| *** edmondsw has joined #openstack-powervm | 01:02 | |
| *** edmondsw has quit IRC | 01:07 | |
| *** adreznec has quit IRC | 02:44 | |
| *** adreznec has joined #openstack-powervm | 02:45 | |
| *** tonyb has quit IRC | 02:46 | |
| *** tonyb has joined #openstack-powervm | 02:47 | |
| *** edmondsw has joined #openstack-powervm | 02:51 | |
| *** edmondsw has quit IRC | 02:55 | |
| *** chhagarw has joined #openstack-powervm | 05:29 | |
| *** AlexeyAbashkin has joined #openstack-powervm | 07:46 | |
| *** edmondsw_ has joined #openstack-powervm | 12:11 | |
| *** Alexey_Abashkin has joined #openstack-powervm | 12:33 | |
| *** AlexeyAbashkin has quit IRC | 12:36 | |
| *** Alexey_Abashkin is now known as AlexeyAbashkin | 12:36 | |
| *** efried has joined #openstack-powervm | 12:46 | |
| efried | ō/ | 12:46 |
|---|---|---|
| *** esberglu has joined #openstack-powervm | 13:06 | |
| *** esberglu has quit IRC | 13:20 | |
| *** esberglu has joined #openstack-powervm | 13:30 | |
| *** edmondsw_ is now known as edmondsw | 13:42 | |
| *** tjakobs has joined #openstack-powervm | 13:44 | |
| *** mujahidali has joined #openstack-powervm | 13:54 | |
| edmondsw | #startmeeting PowerVM Driver Meeting | 14:01 |
| openstack | Meeting started Tue May 29 14:01:19 2018 UTC and is due to finish in 60 minutes. The chair is edmondsw. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:01 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:01 |
| *** openstack changes topic to " (Meeting topic: PowerVM Driver Meeting)" | 14:01 | |
| openstack | The meeting name has been set to 'powervm_driver_meeting' | 14:01 |
| edmondsw | #link agenda: https://etherpad.openstack.org/p/powervm_driver_meeting_agenda | 14:01 |
| edmondsw | #topic In-Tree Driver | 14:01 |
| *** openstack changes topic to "In-Tree Driver (Meeting topic: PowerVM Driver Meeting)" | 14:01 | |
| edmondsw | pinging esberglu efried gman_tx mdrabe mujahidali | 14:02 |
| esberglu | vSCSI is out of the runway. Unfortunate that our runway slot was during the summit / memorial day | 14:02 |
| edmondsw | that counted? | 14:03 |
| esberglu | Yep | 14:03 |
| edmondsw | that's some feedback on the runway process... PTG and summit weeks should not count | 14:03 |
| edmondsw | efried ^ | 14:03 |
| esberglu | But all issues have been addressed and CI is passing, so I'm not worried about getting it in | 14:03 |
| efried | Yeah, there was some chatter about dispensations for summit week. Guess that didn't fly. | 14:04 |
| edmondsw | yeah, in a runway or not, we should ask mriedem and melwitt to take a look and confirm their issues are addressed | 14:04 |
| *** gman-tx has joined #openstack-powervm | 14:05 | |
| esberglu | I won't be able to do any more in-tree work this release (resize/mig) so that will probably slide until next release | 14:05 |
| edmondsw | esberglu understood | 14:05 |
| esberglu | And we don't have CI yet for that anyways | 14:05 |
| edmondsw | esberglu I'm hoping you'll still be able to finish the CI work for that | 14:05 |
| edmondsw | but we can talk about that when we get to CI | 14:05 |
| edmondsw | I'm not giving up yet on some of the resize/mig work getting in if we can get CI done | 14:06 |
| edmondsw | maybe efried or I can pick that up (the porting, not CI) | 14:06 |
| esberglu | That's all I have for in-tree | 14:06 |
| edmondsw | oh, forgot to link | 14:07 |
| edmondsw | #link https://etherpad.openstack.org/p/powervm-in-tree-todos | 14:07 |
| edmondsw | #topic Out-of-Tree Driver | 14:07 |
| *** openstack changes topic to "Out-of-Tree Driver (Meeting topic: PowerVM Driver Meeting)" | 14:07 | |
| edmondsw | #link https://etherpad.openstack.org/p/powervm-oot-todos | 14:07 |
| edmondsw | chhagarw gman-tx can one of you update us on iSCSI? | 14:08 |
| edmondsw | I think we're waiting on DevStack testing for https://review.openstack.org/#/c/567575/ | 14:08 |
| edmondsw | I know chhagarw was trying to set that up last week and had some issues | 14:09 |
| edmondsw | esberglu I pointed her to you for help... did that happen? | 14:09 |
| esberglu | Yeah she's had bad luck and hit about every issue I know stacking, still working on it today | 14:10 |
| edmondsw | alright tx | 14:10 |
| edmondsw | I talked to efried about what we need to do for MSP support | 14:11 |
| edmondsw | and we've got some patches in flight related to tox configs | 14:12 |
| edmondsw | and lower constraints jobs | 14:12 |
| edmondsw | we've been testing networking-powervm against a queens neutron release instead of master, which I'm trying to fix | 14:12 |
| edmondsw | etc. | 14:12 |
| edmondsw | another thing that I don't think I've mentioned here before is that not that we have things official with PowerVMStackers we should be able to move our docs to docs.openstack.org from readthedocs | 14:13 |
| edmondsw | lots of docs cleanup we can do | 14:13 |
| edmondsw | I think that's enough on OOT for today unless anyone has anything else | 14:14 |
| esberglu | edmondsw: I'm putting together a list of backports for the merged IT changes | 14:14 |
| edmondsw | esberglu great, thank you | 14:14 |
| edmondsw | that is important | 14:15 |
| edmondsw | added that to the TODO list so I don't forget | 14:15 |
| edmondsw | you can link things there | 14:16 |
| chhagarw | code is done, its been waiting for devstack to complete, facing issues with devstack | 14:16 |
| edmondsw | chhagarw hopefully we can help you get those resolved today/tomorrow | 14:16 |
| edmondsw | #topic Device Passthrough | 14:17 |
| *** openstack changes topic to "Device Passthrough (Meeting topic: PowerVM Driver Meeting)" | 14:17 | |
| edmondsw | efried any update? | 14:17 |
| efried | trying to think whether I had already mentioned: granular has merged on both sides (placement & scheduler). | 14:18 |
| efried | NRP series still needs reviews (no action during summit). | 14:18 |
| efried | Libvirt use of update_provider_tree merged. | 14:18 |
| efried | I think that's it for now. | 14:19 |
| edmondsw | we learned something interesting at the summit... when you attach a NovaLink adapter, you actually have to attach 3 specific PCI devices as a set to get full performance | 14:19 |
| edmondsw | that can probably be abstracted away from Nova but representing the set as a single thing in placement | 14:20 |
| edmondsw | when I say nova, I mean outside the driver | 14:20 |
| efried | You're talking about the GPUs and the NVLink adapters | 14:20 |
| edmondsw | yes | 14:20 |
| edmondsw | sorry... NVLink, not NovaLink | 14:20 |
| efried | I don't know HOW you could have confused those. | 14:20 |
| edmondsw | lol | 14:20 |
| edmondsw | fingers doing their own thing | 14:21 |
| edmondsw | there was a very interesting presentation by NTT (Japanese company) about using GPUs with OpenStack on Power systems if anybody wants to go back and watch the replay | 14:22 |
| edmondsw | and some good conversation in the hall afterward | 14:22 |
| edmondsw | #topic PowerVM CI | 14:22 |
| *** openstack changes topic to "PowerVM CI (Meeting topic: PowerVM Driver Meeting)" | 14:22 | |
| edmondsw | #link https://etherpad.openstack.org/p/powervm_ci_todos | 14:23 |
| esberglu | vSCSI CI is up but for master branch only | 14:23 |
| esberglu | I've been having mujahidali work on the stable branches | 14:23 |
| edmondsw | (OOT) | 14:24 |
| esberglu | huh? | 14:24 |
| esberglu | Oh the OOT stable branches | 14:24 |
| esberglu | Yeah | 14:24 |
| edmondsw | right | 14:24 |
| mujahidali | I am trying to enable vscsi on stable branches but I got the error: etcd3 not found.I browse to the location and found that 3.1 version of etcd3 was not there | 14:24 |
| edmondsw | no IT since we didn't have vSCSI IT for queens a prev | 14:24 |
| mujahidali | https://github.com/coreos/etcd/releases | 14:24 |
| esberglu | mujahidali: I haven't had a chance to look at that issue yet | 14:26 |
| esberglu | mujahidali: Do you have ETCD_VERSION set in your local.conf? | 14:27 |
| esberglu | edmondsw: Regarding multinode CI. I've been getting pressured from my new position to close all current work | 14:27 |
| esberglu | In our meeting a few weeks ago we agreed that I would deliver vSCSI CI and help mujahidali finish up multinode | 14:28 |
| edmondsw | yep | 14:28 |
| esberglu | I can maybe sacrifice a few hours here or there but that's it | 14:28 |
| edmondsw | that's fine as long as it's enough to get things done | 14:29 |
| edmondsw | does mujahidali know how to take that over? | 14:29 |
| esberglu | no | 14:29 |
| mujahidali | esberglu, ETCD_VERSION=v3.2.0-rc.1 but need to figure out why it's picking 3.1.10 | 14:29 |
| esberglu | I'll put together some notes about what's left for multinode and send it out to both of you | 14:30 |
| edmondsw | esberglu ok thanks | 14:30 |
| esberglu | mujahidali: I'm going to be redeploying the staging CI environment today, so if you have anything on it you need, save it elsewhere | 14:31 |
| mujahidali | go ahead | 14:31 |
| esberglu | There's a bunch of stuff we can clean up around vSCSI CI, I've added all of it to the etherpad so I won't go into it here | 14:32 |
| edmondsw | agreed, tx | 14:32 |
| esberglu | Thats all | 14:33 |
| edmondsw | esberglu thanks. And great job getting vSCSI done | 14:33 |
| esberglu | thanks | 14:33 |
| edmondsw | #topic Open Discussion | 14:33 |
| *** openstack changes topic to "Open Discussion (Meeting topic: PowerVM Driver Meeting)" | 14:33 | |
| edmondsw | anybody have anything? | 14:34 |
| esberglu | nope | 14:34 |
| edmondsw | alright, thanks everyone! | 14:36 |
| edmondsw | #endmeeting | 14:36 |
| *** openstack changes topic to "This channel is for PowerVM-related development and discussion. For general OpenStack support, please use #openstack." | 14:36 | |
| openstack | Meeting ended Tue May 29 14:36:26 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 14:36 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/powervm_driver_meeting/2018/powervm_driver_meeting.2018-05-29-14.01.html | 14:36 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/powervm_driver_meeting/2018/powervm_driver_meeting.2018-05-29-14.01.txt | 14:36 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/powervm_driver_meeting/2018/powervm_driver_meeting.2018-05-29-14.01.log.html | 14:36 |
| edmondsw | chhagarw what's your current issue with DevStack? | 14:36 |
| *** mujahidali has quit IRC | 14:45 | |
| chhagarw | mysqld not running | 14:59 |
| chhagarw | stack failed since rabbitmq failed because mysqld is not working | 15:02 |
| *** tjakobs has quit IRC | 15:13 | |
| *** tjakobs has joined #openstack-powervm | 15:17 | |
| edmondsw | esberglu ^ you know anything about that? | 15:39 |
| esberglu | chhagarw: Did you try restarting mysql? | 15:39 |
| *** AlexeyAbashkin has quit IRC | 15:49 | |
| efried | esberglu, edmondsw: FYI, the pike backport of the UUID stringification is merging. That allows us to remove a patch from our env, yah? | 16:09 |
| esberglu | edmondsw: efried: 6589 | 16:13 |
| esberglu | You guys can +2, I'll wait to merge it until the nova patch merges | 16:14 |
| *** mujahidali has joined #openstack-powervm | 16:57 | |
| chhagarw | tried, but each time rabbitmq is restart mysql went down | 17:05 |
| *** prashkre has joined #openstack-powervm | 17:32 | |
| *** mujahidali has quit IRC | 17:33 | |
| *** openstackgerrit has joined #openstack-powervm | 17:58 | |
| openstackgerrit | Matthew Edmonds proposed openstack/networking-powervm master: Test our master with neutron's master https://review.openstack.org/570515 | 17:58 |
| *** prashkre has quit IRC | 17:59 | |
| *** prashkre has joined #openstack-powervm | 18:05 | |
| *** prashkre_ has joined #openstack-powervm | 18:07 | |
| *** prashkre has quit IRC | 18:10 | |
| *** prashkre_ has quit IRC | 18:25 | |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm master: Test our master with neutron's master https://review.openstack.org/570515 | 18:33 |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm master: add lower-constraints job https://review.openstack.org/555964 | 18:36 |
| efried | edmondsw: ^ ftfy | 18:36 |
| efried | edmondsw: Out of curiosity, going to let that spin again without trying to "fix" the pep8 stuff yet, in case those "failures" are somehow caused by same weirdness. | 18:37 |
| efried | edmondsw: Did you ever get resolution on the pathing thing? Is what we're doing now (using git.o.o) the right way? | 18:37 |
| edmondsw | efried yeah I'm not sure what's causing the pep8 issues all the sudden. They aren't neutron-specific pep8 issues, so why would the neutron version matter? It's weird | 18:46 |
| edmondsw | as for using git.o.o... the reason I changed to using that is because fungi said we should | 18:46 |
| edmondsw | so I did... didn't intend to dig into it any more than that | 18:46 |
| efried | edmondsw: I may be misremembering, but I want to say new hacking rules were introduced recently. And if we were using Q and are now using R, that'd do it. | 18:46 |
| efried | edmondsw: Yeah, I have no beef with using git.o.o, was just checking whether that was The Final Word or if we were still expecting a fup from fungi. | 18:47 |
| edmondsw | yeah I need to find where pep8 checks are excluded and see if that changed | 18:47 |
| edmondsw | efried ok, neutron has definitely been fixing pep8 issues and removing things from the ignore line in tox.ini | 18:49 |
| edmondsw | so I'll have to go fix that | 18:49 |
| edmondsw | even the one where pep8 is actually wrong, I guess :( | 18:50 |
| efried | edmondsw: Cool beans. And nice proof that we're dtrt now. | 18:50 |
| edmondsw | first error complains about import ordering where the import ordering is actually correct | 18:50 |
| efried | oh? Looking... | 18:50 |
| edmondsw | networking_powervm/plugins/ml2/drivers/mech_pvm_base.py:20:1: I100 Import statements are in the wrong order. from neutron.plugins.ml2 should be before from neutron.plugins.ml2.drivers | 18:51 |
| edmondsw | because that from actually imports rpc, which should come after drivers | 18:51 |
| edmondsw | I'm pretty sure I've seen that handled correctly other places, so it was odd to see it complain here | 18:51 |
| efried | edmondsw: I must be looking at something wrong. | 18:54 |
| edmondsw | ? | 18:54 |
| efried | okay, you threw me off when you said "first error". | 18:55 |
| edmondsw | efried sorry... that was the first when I ran locally | 18:55 |
| efried | k | 18:55 |
| edmondsw | I don't know if the ordering is the same with zzul | 18:55 |
| edmondsw | zuul | 18:55 |
| efried | yeah, that one ought to be correct. | 18:55 |
| efried | weird | 18:55 |
| edmondsw | efried can you fix these pep8 issues and rebase onto that? | 18:56 |
| efried | edmondsw: As in, you want me to make the patch that fixes the errors? | 18:56 |
| edmondsw | yep, if you would | 18:57 |
| efried | okay, on it. | 18:57 |
| edmondsw | tx | 18:57 |
| edmondsw | esberglu https://review.openstack.org/#/c/570515/ is going to revert something you did... let me know if you have any concerns there | 18:59 |
| edmondsw | I think the community wanted us to do that, but turns out it doesn't work | 19:00 |
| *** chhagarw has quit IRC | 19:00 | |
| *** AlexeyAbashkin has joined #openstack-powervm | 19:06 | |
| *** AlexeyAbashkin has quit IRC | 19:10 | |
| *** AlexeyAbashkin has joined #openstack-powervm | 19:11 | |
| esberglu | edmondsw: I'm fine with that revert. You're failing pep8 though | 19:15 |
| efried | esberglu: I'm working on that. | 19:17 |
| efried | edmondsw: I think the current l-c failure actually found an incompatible constraint :) | 19:17 |
| efried | as in, actually doing what it's supposed to. | 19:18 |
| efried | edmondsw: Did you get this while trying to recreate those pep8 failures locally? | 19:23 |
| efried | AttributeError: 'int' object has no attribute 'endswith' | 19:23 |
| efried | never mind | 19:23 |
| edmondsw | oh nice... what's the incompatibility? | 19:27 |
| efried | edmondsw: Looks like sqlalchemy. | 19:32 |
| efried | but I'm not positive yet. | 19:32 |
| efried | I'll need to figure out if I can run the l-c job locally. | 19:32 |
| efried | wasn't there some problem with that? | 19:32 |
| *** AlexeyAbashkin has quit IRC | 19:34 | |
| edmondsw | efried yeah, can't really be done | 19:34 |
| efried | edmondsw: Yeah, these pep failures are totally wrong. Supposed to be in sections - base, 3rd party, then local - right? | 19:36 |
| edmondsw | yes | 19:38 |
| edmondsw | you seeing them complain otherwise? | 19:39 |
| edmondsw | as for the sqlalchemy thing... somehow we pulled in 1.0.10 http://logs.openstack.org/15/570515/3/check/openstack-tox-lower-constraints/6c98b40/job-output.txt.gz#_2018-05-29_18_47_51_581449 | 19:39 |
| edmondsw | even though neutron master requires >= 1.2.0 https://github.com/openstack/neutron/blob/master/requirements.txt#L21 | 19:40 |
| edmondsw | I will say it's really silly of them to list all those != that are anyway <1.2.0 | 19:40 |
| edmondsw | but still, how did we get 1.0.10? | 19:40 |
| edmondsw | efried ^ | 19:40 |
| edmondsw | looks like sqlalchemy added pending_to_persistent in 1.1.0 | 19:41 |
| edmondsw | oh, our lower-constraints.txt says 1.0.10... so that's wrong, because that can't be lower than what neutron requires | 19:42 |
| efried | edmondsw: right. Need to go through and update things that have moved since this patch was generated. Probably would be best to run the tools again. But that may be more trouble than worth. | 19:44 |
| edmondsw | I'll fix that while you worry about pep8 | 19:45 |
| edmondsw | already found another we'll need to update as well | 19:45 |
| efried | edmondsw: Roger that. Meanwhile... I'm trying to figure out why these bogus import errors are showing up. I don't see any big patch that "fixed" neutron for this. But I also don't see them ignoring the errors. Still playing. | 19:46 |
| esberglu | efried: edmondsw: Procedural -2 was dropped from vSCSI | 19:47 |
| efried | good | 19:47 |
| edmondsw | yay | 19:47 |
| efried | edmondsw: When I try to run pep8 under py2.7, I get: | 19:48 |
| efried | Certificate did not match expected hostname: git.openstack.org. Certificate: {'notAfter': 'May 19 23:59:59 2019 GMT', 'subjectAltName': (('DNS', 'git.airshipit.org'), ('DNS', 'www.git.airshipit.org')), 'subject': ((('organizationalUnitName', u'Domain Control Validated'),), (('organizationalUnitName', u'PositiveSSL'),), (('commonName', u'git.airshipit.org'),))} | 19:48 |
| efried | Could not install packages due to an EnvironmentError: HTTPSConnectionPool(host='git.openstack.org', port=443): Max retries exceeded with url: /cgit/openstack/requirements/plain/upper-constraints.txt (Caused by SSLError(CertificateError("hostname 'git.openstack.org' doesn't match either of 'git.airshipit.org', 'www.git.airshipit.org'",),)) | 19:48 |
| efried | that looks like something in your wheelhouse | 19:48 |
| edmondsw | yeah... let me check | 19:49 |
| edmondsw | efried that's a totally valid error. The cert returned for git.openstack.org is not valid for git.openstack.org | 19:50 |
| efried | edmondsw: Wonderful. So why does it work in py3? And is there a way I can get around it? | 19:51 |
| edmondsw | I have no idea why it would work in py3 or how to work around it besides going back to using github | 19:54 |
| edmondsw | maybe they're doing maintenance on git.o.o and the cert was just changed? | 19:54 |
| efried | I seem to recall getting this a week or two ago as well. | 19:54 |
| edmondsw | this cert was only generated 10 days ago | 19:54 |
| edmondsw | May 19 00:00:00 2018 GMT | 19:55 |
| edmondsw | can you complain on infra? | 19:55 |
| efried | I can. Though it would help to have backup from someone who can spell SSL. | 19:56 |
| edmondsw | I can do that. I just can't say exactly how you hit it | 19:56 |
| edmondsw | I didn't hit it myself | 19:56 |
| edmondsw | but I think I was running under py3 | 19:56 |
| edmondsw | why did you change that? | 19:57 |
| efried | edmondsw: Experimenting, trying to get these import errors to go away without actually "fixing" them (to be wrong). | 19:58 |
| efried | edmondsw: Trying to match the neutron env by stages. The neutron tox.ini uses py2.7 for the pep8 env. | 19:58 |
| edmondsw | efried esberglu any idea why networking-powervm requires a newer version of oslo.config than neutron does? | 19:58 |
| *** tjakobs has quit IRC | 19:59 | |
| efried | edmondsw: to what? | 20:00 |
| efried | Current version looks to have been triggered by I4bf5aaf454e2187978571667a554926b634c13ec | 20:00 |
| efried | possible that neutron quit paying attention to g-r before n-p did? | 20:01 |
| edmondsw | that could be | 20:02 |
| edmondsw | yeah, ours was updated by bot in March: https://github.com/openstack/networking-powervm/commit/0c40c0281b5f7ee88e0c9e6af83610eeb1a305f2 | 20:04 |
| edmondsw | efried I think I'm leaning toward changing back to be consistent with neutron... thoughts? | 20:04 |
| efried | I guess if it passes CI... | 20:05 |
| efried | Need to do that in a separate patch tho | 20:05 |
| efried | Certainly l-c can have the lower version in the l-c patch. | 20:06 |
| efried | but for the main requirements.txt should be different patch | 20:06 |
| edmondsw | efried I think the check job will make us change them together | 20:07 |
| efried | edmondsw: if requirements.txt is higher than l-c.txt? No. | 20:08 |
| *** tjakobs has joined #openstack-powervm | 20:09 | |
| efried | edmondsw: You pushing a new rev some time soon? | 20:11 |
| edmondsw | efried why make this a separate change? | 20:11 |
| efried | edmondsw: Oh, traceability and revertability. | 20:11 |
| efried | It's not related to lower constraints. | 20:12 |
| edmondsw | this patch isn't all about lower constraints | 20:12 |
| edmondsw | it's about testing on master | 20:12 |
| edmondsw | there are already other changes in requirements.txt and in tox.ini | 20:12 |
| efried | edmondsw: You're right. We should do the testing-on-master stuff in one patch, and the l-c stuff in a separate patch. | 20:13 |
| efried | but... we already have l-c stuff merged, huh? | 20:13 |
| efried | It's just... a separate change. | 20:13 |
| edmondsw | yeah, so you've got a chicken-egg problem to split this | 20:13 |
| edmondsw | if we take all the l-c stuff out of this patch, it won't pass tox | 20:14 |
| edmondsw | it's not a separate change | 20:14 |
| efried | Testing against neutron master is one thing. Fixing up reqs to match neutron master *if they break* is related (and inextricable). Fixing up reqs to match neutron master for no reason other than because they're different -- that's a different change. | 20:14 |
| edmondsw | it could be, but... whatever | 20:16 |
| edmondsw | extra work for no purpose | 20:16 |
| openstackgerrit | Matthew Edmonds proposed openstack/networking-powervm master: Test our master with neutron's master https://review.openstack.org/570515 | 20:16 |
| efried | edmondsw: I'm just anticipating needing to revert that sucker. It's a cleaner paper trail when you just punch the revert button rather than doing a custom partial-revert where your reviewers (and future we) have to dig into the original's commit message, review history, and code to figure out what it was about and which part is being reverted and why that and not the rest. | 20:19 |
| edmondsw | I just can't see why you'd anticipate that. I see zero chance of that happening | 20:19 |
| edmondsw | but i left it out | 20:20 |
| *** AlexeyAbashkin has joined #openstack-powervm | 20:20 | |
| efried | I'll be delighted if the revert never happens. | 20:20 |
| edmondsw | you may want me to have left out l-c changes to ovsdbapp and oslo.config as well for the same reason? | 20:21 |
| edmondsw | those are the 2 where I actually reduced l-c | 20:21 |
| efried | reduced it to match master neutron's l-c? | 20:23 |
| efried | Okay, you're right, it's really not worth it. Chances of having to revert all (vs. just part) of either change are basically nil. Just do it all together. | 20:24 |
| *** AlexeyAbashkin has quit IRC | 20:25 | |
| edmondsw | efried tx | 20:29 |
| efried | edmondsw: Do you have local edits yet? I have the pep8 fix ready and based under the other. | 20:29 |
| openstackgerrit | Matthew Edmonds proposed openstack/networking-powervm master: Test our master with neutron's master https://review.openstack.org/570515 | 20:29 |
| edmondsw | just pushed | 20:29 |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm master: Test our master with neutron's master https://review.openstack.org/570515 | 20:31 |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm master: Fix pep8 before moving to master https://review.openstack.org/571017 | 20:31 |
| efried | edmondsw: k, I think I did this right. ^ | 20:31 |
| efried | yeah, PS6 shows rebase only. | 20:31 |
| efried | edmondsw: Note the main thing in tox.ini was adding that import-order-style thing - that got rid of all the bogus errors. The rest were legit (things were in the wrong sections). | 20:32 |
| edmondsw | efried nice | 20:32 |
| efried | edmondsw: Now we have to decide how we're going to +A these things, since we've both had our hands all over all of them. | 20:33 |
| efried | Calling mdrabe... | 20:33 |
| mdrabe | Hello | 20:33 |
| efried | oh, he's not a networking-powervm core | 20:33 |
| mdrabe | bye bye | 20:34 |
| efried | edmondsw: Wonder if we should check https://review.openstack.org/#/c/555964/ against nova's latest l-c | 20:34 |
| efried | mdrabe: Don't leave yet -- we'll need you to +A ^ | 20:34 |
| efried | edmondsw: I don't think I've seen any l-c bumps go by in nova lately. | 20:35 |
| efried | okay, no, the only update is to wsgi-intercept, which we don't use. So we should be good there. mdrabe, would you +A that sucker please? | 20:37 |
| edmondsw | efried I think you can +2 570515, that was mostly me, and then we can get someone else to +A | 20:37 |
| efried | edmondsw: Done. Maybe we can get Doug and/or Jeremy to +1 and accept that as sufficient to merge. | 20:38 |
| edmondsw | probably ought to wait to make sure zuul likes these before bugging mdrabe :) | 20:38 |
| efried | what could possibly go wrong? | 20:39 |
| mdrabe | efried: Where'd all these lower constraints values come from? | 20:43 |
| edmondsw | mdrabe I updated our lower-constraints to match what I found in neutron's requirements.txt (master branch) | 20:50 |
| edmondsw | if you're talking about 570515 | 20:50 |
| edmondsw | if you mean in general, lower-constraints was based off the min value in global-requirements originally, though it has probably diverged since | 20:51 |
| edmondsw | mdrabe but look at https://review.openstack.org/#/c/571017/ first | 20:52 |
| mdrabe | edmondsw: Was looking at 555964, I asked because the values don't appear to be the same as nova's | 20:53 |
| mdrabe | So I was just wondering where they came from | 20:53 |
| mdrabe | edmondsw: https://review.openstack.org/#/c/571017 LGTM, but I can only +1 there | 20:54 |
| openstackgerrit | Matthew Edmonds proposed openstack/networking-powervm master: Test our master with neutron's master https://review.openstack.org/570515 | 20:55 |
| edmondsw | mdrabe oh, drat, that's networking-powervm not nova-powervm | 20:55 |
| edmondsw | yeah | 20:55 |
| edmondsw | I've asked svenkat to look at the networking ones | 20:59 |
| edmondsw | mdrabe as for 555964... yeah, we should compare to nova, let me look | 21:00 |
| esberglu | edmondsw: Sorry I haven't been following along. But I can W+1 networking and nova-powervm changes | 21:00 |
| edmondsw | esberglu was trying to spare you :) | 21:01 |
| edmondsw | I know you're tied up in new things | 21:01 |
| mdrabe | edmondsw: I'll leave a comment on the review | 21:01 |
| edmondsw | mdrabe tx... I've gotta run but will look later | 21:01 |
| esberglu | edmondsw: Just 571017 for now? | 21:01 |
| edmondsw | esberglu and then 570515 | 21:02 |
| edmondsw | svenkat is also looking | 21:02 |
| esberglu | I'll just leave it then | 21:02 |
| esberglu | edmondsw: I've got the staging CI forcing the subnode to be on the same host as the main node | 21:06 |
| esberglu | It's going to make our nodepool configuration super groos | 21:06 |
| esberglu | gross | 21:06 |
| esberglu | Well I guess not that gross, just tons of duplication | 21:07 |
| *** edmondsw has quit IRC | 21:13 | |
| openstackgerrit | Merged openstack/networking-powervm master: Fix pep8 before moving to master https://review.openstack.org/571017 | 21:23 |
| openstackgerrit | Merged openstack/networking-powervm master: Test our master with neutron's master https://review.openstack.org/570515 | 21:23 |
| *** tjakobs has quit IRC | 21:37 | |
| esberglu | efried: They bumped out the runway end date because of the summit | 21:47 |
| efried | yup | 21:47 |
| esberglu | So vSCSI should get in this week | 21:47 |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm master: add lower-constraints job https://review.openstack.org/555964 | 21:52 |
| efried | mdrabe: ^ | 21:52 |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm master: add lower-constraints job https://review.openstack.org/555964 | 21:52 |
| efried | rebased ^ | 21:53 |
| efried | We should totally do a sweep to see which of these silly things we actually need. | 21:53 |
| efried | I can't imagine we're actually using e.g. "snowballstemmer" (wtf is that anyway) | 21:53 |
| efried | ahdang, I wonder if there's even any reason for us to be listing anything we don't directly require in requirements.txt/test-requirements.txt | 21:54 |
| *** gman-tx has quit IRC | 21:58 | |
| efried | apparently there is. | 22:04 |
| *** esberglu has quit IRC | 22:10 | |
| *** gman-tx has joined #openstack-powervm | 22:17 | |
| *** gman-tx has quit IRC | 22:23 | |
| efried | mdrabe: I've gotten to a point where I don't know wtf to do with these requirements anymore. | 22:28 |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm master: add lower-constraints job https://review.openstack.org/555964 | 22:31 |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm master: add lower-constraints job https://review.openstack.org/555964 | 22:31 |
| efried | let's try that | 22:32 |
| efried | shrug | 22:32 |
| *** openstackstatus has joined #openstack-powervm | 23:43 | |
| *** ChanServ sets mode: +v openstackstatus | 23:43 | |
| *** gman-tx has joined #openstack-powervm | 23:45 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!