstack-powervm13:47
*** arunman has quit IRC13:50
*** arunman has joined #openstack-powervm13:51
*** esberglu has joined #openstack-powervm13:58
edmondsw#startmeeting PowerVM Driver Meeting14:03
openstackMeeting started Tue Mar 27 14:03:21 2018 UTC and is due to finish in 60 minutes.  The chair is edmondsw. Information about MeetBot at http://wiki.debian.org/MeetBot.14:03
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.14:03
*** openstack changes topic to " (Meeting topic: PowerVM Driver Meeting)"14:03
openstackThe meeting name has been set to 'powervm_driver_meeting'14:03
edmondswhttps://etherpad.openstack.org/p/powervm_driver_meeting_agenda14:03
edmondsw#topic In-Tree Driver14:03
*** openstack changes topic to "In-Tree Driver (Meeting topic: PowerVM Driver Meeting)"14:03
edmondswhttps://etherpad.openstack.org/p/powervm-in-tree-todos14:03
edmondswesberglu you want to give us a quick update?14:04
esbergluYep14:04
efriedรถ/  (new haircut)14:04
edmondsw:)14:05
esbergluNeed to respond to a few things on network hotplug for mriedem14:05
*** arunman has quit IRC14:05
esbergluBut looks like he's ready to +2 after that14:05
edmondswwell, after we get the CI testing it14:05
edmondswhow's that going?14:05
esbergluedmondsw: Sounded like he was okay with putting it through anyways, he asked about that in -nova yesterday14:06
esbergluedmondsw: Still haven't gotten active RMC was gonna talk to mdrabe today14:06
edmondswyeah, I saw, but the way I read it he was only ok with +2 once the CI is testing it14:06
esbergluCould be, either way we need to get that working14:07
edmondswyep14:07
esbergluvSCSI, Snapshot, DiskAdapter and Localdisk are all ready for review14:07
edmondswlet's focus on getting that working in CI14:07
esbergluYep14:07
esbergluNext up is resize but I'm gonna sit on that until we start moving on the rest14:08
edmondsw+1 on vscsi14:08
esbergluedmondsw: Do the pypowervm 1.1.12 in a follow on?14:09
edmondswI guess. that's what efried suggested, and I'll yield to him on that14:09
esbergluAdding a todo to the etherpad14:10
efrieddid I say that?14:10
edmondswyep, in the review14:10
efriedOh, because we were going to wait until 1.1.13 to bump the minimum.14:11
edmondsw1.1.13 should be cut today, and as soon as that's in u-c we can bump the minimum14:11
edmondswif that happens before we see anyone looking at vscsi, we can make the change there, else we can try to do a followon14:11
edmondswsound like a plan?14:11
esbergluSounds good14:11
esbergluThat's all I had for IT14:12
edmondswI'm still trying to get to reviewing snapshot :(14:12
efriedrunways start tomorrow.14:12
edmondswcool14:12
efriedHave we queued up the ready reviews in there yet?14:13
edmondswefried I think vscsi is the only one really ready14:13
edmondswI will add it14:13
esbergluNope was waiting until I had +1s from both of you on vSCSI14:13
esbergluAnd CI for network hotplug14:13
efriedam I not +1 on that yet?14:13
edmondswyou are... I wasn't until just now14:13
efriedokay.14:14
efriedWanna wait a week, I can be +2 :)14:14
edmondswesberglu actually you go ahead and add vscsi to the runway14:14
esbergluOh yeah congrats efried!14:14
edmondswefried oh? :)14:14
edmondswcongrats then!14:14
efriedthanks, provisionally.  The nom is out on the ML.14:14
edmondswhadn't noticed yet... that's awesome14:14
efriedI'm right chuffed, as they used to say on the east side (of London)14:15
edmondswas you should be14:15
edmondswone thing about localdisk14:16
edmondswdid we work out what we're doing so that conf settings don't conflict between IT and OOT?14:17
esbergluOh right. No we didn't14:17
edmondsware you going to try my suggestion of an OOT change to check and see if the setting is already defined and if so not redefine it?14:18
esbergluSure14:18
edmondswok, let us know how that goes14:18
edmondswanything else for IT?14:18
efriedThat doesn't sound like it'll work.  I don't think you can predict the load order.14:18
efriedPerhaps ask around in -oslo14:19
edmondswthat was my fear14:19
efriedOr on the dev ML.14:19
efriedI bet dhellmann has this answer off the top.14:19
edmondswI thought maybe we could assume nova was loaded before a nova virt driver, though14:19
esbergluWe could name the IT group something slightly different14:19
edmondswyeah, he'd be the person I'd go to14:19
edmondswesberglu I really don't want to name things differently if we can avoid it14:19
efriedNamespacing would be okay, except we really want the in14:20
efriedin-tree one to be called [powervm]14:20
edmondswa) I want consistency14:20
efriedoh, now I remember.14:20
edmondswb) I like the existing names better than anything else I've thought of14:20
efried(edmondsw We wouldn't have to change the conf option names, just the group name.  But even so)14:20
efriedI remember how we need to tackle this.14:21
edmondswefried right... but like you said, we want the group name to stay the same14:21
efriedBecause we already lockstep nova and nova-powervm.14:21
efriedWe introduce the conf option into nova and then *remove* same from nova-powervm.14:21
esbergluCan we just add it to nova and remove from nova-powervm14:21
esbergluYeah14:21
efriedThose patches land in quick succession and we're good.14:21
edmondswexcept for the time between the first merge and the second14:22
esbergluefried: There might still be a problem with that, the group would get registered twice I think14:22
efriedwhich should be negligible.  Put Depends-On from the nova-powervm patch to the nova patch.14:22
edmondswand CI will be failing OOT for the IT change14:22
edmondswwhich I guess we can ignore14:23
efriedyeah14:23
efriedesberglu: Not sure about the group redef thing.  Another one for dhellmann.14:23
edmondswdo we have Depends-On working?14:23
esbergluno14:23
efriedoh, right.  No.  But that's okay.  We'll just land 'em quickly.14:23
edmondswyeah, I didn't think so14:24
edmondswk14:24
edmondswso esberglu will check with dhellmann on the group redef thing14:24
edmondswand we'll take it from there14:24
edmondsw#topic Out-of-tree Driver14:25
*** openstack changes topic to "Out-of-tree Driver (Meeting topic: PowerVM Driver Meeting)"14:25
edmondswhttps://etherpad.openstack.org/p/powervm-oot-todos14:25
efriedor just try it14:25
edmondswwe've got a couple things waiting for the pypowervm bump14:26
edmondswalready said that should release today, so the bump should happen later this week14:26
edmondswI think a couple of us have dropped more comments on the volume refactor, and it will need to be rebased again14:27
efriedAre we really going forward with that?14:27
efriedI mean, I'm all for the idea, but does tjakobs really have the bandwidth to go test all the drivers live?14:28
edmondswhe hasn't said otherwise14:28
edmondswhe was willing14:28
esbergluI thought there was something in slack the other day suggesting otherwise14:28
efriedokay, wfm.  *I* ain't gonna do it.14:28
edmondswwe'll see how that goes, but I would really like to see that merge14:28
edmondsw(once thoroughly tested)14:29
efriedyeah, cause I'm going to be pretty hardline about not merging it unless it's tested.14:29
edmondswsame14:29
edmondswI think everyone's in agreement there14:29
edmondswbesides tjakobs, I want PowerVC to test it14:29
edmondswso we know the things they use are still working14:30
edmondswthat should give us another layer of confidence14:30
*** tjakobs has joined #openstack-powervm14:30
edmondswanything else OOT?14:30
edmondswoh chhagarw is working on some stuff for iscsi that I should bring up14:31
edmondswwe've got an issue there where I don't think it will be possible to setup the same volume through both NovaLink as a VIOS and through traditional VIOS14:31
edmondswso we're trying to work out how to handle that14:32
efriedat the same time?14:32
efrieduh, sorry, maybe that didn't make any sense at all.14:32
efriedNever mind.14:32
edmondswk14:32
edmondswat any rate, there will be nova-powervm changes coming for iscsi in one form or another14:33
edmondswanything else OOT?14:33
esberglunope14:33
edmondsw#topic Device Passthrough14:33
*** openstack changes topic to "Device Passthrough (Meeting topic: PowerVM Driver Meeting)"14:33
edmondswI haven't been able to do much here the last week14:34
edmondswefried you have anything?14:34
efriedToday is spec review day.  Tomorrow runways start.  Other than continuing to shepherd outstanding patches and review related stuff, no update from last time.14:34
efriedupdate_provider_tree is queued up for the first runway, I think.14:35
edmondswyep, saw that14:35
edmondswvery good14:35
efriedSo I expect that to be merged within the next week or two.14:35
*** arunman has joined #openstack-powervm14:35
efriedBut progress on NRP-in-a_c is still very slow.14:35
efriedAnd that's going to be a prereq for... everything else.14:35
edmondswa_c ?14:35
efriedIncluding granular14:35
efriedGET /allocation_candidates14:35
edmondswah14:35
efriedThe placement API that does scheduling14:36
efriedbasically14:36
efriedmeanwhile pretty much every spec on the slate talks about using granular as if it's already a thing14:36
efriedSo no pressure.14:36
edmondsw:)14:37
edmondswalright, moving on to CI14:37
edmondsw#topic PowerVM CI14:37
*** openstack changes topic to "PowerVM CI (Meeting topic: PowerVM Driver Meeting)"14:37
edmondswhttps://etherpad.openstack.org/p/powervm_ci_todos14:37
edmondswesberglu?14:37
esbergluTop priority is the network hotplug CI14:37
edmondsw+114:37
esbergluThen I can hop back into the nodepool update14:38
esbergluEverything is running smoothly again for the most part14:38
esbergluIT and OOT are actually seeing the vOPT naming conflict thing at about the same rate14:39
esbergluMeaning https://review.openstack.org/#/c/540453/ isn't really doing anything14:39
esbergluAnd that it might not be the naming conflict we thought it was14:39
edmondswhmm14:39
esbergluNeed to hop back into that and debug14:40
edmondswshould we prioritize multinode testing over the nodepool update?14:41
esbergluedmondsw: I really wanted to finish the mgmt service upgrade and queens upgrade first14:42
esbergluBut keep getting pulled off elsewhere14:42
edmondswthink we're getting close there?14:42
esbergluProbably time to start weighing that choice though14:42
esbergluIdk it depends on if the management service update fixes the issues that the queens undercloud is seeing14:43
edmondswyou've already got a commit up for resize, and it will need multinode CI before it has a chance of merging14:43
esbergluedmondsw: Yeah I will start looking at it more in depth, I'll try to have a plan moving forward for next weeks meeting14:44
edmondswsounds good14:44
esbergluNothing else from me14:45
edmondsw#topic Open Discussion14:46
*** openstack changes topic to "Open Discussion (Meeting topic: PowerVM Driver Meeting)"14:46
edmondswefried and I just got approval to book travel for the summit14:46
efriedOh, cool, hadn't seen that.14:46
edmondswshould be in your inbox14:46
efried222 unread as of right now.14:47
edmondswI plan to go ahead and do that today, see how booked the hotels are (probably full)14:47
edmondswwill let you know what I find14:47
edmondswanything else to discuss?14:47
efriedI may ping you later for info about which hotels etc.  I don't remember seeing anything on the website last time I looked.14:47
edmondswhmm... I registered, but haven't looked for flights/hotels yet14:48
* efried headless chicken / one-armed wallpaper hanger / one-legged man in ass-kicking contest today14:48
edmondswyeah, today is gonna be rough for me too14:48
edmondswhave a dentist appointment, Ben is visiting so we have extra meetings with that, etc.14:49
efriedso, let's close this down if nothing else to discuss (that can't wait)14:49
edmondswyep14:49
edmondsw#endmeeting14:49
*** openstack changes topic to "This channel is for PowerVM-related development and discussion. For general OpenStack support, please use #openstack."14:49
openstackMeeting ended Tue Mar 27 14:49:36 2018 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)14:49
openstackMinutes:        http://eavesdrop.openstack.org/meetings/powervm_driver_meeting/2018/powervm_driver_meeting.2018-03-27-14.03.html14:49
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/powervm_driver_meeting/2018/powervm_driver_meeting.2018-03-27-14.03.txt14:49
openstackLog:            http://eavesdrop.openstack.org/meetings/powervm_driver_meeting/2018/powervm_driver_meeting.2018-03-27-14.03.log.html14:49
*** arunman has quit IRC15:04
edmondswefried the email I got after registering for the summit has a link for discounted hotels15:33
efriededmondsw: okay, so there's not just one main preferred hotel or whatever.15:34
edmondswefried there are 8 listed15:34
edmondswsummit itself is held at a convention center15:34
efriededmondsw: It's nice to be in the same hotel as other people.  But I guess if it's that spread out.15:34
edmondswagreed15:35
efriededmondsw: Any sense for whether car rental will be practical/necessary?15:35
edmondswprobably the closer the better15:35
edmondswlooks like the 8 hotels are all within a few blocks, so probably only need a rental if we have to book further out15:35
edmondswI have no idea what public transportation is like in Vancouver15:35
edmondswI expect nice but haven't looked15:36
edmondswhttps://www.openstack.org/summit/vancouver-2018/travel/15:36
*** arunman has joined #openstack-powervm16:00
*** arunman has quit IRC16:01
*** arunman has joined #openstack-powervm16:08
*** arunman has quit IRC16:17
*** k0da has quit IRC16:22
*** openstackgerrit has joined #openstack-powervm17:19
openstackgerritTaylor Jakobson proposed openstack/nova-powervm master: Refactor volume drivers  https://review.openstack.org/53081617:19
*** AlexeyAbashkin has quit IRC17:21
openstackgerritTaylor Jakobson proposed openstack/nova-powervm master: Refactor volume drivers  https://review.openstack.org/53081617:21
esbergluefried: edmondsw: mdrabe: I'm still trying to get active RMC for the tempest instances. Any chance it could be an issue that we're doing remote pypowervm?17:54
esbergluI know next to nothing about RMC17:54
efriedYou're saying RMC isn't coming up at all?18:04
esbergluRight18:05
mdrabeesberglu: RMC on what?18:06
esbergluBut I can spawn from the same image on the undercloud and RMC will become active18:06
mdrabeDo the images have the rsct packages installed?18:07
esberglumdrabe: We have a set of neos (CI undercloud) in a multinode openstack that spawn instances18:07
esbergluThese instances then stack AIO devstack (cloud on top of cloud) and but use remote pypowervm from the undercloud neos18:08
esbergluThe AIO devstack then spawns instances for tempest testing purposed18:08
esberglupurposes18:09
esbergluIt's this top level of instances that won't reach active RMC18:09
efriedesberglu: Do we have this same issue OOT?18:09
esbergluThe image has rsct packages installed, and if a spawn an instance on the undercloud can get an active RMC connection18:09
esbergluefried: Yes18:09
efriedoh, okay18:10
esbergluI'm working this on OOT right now not IT18:10
esbergluBut it shouldn't matter18:10
efriedoh, okay18:10
esbergluI thought that the remote pypowervm might have something to do with it since the image works on the undercloud, but not the overcloud18:11
mdrabeesberglu: Was this working before and now all of a sudden it's not?18:14
esberglumdrabe: Nope we've never been waiting for active rmc connections on the overcloud18:14
esbergluBut we need to now because network interface attach requires active rmc18:15
mdrabeAre these NL partitions?18:15
esbergluThe undercloud neos are18:16
esbergluNot the overcloud18:17
esbergluWhich is why we are using remote pypowervm (the AIO instances use remote pypowervm running on the neos)18:17
edmondswI can't see how it would matter that this is remote pypowervm vs. normal18:18
esbergluedmondsw: Idk I'm just spitballing here18:18
edmondswesberglu yeah, understood. It's a good question, but I think the answer is "no"18:19
edmondswwell... it might be something related, though18:20
edmondswthinking18:20
edmondswnah, I got nothing18:21
*** chhagarw has quit IRC18:21
edmondswthe LPARs run on the real NovaLink, not the proxy where we do remote pypowervm, so rsct should be working the same way it always does18:22
edmondswsomething about the host system?18:23
edmondswsince it's not the image18:23
edmondswmdrabe could there be something misconfigured on the NovaLink side rather than on the LPAR rsct side?18:23
esbergluedmondsw: Same novalink though18:28
edmondswhm18:28
edmondswI wonder if someone in the #novalink channel would have better ideas18:29
mdrabeesberglu: Wanna do a meeting about it? Might be better to chit chat18:31
esberglumdrabe: Sure18:32
*** AlexeyAbashkin has joined #openstack-powervm18:35
*** AlexeyAbashkin has quit IRC18:39
*** k0da has joined #openstack-powervm18:42
*** k0da has quit IRC19:03
*** k0da has joined #openstack-powervm19:05
*** edmondsw has quit IRC19:58
*** edmondsw has joined #openstack-powervm19:59
*** edmondsw_ has joined #openstack-powervm20:01
*** edmondsw has quit IRC20:04
*** edmondsw_ has quit IRC20:04
*** edmondsw has joined #openstack-powervm20:04
*** edmondsw has quit IRC20:04
*** edmondsw has joined #openstack-powervm20:59
*** edmondsw_ has joined #openstack-powervm21:00
*** edmondsw has quit IRC21:04
*** edmondsw_ has quit IRC21:05
mdrabeesberglu: Was taking a look at the latest failure on https://review.openstack.org/#/c/555309/3 which has dependency on pypowervm 1.1.1321:21
mdrabeDoes the undercloud need to be updated for that?21:21
esberglumdrabe: Nope pypowervm needs to be bumped to 1.1.13 in openstack/requirements upper-constraints.txt21:23
mdrabeAh k thnx21:24
efriedand lower-constraints in nova-powervm21:25
mdrabeefried: you mean requirements.txt in nova-powervm?21:32
efriedmdrabe: requirements.txt will get bot-updated when we bump g-r in the requirements project.  But there's a shiny new lower-constraints.txt file that we have to update ourselves.21:32
mdrabeefried: It's so shiny I can't find it. It's not under nova-powervm?21:34
*** edmondsw has joined #openstack-powervm21:35
efriedmdrabe: Mahbad: https://review.openstack.org/#/c/555964/21:36
efriednot yet merged.21:36
efriedbecause nova needs to be in pypi first.21:36
efriedSo yeah, for right now we just need to funnel the g-r bump through the requirements project and let the bot propose our requirements.txt update.21:36
*** edmondsw has quit IRC21:36
mdrabeAh okay thnx21:37
*** arunman has joined #openstack-powervm21:41
*** esberglu has quit IRC21:47
*** esberglu has joined #openstack-powervm21:48
*** esberglu has quit IRC21:48
*** esberglu has joined #openstack-powervm21:49
*** esberglu has quit IRC21:53
*** arunman has quit IRC22:01
*** esberglu has joined #openstack-powervm22:01
*** esberglu has quit IRC22:05
*** k0da has quit IRC22:32
*** tjakobs has quit IRC22:48
*** AlexeyAbashkin has joined #openstack-powervm23:12
*** AlexeyAbashkin has quit IRC23:17

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