Thursday, 2026-06-18

carloss#startmeeting manila15:00
opendevmeetMeeting started Thu Jun 18 15:00:04 2026 UTC and is due to finish in 60 minutes.  The chair is carloss. Information about MeetBot at http://wiki.debian.org/MeetBot.15:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.15:00
opendevmeetThe meeting name has been set to 'manila'15:00
carlosscourtesy ping: dviroel gouthamr carthaca Sai ashrodri gireesh Kumar_T15:00
gouthamro/15:00
carlossAnoop_Shukla dari-c Silvia-Wachira15:00
carthacahi15:02
gireesho/15:02
* fungi is around again15:02
Anoop_ShuklaO/15:03
carlosshello folks and welcome!15:03
carlossour meeting agenda for today:15:03
carloss#link https://wiki.openstack.org/wiki/Manila/Meetings (Meeting agenda)15:04
Silvia-Wachirao/15:04
carloss#topic Announcements15:04
carlossSchedule and Deadlines15:04
carloss#link https://releases.openstack.org/hibiscus/schedule.html (Hibiscus Schedule)15:04
carlossnow we have all dates in the schedule (i mentioned this in the previous meeting as well)15:04
carlossspec freeze was last week15:04
carlossbut15:05
carlossI mentioned we'd extend the deadline for this week15:05
carlossto try accommodating more times for reviews and the specs that didn't get much feedback yet15:05
carlossor that were being discussed still15:06
carlosswe'll go through specs in a few15:06
carlossanother announcement I have for today is about my working windows for the next two weeks15:08
kpdevhi15:08
carlosso/15:08
carlossI'll be on PTO from June 22nd through July 6th, returning on the 7th.  Meaning you won't see me around much in the next two weeks. I'd like to ask you if you feel we should keep the next two weekly meetings,  skip one of them or skip both?15:09
carlossI've already asked many favors from gouthamr in the past few times I needed to skip the meeting15:10
carlossand I'm really thankful for the help15:11
carlossI'm just trying not to add many more work to other people15:11
Anoop_Shukla_I think we can keep one of them just to catch up on things that may require discussion?15:11
carlossIf you'd like to help us running the meetings while I'm away, I'll really appreciate that too, and it's a nice experience15:11
carthacaEnjoy your time off. Perfect time to also announce my absence: I will be on PTO from July 4th, returning August 17th15:11
carlossAnoop_Shukla_: yes, that makes sense15:11
carlosscarthaca: thank you and likewise! hope you enjoy your time off too15:11
Anoop_Shukla_Hmm..holiday season is on :)15:12
carlossshould we skip the 25th or the Jul 2nd meeting?15:13
carlossJul 2 is close to the 4th of July holiday15:13
Anoop_Shukla_Jul 2nd I doubt we will have much traction15:13
carlossmeaning folks in the US might have the long weekend ahead15:13
carlossAnoop_Shukla_: yes15:13
carlosslet's keep next week's meeting and cancel the Jul 2nd then15:13
carlossI'll send the email to the mailing list15:13
carlossthank you for weighing in15:14
carlossthat's all I had for announcements. Do you have an announcement to share with us today?15:15
carlosstaking silence as no15:16
carlosslet's talk review focus15:16
carloss#topic Review focus15:16
carloss#link https://etherpad.opendev.org/p/manila-hibiscus-review-focus (Hibiscus Review Focus etherpad)15:17
carlossall eyes on the specs at the moment I guess15:17
carlossthe following specs are still under review:15:19
carloss#link https://review.opendev.org/c/openstack/manila-specs/+/988495 (Specs for share server replica)15:19
carloss#link https://review.opendev.org/c/openstack/manila-specs/+/987622 (Add per-process healthcheck for Manila services)15:19
carloss#link https://review.opendev.org/c/openstack/manila-specs/+/987307 (Add Lustre share driver spec)15:19
carloss#link  https://review.opendev.org/c/openstack/manila-specs/+/985196 (Add VirtioFS testing scenarios)15:19
carlossgireesh: thanks for keeping up with the updates15:19
gireeshThanks Carloss and Goutham for review comments,  I have few questions which I asked to Goutham, if possible Goutham can answer those.15:20
carlossgood reviews on all of the specs. as I noted in the previous weeks, VirtioFS might not qualify for the deadline now, but good to keep the reviews going15:21
carlossgireesh: you're welcome15:22
carlossany other thoughts to be discussed as part of this meeting?15:23
carlossI'd love to get closure on these specs by the weekend15:24
carlossand please let me know if you need more time15:24
carlossit's hard to keep extending spec freeze because it ends up eating up implementation time15:24
sylvanld[m]hello sorry i am late to the party15:26
sylvanld[m]I had trouble to setup the IRC client15:26
carlosssylvanld[m]: no worries15:26
carlosso/15:26
carlosswelcome15:26
sylvanld[m]thx!15:26
gouthamron the specs, i'll respond on gerrit15:27
sylvanld[m]i recently submitted a PR regarding listing of share servers. It has been approved (review+2) but don't have workflow+115:27
sylvanld[m]https://review.opendev.org/c/openstack/manila/+/97553015:27
carlossgouthamr: great, thank you15:27
gouthamrsylvanld[m]: looking15:27
sylvanld[m]thank you!15:28
Anoop_ShuklaSome open comments on the Luster spec - though spec looks in good shape15:28
carlossgouthamr: thanks for looking too15:28
gouthamrAnoop_Shukla: on the Lustre spec, my bad.. been doing a bunch of things and letting that one languish15:28
Anoop_ShuklaOkay..sure goutham15:28
gouthamri can respond to the comments today and close things out.. to be fair, we don't _need_ a spec for a new driver.. i did want to flush out the design in the open15:28
gouthamrand it was super cool to get engagement from the LUSTRE maintainers15:29
gouthamroneswig to thank for that15:29
carloss++15:30
gouthamrthe part that touches core manila is a new LUSTRE protocol.. the code changes for which are trivial15:30
gouthamrand the pattern follows CEPHFS, GLUSTERFS, WEKAFS (being proposed) etc.. 15:30
carloss^ which we have plenty of examples as you noted and requires work, yes, but it's not going to be changing a lot the way we do, so I agree with you15:32
carlossIt's nice that you went the spec route for documenting it15:32
carlossbut I don't think there should be major blockers in terms of manila for the lustre spec15:32
carlosswe've a lot of interest on this too, so it's another motivation15:33
gouthamrindeed15:34
gouthamrseems like an AI enablement theme going, if someone is paying attention :D 15:34
carlossyep15:34
carlossall great specs though and great work on all of them... zorillas rock!15:34
carlosslet's try getting them in15:34
carlossin case there aren't major blockers, of course15:35
carlossstill on review focus15:35
carlosskpdev, I did not forget https://review.opendev.org/c/openstack/manila/+/97878015:35
carlossI'll get back to it15:36
carlossapologies for the delay15:36
carlossand thanks for adding your thoughts Anoop_Shukla_ and carthaca15:36
kpdevack15:36
carlossI'll also get to the other fixes in the review list soon15:36
carlosswill try posting as many reviews as I can before I take off15:37
carlossany other reviews needing attention that we should note as part of $topic?15:38
gouthamrmaybe reprising one that might help new driver authors15:39
gouthamr#link https://review.opendev.org/c/openstack/manila/+/976320 ([doc] Document share driver interfaces) 15:39
carlossyes!15:39
gouthamri had to revisit this when reviewing the weka driver15:39
carlossand that's a great one15:39
gouthamri wanted to know what we can do to make it easier to review and merge15:40
gouthamrcan i split it up into multiple reviews?15:40
carlossuhmm15:40
gouthamrwould be nice to have a few shepherds that can move it along, as well15:40
carlossI think that can help with the readability, yes15:41
gouthamri worked on this to help Srinesh over the last cycle with their HPE driver15:41
gouthamrso there's some valuable feedback in there that i'll address15:41
gouthamrbesides, i think i've carloss and Anoop_Shukla taking a look15:42
carlossyep15:42
carlossoh, there's another set of changes I was about to forget and we definitely need it as early as we can in the cycle:15:43
carloss#link https://review.opendev.org/q/topic:bp/privsep-migration+project:openstack/manila15:43
carlossgouthamr: thanks for being patient with the reviews on the driver interfaces. gireesh can we have your eyes on it too?15:43
Anoop_Shuklasure15:43
gouthamr++15:43
gireesh++15:43
carlosson the privsep changes, gouthamr Anoop_Shukla_ kpdev can I get your eyes on them? I tried making things very modular15:44
carlossand CI should help15:45
carlossonly issue at the moment is that I am not being able to test the container driver, because the CI is badly broken15:45
carlossand I've started looking at how to fix it but no luck yet15:45
Anoop_Shuklaack15:46
carlossthank you15:46
gouthamrack carloss; since you're gone for a bit, i can take over updating these patches15:46
carlossgouthamr: you're very kind to offer that, thank you15:46
carlossalright, a couple of bugs to triage today but I'll try doing it in #openstack-manila15:47
carlossif we have anything that requires more eyes15:47
carlossany other thoughts for $topic?15:47
carloss#topic Bridging the Gap Gazpacho Cycle follow-up discussion (fungi)15:48
fungibriefly revisiting this topic from last week now that folks have hopefully had time to digest it better, i've got a few more points that came out of a closer look into the data but also wanted to see if anyone had additional thoughts15:48
fungione thing i missed calling out, which is exciting, is that the typical time to response on reviews improved drastically in gazpacho compared to flamingo; twice as fast on average (mean) but the median was actually four times faster!15:49
fungii don't know if the team's reviewing habits changed in gazpacho, but that would be something to continue if so as it seems to have made a big impact, both the mean and median were significantly better than in any of the prior 5 cycles15:49
fungiwe had takeaway suggestions, some similar to the previous round... contributors still don't seem to be clear on where maintainers are tracking review priorities, so continuing to work on addressing that communication gap may help15:49
fungialso the change and reviewer numbers are steadily growing while the active maintainer count is falling, so if possible try to find some time to evaluate new maintainer candidates from the pool of established reviewers and code contributors15:49
fungiif anybody has follow-up thoughts or questions on any of the above, as well as for the survey and metrics analysis from last week now that you've had time to mull it over, i'm happy to answer questions and entertain feedback here15:49
fungi(or in the #openstack-manila irc channel after the meeting, or on the openstack-discuss mailing list, or even directly/in private for that matter)15:49
gouthamri think we ought to fix the review prioritization thing asap15:50
gouthamrthe MTTR improvement is great to see! 15:50
fungiyeah, the median time to review on manila changes in gazpacho was 3.24 days15:51
fungivs 12.36 in flamingo15:51
carlossyeah, that is a great enhancement15:51
gouthamrwow15:51
fungigranted, flamingo was rather bad when compared to prior cycles, but the next best of the previos 5 cycles was median 4.82 days back in caracal15:52
Anoop_ShuklaGreat!15:52
gouthamrsuper open to new maintainers, happy to mentor folks to take on that role too.. we did propose two new core reviewers during gazpacho15:53
carlossany other bright ideas collected from other teams on how to communicate the review priorities?15:53
carloss> super open to new maintainers, happy to mentor folks to take on that role too.. we did propose two new core reviewers during gazpacho15:53
carloss++15:53
carlossI think the new cores definitely helped with that enhancement in the numbers15:53
fungiwe had discussed some in the flamingo retrospective, for example in initial review comments from maintainers including a link to the documentation about how and where review priorities are tracked15:54
gouthamrand in general, we'd looove for folks to review; we have a lot of code contributors but less reviewers 15:54
gouthamrfungi: ah yes! 15:54
fungibut also importantly, making sure that the documentation about review prioritization is as discoverable to occasional contributors as it can me15:54
fungier, as it can be15:54
gouthamrit can be you too if you want :P we'll ship a zorilla hoodie your way15:55
gouthamrwe'll work on those two items 15:55
Anoop_ShuklaHoodie is a great idea ;)15:55
carlosshmmm, maybe I could come up with something for manila changes in that direction then... like have something alerting me of new changes so I can go post a comment directing to our docs and15:56
fungiin gazpacho there were 85 people who reviewed at least one change, up from 77 in flamingo and 68 in epoxy, so the reviewer count on the whole is growing, though that doesn't sadly tell us much about their capacity to grow15:56
carloss#link https://etherpad.opendev.org/p/manila-hibiscus-review-focus ?15:56
fungier, about their capacity to grow into maintainers15:56
gouthamrack, we'll see how much further we get in hibiscus.. i think we discussed adopting the reviewer/approver pattern 15:59
carloss++15:59
gouthamrbut, we've actually onboarded cores faster too when we've felt the pinch15:59
gouthamrthanks to folks showing up 15:59
fungiyeah, you had 10 active reviewers back in dalmation, and that's fallen steadily to 7 in gazpacho16:00
carlossagreed16:00
fungier, i mean active maintainers16:00
gouthamrwhat's the definition of that? 16:00
fungi(where "active maintainer" is anyone who exercised a code-review +/-2 or workflow +1 in gerrit on at least one change)16:01
gouthamrack! 16:01
carlossI feel we have even more potential though, we can go further and onboard more people16:01
carlossconsidering the head count of contributors at the moment16:01
carlossthat's great numbers... I'll try coming up with ideas to address the "knowing the priorities" problem16:03
gouthamrwe'll merge your change carloss 16:03
gouthamrand set it as the channel topic on IRC etc16:04
carloss++16:04
carlossthank you16:04
* gouthamr notices it is stale at the moment16:04
carlossand great job to all zorillas... your answers to the survey are definitely helping us in this discussion16:04
carlossand openstack16:04
fungiyes, absolutely, such a huge help!16:04
carlossand thanks for sharing your thoughts and the numbers, fungi 16:05
carlossdefinitely helpful16:05
fungimy pleasure as always16:05
carlosswe're a couple of minutes past the hour... let's continue in #openstack-manila in case there are more topics and questions16:06
carlossthank you everyone for your participation16:06
fungiyeah, feel free to hit me up any time16:06
gouthamrty fungi, everyone!16:06
carlossZorillas ftw!16:06
carlosslet's wrap up16:06
carlosshope you folks have a great day16:06
carlossthank you for participating16:06
carloss#endmeeting16:07
opendevmeetMeeting ended Thu Jun 18 16:07:05 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)16:07
opendevmeetMinutes:        https://meetings.opendev.org/meetings/manila/2026/manila.2026-06-18-15.00.html16:07
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/manila/2026/manila.2026-06-18-15.00.txt16:07
opendevmeetLog:            https://meetings.opendev.org/meetings/manila/2026/manila.2026-06-18-15.00.log.html16:07

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!