15:00:58 #startmeeting manila 15:00:58 Meeting started Thu Sep 23 15:00:58 2021 UTC and is due to finish in 60 minutes. The chair is gouthamr. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:58 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:58 The meeting name has been set to 'manila' 15:01:01 o/ 15:01:04 o/ 15:01:15 o/ 15:01:23 o/ 15:01:26 o/ 15:01:46 o/ 15:01:50 o/ 15:01:52 o/ 15:01:52 0/ 15:01:53 courtesy ping: ganso vkmc tbarron felipe_rodrigues ecsantos 15:01:57 o/ 15:02:49 hello everyone! thank you for joining 15:02:55 here's our agenda for today: https://wiki.openstack.org/wiki/Manila/Meetings#Next_Meeting 15:03:01 #topic Announcements 15:03:23 we shipped the rc1 builds of manila, manila-ui 15:03:59 so far we haven't seen any release critical bugs that warrant an rc2 15:04:13 if you're aware of any, we can discuss them in a bit 15:05:02 several manila contributors are attending the GHC Open Source Day next week 15:05:05 #link https://ghc.anitab.org/programs-and-awards/open-source-day/ 15:06:16 o/ 15:06:27 the open source day is a great opportunity for attendees at GHC who may be either be starting their careers or are mid-career to explore open source communities 15:07:56 the attendees will be going through an overview/onboarding session and will pair-program with us to commit some bug fixes 15:08:27 so if you're aware of any minor issues that can be tackled by new contributors in a couple of hours, please do flag them 15:09:32 and call attention to them on #openstack-manila 15:10:28 that's all the announcements for today, any one else got any? 15:12:15 #topic AIso from Last Week 15:12:19 i think that was autocorrect 15:12:30 #topic AIs for Last Week 15:12:48 * gouthamr checks ethercalc 15:13:38 we were looking to check up on the stale bugs vhari identified 15:13:46 gouthamr, https://ethercalc.openstack.org/kxz9jz7tuopk :) 15:14:04 * gouthamr is unable to open that link, again 15:14:06 :| 15:14:06 I think ethercalc is messing with us :D 15:14:14 oh its not just me 15:14:23 ethercalc-- 15:14:23 gouthamr, I have it open 15:14:32 it was okay for most of the week, but since yesterday afternoon it stopped responding (at least to me) 15:14:34 * fabiooliveira having hard time to open 15:14:45 a minute ago .. and now times out! 15:14:53 omergod 15:15:00 special effect for us! 15:15:03 ethercalc is trolling us 15:15:21 sigh, vhari - if you have it open, is it worth transfering the material to etherpad 15:15:22 i can not open too. 15:15:45 I lost connection after refresh .. will do 15:16:28 ack, thank you 15:17:30 vhari: lets follow up on #openstack-manila when the thing comes back alive 15:17:50 gouthamr, ack 15:17:56 #topic Reviews needing attention 15:18:26 we're not in the yoga cycle yet officially, so lets use the same etherpad as we were last week 15:18:27 #link https://etherpad.opendev.org/p/manila-xena-review-focus 15:19:13 i wanted to check on the status of the python-manilaclient changes at the bottom of the pad 15:19:49 fabiooliveira++ on reviewing these 15:20:15 :D 15:20:16 looks like we're hitting a lower constraints job failure 15:20:39 openstacksdk 0.11.2 depends on decorator>=3.4.0 15:20:39 The user requested (constraint) decorator==3.4.0 15:20:52 that job's also trolling? 15:21:34 #link https://zuul.opendev.org/t/openstack/build/73bfe7613105450fa4adfefe5c9cbb8d (lower constraints job fails on python-manilaclient / trunk branch) 15:22:19 some lower-constraints failures in the client and the manila repo seemed transient last week 15:22:26 but this one persists 15:25:12 we might have to adjust the constraints here as we've done in the past - lets see if we can reproduce this locally 15:26:12 that said, do keep reviewing these patches, and update them 15:26:41 while we resolve this unrelated gate issue 15:26:57 any other reviews that need attention? 15:28:24 there are some changes I think would be nice to have some eyes 15:28:26 doc ones :) 15:28:52 ack, which ones, carloss? 15:28:58 https://review.opendev.org/c/openstack/manila/+/786052 and https://review.opendev.org/c/openstack/manila/+/786048 15:29:12 thanks for reviewing felipe_rodrigues and dviroel 15:29:23 yw 15:30:07 it'd be nice to get these two moving... they are related to the security service update/add changes introduced during wallaby 15:31:07 great thanks for calling them out, and for working on them 15:32:08 i'll take a look in a bit 15:32:13 my pleasure :) 15:32:16 thanks gouthamr 15:32:20 anything else? 15:32:57 #topic Bug Triage (vhari) 15:33:50 gouthamr, thanks .. lets look at a few bugs opened last week 15:33:54 #link https://bugs.launchpad.net/manila/+bug/1942417 15:35:17 ack, this one needs some investigation - and the warning *may* be benign 15:35:46 it was observed in the CI 15:36:06 after we started using the threadgroup approach over our prior fixed interval looping call 15:36:34 manila.service.Service.report_state is expected to run every second, and in some cases, it takes longer 15:37:56 the delay could be anywhere in this https://opendev.org/openstack/manila/src/branch/master/manila/service.py#L264-L303 15:38:12 my initial suspicion was a busy db handle 15:39:52 i think we should triage this as pending investigation 15:40:06 although there's no such status on launchpad 15:41:10 as far as the severity goes, none of our tests surfaced any issue with this 15:41:54 so this warning is probably harmless 15:42:03 and it doesn't occur on all the gate jobs either 15:45:25 guess the bug needs some more details 15:45:58 vhari: can we keep this in "New" and bring it up next week? 15:46:13 gouthamr, sure will do 15:46:25 next up #link https://bugs.launchpad.net/manila/+bug/1943829 15:48:05 hmm, TypeError: unsupported operand type(s) for +: 'NoneType' and 'int' 15:48:07 from https://opendev.org/openstack/manila/src/branch/master/manila/scheduler/filters/capacity.py#L97-L98 15:49:17 this is in the zfsonlinux driver 15:49:33 s/in the/with the 15:50:29 haven't seen this issue before; provisioned_capacity_gb should always be an integer, and default to 0 15:50:46 its possible the driver's reporting 'None' and we're not preventing that in the scheduler 15:52:10 its not very obvious looking at the code 15:52:29 https://opendev.org/openstack/manila/src/branch/master/manila/scheduler/host_manager.py#L456-L462 15:54:19 anyone looking to investigate this bug? you may not need to reproduce it as much as check if there's any place this "provisioned_capacity_gb" has the potential of being set to None 15:55:00 gouthamr, stale bug list is backup https://ethercalc.openstack.org/kxz9jz7tuopk 15:55:14 gouthamr, copied contents here https://etherpad.opendev.org/p/Manila-Xena-Cycle-Stale-Bug-List 15:55:31 great thank you vhari 15:55:39 gouthamr, yw 15:56:38 alright, i'd suggest setting the zfsonlinux bug to "Medium" and changing the bug title to "Provisioned capacity GB is mishandled in the scheduler" 15:57:42 lets chat about the stale bugs on #openstack-manila 15:57:51 any other new bugs for today, vhari? 15:58:01 gouthamr, that's all for today 15:58:13 thank you 15:58:16 #topic Open Discussion 15:58:22 gouthamr, I added links to stale bugs on our usual bug etherpad https://etherpad.opendev.org/p/manila-bug-triage-pad-new 15:58:33 #link https://etherpad.opendev.org/p/manila-bug-triage-pad-new 15:58:52 ++ 15:59:39 if you're looking for a heart melt 15:59:41 #link https://www.nps.gov/katm/learn/fat-bear-junior-2021.htm 16:01:07 my grandmother would think those stonks can surely eat some more fish 16:01:23 that's a wrap for today 16:01:31 thank you all for attending, see you on #openstack-manila 16:01:39 #endmeeting