Thursday, 2026-07-16

carloss#startmeeting manila15:00
opendevmeetMeeting started Thu Jul 16 15:00:21 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 vhari carthaca Sai ashrodri gireesh Kumar_T15:00
carlossAnoop_Shukla dari-c Silvia-Wachira15:01
vharihi 15:01
Kumar_THi15:01
SaiI/15:03
Saio/15:03
carlosso/15:03
carlosssmaller quorum today15:03
carlosslet's get started15:03
carlossand we'll likely wrap up early and get back to our bugs!15:03
carloss#link https://wiki.openstack.org/wiki/Manila/Meetings (Meeting)15:03
carloss#undo15:04
opendevmeetRemoving item from minutes: #link https://wiki.openstack.org/wiki/Manila/Meetings15:04
carloss#link https://wiki.openstack.org/wiki/Manila/Meetings (Meeting agenda)15:04
carlossour meeting agenda for today15:04
carlosslet's get started15:04
carloss#topic Announcements15:04
itaiweismanwekaם/15:04
itaiweismanwekao/15:04
carlossSchedule and Deadlines15:04
carlosso/ itaiweismanweka 15:04
carlosswelcome15:04
kpdevhi15:05
carlosso/ kpdev 15:05
carloss#link https://releases.openstack.org/hibiscus/schedule.html (Hibiscus Schedule)15:05
sakumbha_o/15:05
carlosshey there sakumbha_15:05
carlossso this is our bugsquash week15:05
carlossand we've been firing on all cylinders on bug fixes reviews and pushing patches15:05
carlossthank you everyone that is participating15:06
carlosswe'll focus more on that towards the second part of this meeting15:06
vhari++15:06
carlossfeature proposal freeze and new driver deadlines are in two weeks15:06
carlossfor feature proposal freeze: all new Manila features must be proposed and substantially completed, with unit, functional and integration tests by the end of the week. Collaborative review sessions will happen in the mid cycle, in order to speed up the review process.15:07
carlossfor the new driver deadline: by the end of the week all new backend drivers for Manila must be substantially complete, with unit tests, and passing 3rd party CI. Drivers do not have to actually merge until feature freeze.15:07
carlossthese two deadlines are two weeks from now15:07
carlossjust a reminder: features and drivers must be proposed to gerrit15:07
carlossbut they have until feature freeze to merge15:07
carlossi think we're in a good position at the moment, considering all features and so on15:08
carlossthat's all I had in terms of our schedule15:08
carlossif you have any questions about such deadlines or need more time, please let me know in advance in #openstack-manila15:08
carlossthat said15:08
carlossI don't have anything else to share for our schedule15:09
carlossI have one other announcement for today that I did not add to the agenda but I think it's worth sharing15:09
carlossI'd like to introduce you to sakumbha_ 15:09
carlossSahil is joining our team at Red Hat and will be focusing in Manila15:09
carlossyou've seen Sahil active in the bugsquash changes and chatter already15:10
carlossso welcome sakumbha_, the zorilla team is lucky to have you and we're sure you'll have fun with us15:10
sakumbha_Hey everyone, happy to join zorillas !15:11
Anoop_ShuklaO/15:12
* vhari brb15:12
carlossalright, that's all I had for $topic. is there an announcement you'd like to share with us today?15:14
carlosstaking silence as no15:16
carlosslet's leave our bugsquash for the last slot15:17
carloss#topic Review Focus15:18
carloss#link https://etherpad.opendev.org/p/manila-hibiscus-review-focus (Hibiscus Review Focus etherpad)15:18
carlossitaiweismanweka thanks for adding a topic to the agenda15:18
carlossWeka share driver PS12 — re-review request on change 989998, CI green15:18
itaiweismanwekao/ yes, that's me! PS12 addresses all PS11 feedback — eventlet replaced with native threads, CI green (903 tests, 0 failures). Would appreciate a re-review from gouthamr or Stig. We're tracking toward the July 30 deadline.15:19
carlossthat's awesome progress15:20
carlossI haven't looked yet but I am planning too15:20
carlossI had a quick exchange with gouthamr about it 15:20
itaiweismanwekathanks carloss! much appreciated. Any timeline in mind, given the July 30 deadline?15:20
carlosswhere we also discussed limitations for some types of access control15:20
carloss> thanks carloss! much appreciated. Any timeline in mind, given the July 30 deadline?15:21
carlossI think you folks are good honestly. Code is proposed, CI passing15:21
carlossunit tests implemented15:21
carlossthat's all we'd expect for the upcoming July 30th deadline15:21
itaiweismanwekagreat to hear, thank you! You mentioned limitations for some types of access controlת  could you elaborate? We want to make sure we address any concerns.15:23
carlossyeah, so iirc, it's harder to place access control to shares with WEKA protocol15:24
carlossor I am confusing things with LUSTRE15:24
carlossnvm I think I am mixing things up15:25
carlosswe intend to add this topic as part of the future weekly meetings, likely the next one15:25
carlossHPE folks might be interested on the discussion as well15:25
carlossyeah, I got it right, it's related to access rules15:26
carlossso we can discuss this in the upcoming weekly meetings15:27
carlossis there something else you folks would like to add for $topic?15:28
itaiweismanwekajust to clarify,  for WekaFS protocol, access rules being no-ops is by design, not a limitation. Access control is handled natively at the Weka cluster level. Happy to discuss further in future meetings if helpful!15:28
carlosswe'll cover bug reviews in the bugsquash topics :)15:28
carloss> just to clarify,  for WekaFS protocol, access rules being no-ops is by design, not a limitation. Access control is handled natively at the Weka cluster level. Happy to discuss further in future meetings if helpful!15:28
carlossyep, that's the thing we'd like to expand on15:28
carloss#topic Hibiscus Bug Squash15:31
carloss#link https://etherpad.opendev.org/p/manila-hibiscus-m1-bugsquash (Hibiscus bug squash)15:31
carlosswe are making decent progress on the bugsquash15:32
carlossvhari: was telling me at the beginning of the meeting that we already managed to get in progress bugs from 89 to 70, which means 19 bugs closed in ~2 days15:32
carlossnot bad at all D:15:32
carloss:D *15:32
carlossthank you everyone for reviewing the changes and uploading patches where needed15:33
vhari++ mid-point check progress 15:33
Anoop_ShuklaGreat!15:33
vharity for everyone who is pitching to make a dent in quality 15:33
gouthamr++15:34
carlosswe've quite a few bugs to close still - it's a long list15:34
carlossbut I like the focus and the dedication of the reviewers15:34
carlossso let's keep it up15:34
carlosseven if the changes don't merge, if they get traction is still good progress15:35
carlossand that happened for a lot of changes15:35
carlossis there a change you'd like to add to the bugsquash list or bugs you'd like to have our eyes on?15:35
vhariack, and some merge conflicts have been resolved, helping with reviews15:36
carlosskpdev just linked https://review.opendev.org/c/openstack/manila-tempest-plugin/+/973489 and https://review.opendev.org/q/topic:%22bug/2081827%22 15:37
kpdevyes15:38
carlossthe former is not a bug but important to be closed. I acked gouthamr's comment in the channel about having some comments as draft. I'll try reviewing the tests too15:38
carlossand I'll get to the latter changes15:38
gouthamrwow, bug 2081827 is serious15:40
carlossindeed15:40
gouthamrim not at my computer today, but thanks for patching that up15:40
gouthamrwill take a look when I get in front of a bigger screen15:40
carlossthanks gouthamr15:41
carlossany other bugs you'd like to bring up as part of this meeting15:42
carloss?15:42
carlossdiscussions are going pretty well in the changes15:43
carlossalright15:44
carloss#topic Open Discussion15:44
Anoop_ShuklaWill take a look as well.15:44
gouthamritaiweisman carloss o/ on the access control  discussion, we have an expectation that access control is mandatory, and driven by tenants through Manila. This expectation becomes more of a concern in multi-tenant clouds and, in single tenant clouds, we have a secure-by-design stance by enforcing that each share is inaccessible when it is created - that’s the “deny by design” at play. So, if we’re allowing a different 15:45
gouthamrdesign, I think we owe it to tenants that it not be the default and there are big flags to indicate that we have no way in ensuring that their shares are secured against unauthorized access.15:45
gouthamrjust this cycle we’ve heard that wekafs, DDN’s LUSTRE driver, a new/planned HPE driver can’t map to the established model15:46
gouthamrim thinking we make this super obvious through our share types and tenant visible extra specs mechanism15:47
Anoop_ShuklaUpdate : on the security issue https://bugs.launchpad.net/ossa/+bug/2157914 we are having internal discussion on how to tacle that. We will be defaulting to https transport type by default and a patch should follow soon.15:47
gouthamrgood stuff15:47
Anoop_ShuklaFor NetApp Manila driver15:47
carloss> Update : on the security issue https://bugs.launchpad.net/ossa/+bug/2157914 we are having internal discussion on how to tacle that. We will be defaulting to https transport type by default and a patch should follow soon.15:50
carlossgood stuff Anoop_Shukla, thanks for sharing the progress15:50
carlossgouthamr: yes on the access control stuff... hoping we can get all interested parties in the room next week?15:51
carlossor maybe in the mid cycle15:52
Anoop_ShuklaDeny by design is correct stance IMO15:52
Anoop_ShuklaMost storages follow that practice15:52
carlossagreed15:53
Anoop_ShuklaShould we docu15:53
gouthamrwhen’s our midcycle?15:53
Anoop_ShuklaShould we document this in the maintainers guide?15:53
carlossAug 10 - Aug 1415:53
carloss2 weeks before feature freeze15:53
carlossbut a month from now is little time to react15:54
carlossif the decisions taken from the discussions mean some refactoring happening15:54
carlossmaybe we can have the mid cycle 1 week earlier?15:55
gouthamrI guess so15:55
carlossthat could help... would be the week that follows the fpf and new driver deadline15:55
gouthamrit would allow maintainers to make changes if we decide one way or another15:55
carlossor we could have it at the feature proposal freeze week15:55
gouthamrI can write up our current stance if it’s missing from docs15:55
carlossthat means more time to react15:55
carloss> I can write up our current stance if it’s missing from docs15:56
carloss++, thank you!15:56
Anoop_ShuklaWe shiuld also have operators opinion on this15:56
carlossAnoop_Shukla++15:56
carlossso everyone okay with having the mid cycle at the feature proposal freeze week?15:58
Anoop_Shukla++15:58
carlossawesome... we can make it "later" in the week15:58
carlossso that contributors have the first two or three days of the week to propose the features they didn't propose yet15:58
carlossand then do the mid cycle on thursday/friday or wednesday/thursday15:59
carlosson the good side, there's more time to react to the discussions15:59
carlossalright, let's wrap up the meeting16:00
carlossthanks for the input16:00
carlosslet's get back to our days16:00
carlosshope you have a great day16:00
carlosssee you in #openstack-manila16:00
carloss#endmeeting16:00
opendevmeetMeeting ended Thu Jul 16 16:00:42 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)16:00
opendevmeetMinutes:        https://meetings.opendev.org/meetings/manila/2026/manila.2026-07-16-15.00.html16:00
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/manila/2026/manila.2026-07-16-15.00.txt16:00
opendevmeetLog:            https://meetings.opendev.org/meetings/manila/2026/manila.2026-07-16-15.00.log.html16:00

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