Wednesday, 2026-07-29

opendevreviewYubin Kim proposed openstack/cinder master: NFS driver: extend volume created larger than its snapshot  https://review.opendev.org/c/openstack/cinder/+/98892304:30
opendevreviewSailaja Mada proposed openstack/cinder master: Support for Fibre Channel (FC) as the data plane transport for SnapMirror replication  https://review.opendev.org/c/openstack/cinder/+/99659906:29
*** zseguin is now known as Guest1435007:21
opendevreviewSailaja Mada proposed openstack/cinder master: Support for Fibre Channel (FC) as the data plane transport for SnapMirror replication  https://review.opendev.org/c/openstack/cinder/+/99659907:35
opendevreviewRaghavendra Tilay proposed openstack/cinder master: HPE: Update support matrix  https://review.opendev.org/c/openstack/cinder/+/99908511:16
zigoeharney: May I ask you for reviewing https://review.opendev.org/c/openstack/cinder/+/979641 ?12:01
opendevreviewSilvan Kaiser proposed openstack/cinder stable/2026.1: Fix copy_image_to_volume for Quobyte driver  https://review.opendev.org/c/openstack/cinder/+/99910112:36
opendevreviewAnthony Galica proposed openstack/cinder master: [Hitachi] Add support for NVMe-TCP.  https://review.opendev.org/c/openstack/cinder/+/99746912:36
opendevreviewSilvan Kaiser proposed openstack/cinder stable/2026.1: Report provisioning in driver stats for Quobyte driver  https://review.opendev.org/c/openstack/cinder/+/99910212:37
opendevreviewRaghavendra Tilay proposed openstack/cinder master: [DNM] HPE: Update support matrix  https://review.opendev.org/c/openstack/cinder/+/99908512:56
opendevreviewJon Bernard proposed openstack/cinder master: Skip TestWSGIService under the threading backend  https://review.opendev.org/c/openstack/cinder/+/99590813:19
opendevreviewJon Bernard proposed openstack/cinder master: Skip eventlet-only semaphore tests under threading backend  https://review.opendev.org/c/openstack/cinder/+/99590913:23
auniyalrm_work[m] not all commetns are addressed in patch13:28
jbernard#startmeeting cinder14:00
opendevmeetMeeting started Wed Jul 29 14:00:56 2026 UTC and is due to finish in 60 minutes.  The chair is jbernard. Information about MeetBot at http://wiki.debian.org/MeetBot.14:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.14:00
opendevmeetThe meeting name has been set to 'cinder'14:00
jbernard#link https://etherpad.opendev.org/p/cinder-hibiscus-meetings14:01
jbernard#topic roll call14:02
jbernardo/14:02
agalicao/14:02
hvlcchao1o/14:02
harsho/14:02
jayaanando/14:04
jbernardhey everyone, this might be a quick meeting today14:05
Jyotsnao/14:05
jbernardperiodic jobs are all looking good14:05
jbernardi wanted to raise the topic of a second midcycle meeting14:06
jbernardwe aren't required to hold one, but we typically do and I think it would be useful14:06
jbernardi wanted to ask about timing, what works for everyone? we could sort it out here, i could send out a time poll, any preferences?14:07
agalicanot at 7am :(  lol14:08
jbernardlol, ok so one strong vote for /very/ early :)14:08
agalicayou joke, but I'd rather it be 4am, haha14:09
jbernardoof, that would be 7am my time, that's too early :P14:10
agalica:P14:10
jbernardok, i will take a look at the remaining calendar and propose a date and time before next week's meeting14:11
jbernardwill aim for something similar to what we've done previously14:11
jbernard#topic open discussion14:12
agalicaIt would be highly appreciated if anyone can respond to my query on the mailing list: "[Cinder/Nova, NVME-TCP] Issue with multipath and devstack"14:13
JyotsnaHPE Alletra MP new driver patch https://review.opendev.org/c/openstack/cinder/+/969373 (Recieved +2 and workflow +1 and ready for merge)14:13
harshhi.. is erlon around ?14:13
agalicaasuming you know the answer, anyway14:13
agalicaassuming*14:13
JyotsnaHPE Alletra MP new driver patch https://review.opendev.org/c/openstack/cinder/+/969373 (Recieved +2 and workflow +1 and ready for merge) But have dependency on requirement patch https://review.opendev.org/c/openstack/requirements/+/974322  Received +1 and awaiting +2 from core reviewers, assuming once we have +2 from 2 reviewers , new driver patch will also merge14:13
opendevreviewThomas Goirand proposed openstack/cinder master: Improve heartbeats: RPC pings  https://review.opendev.org/c/openstack/cinder/+/97171014:14
jbernardagalica: i just saw that yesterday, it went into spam for me for some reason... I had a working test config at some point14:14
harshthere are a few reviews in https://etherpad.opendev.org/p/cinder-hibiscus-reviews that are on hold for CI to pass. These reviews have received one +2 and the CI is also green. The reviews are the last ones with (harsh) in Bugs and Features section.14:15
harshwould really appreciate if i can get a second +2 :)14:15
agalicajbernard: I'm guessing the issue is that multipath=False is continually being sent, but I'm out of ideas on how to fix that14:15
Jyotsna<jbernard> can you please convey this to core reviewers sean.mcginnis@gmail.com and frickler@offenerstapel.de for +2 approvals.14:15
jbernardagalica: i remember testing gorka's patches, we had a local.conf that enabled multipath and I remember it working in our test env14:16
jbernardagalica: for nvme-tcp14:17
agalicajbernard: if you can find that it would be highly appreciated :)14:17
jbernardagalica: no sure about root cause, ill try to find my config and see if it still works14:18
jbernardagalica: it's been quite a while14:18
agalicajbernard: thanks14:18
mhenhi14:18
mhenwas anybody aside from eharney able to look at https://bugs.launchpad.net/cinder/+bug/2161646 ?14:19
jbernardharsh: ack, i've been sick/pto for the last few days but will try to take a look this afternoon14:19
harshjbernard: thanks :)14:19
mhenI haven't gotten around to trying the patchset he linked but I'd like to try in the coming days to see if that might fix it14:19
jbernardmhen: i have been sick so i haven't checked on that one, i remember seeing his patch though, worth a try for sure14:20
mhenack14:21
jbernardmhen: if it doesn't address the issue, im guessing eric would want to know about the remaining gaps14:22
jbernardso a useful experiment either way14:22
erlon\o14:23
erlon@harh, was distracted :)14:23
zigoo/14:23
Jyotsna<jbernard> we need attention of reviewers on HPE new driver patch, updated the meeting https://etherpad.opendev.org/p/cinder-hibiscus-reviews 14:24
Jyotsnarequirement patch https://review.opendev.org/c/openstack/requirements/+/974322  Received +1 and awaiting +2 from core reviewers, assuming once we have +2 from 2 reviewers , new driver patch will also merge14:24
harsherlon: no problem :) just wanted to let you know that CI is passed and there are +2s on the patches.14:24
jbernardJyotsna: yes, i'm very aware, i know folks have it on their todo list14:25
Jyotsnaok thanks14:25
erlonharsh: watch patches are those?14:25
harshhttps://etherpad.opendev.org/p/cinder-hibiscus-reviews - Bugs section for (harsh) 14:26
harshAlso two important patches in Features section. Our upcoming features such as Policy Based High Availability and Policy Based Replication are highly dependent on these two patches. We have got multiple +1s and CI is also green. Customers have tested and passed as well.14:27
zigoFYI, I'll package python3-hpe-storage-flowkit-py right away for Debian.14:28
erlongot them14:28
jbernardzigo: nice14:31
zigoWeird, the upstream repo doesn't even have a pyproject.toml or a setup.py.14:32
zigoHow is one supposed to package that ?!?14:32
Jyotsnajbernard: just one last query is it possible to get the driver patch merged before requirement patch? or it is mandatory 14:32
Jyotsnazigo: Driver patch https://review.opendev.org/c/openstack/cinder/+/969373 has pyproject.toml 14:33
jbernardJyotsna: i think the driver depends on the requirement patch so that the gate will pass14:34
Jyotsnaok14:34
jbernardJyotsna: we will need to maintain that dependency14:34
Jyotsnasure thanks 14:34
fqiHi, I really appreciate it if someone can review these patches for the "retype volume including snapshots" feature.14:35
fqiCinder-spec:https://review.opendev.org/c/openstack/cinder-specs/+/98932214:35
zigoJyotsna: I know Cinder has a pyproject.toml, but the HPE python module does not. See:14:35
zigohttps://github.com/hpe-storage/hpe_storage_flowkit_py14:35
zigoSeeing this python module state, I'd vote AGAINST including it.14:36
zigoThere's some "magic" that isn't included in the github repo, that's really bad practice.14:37
jbernardzigo: can you post this on the review so others can see it, i know not everyone can make the meeting14:38
zigoWill do.14:38
jbernardthat way we can iterate as efficiently as possible14:38
jbernardzigo: thanks!14:38
Lucas_Hi all, I'd like to kindly ask for more reviews on my patch: Fujitsu Driver: Improve Snap Pool selection https://review.opendev.org/c/openstack/cinder/+/91049414:39
Lucas_It has passed CI (+1) and already has assigned reviewers (Raghave... and saikrishna). Your feedback and +2 would be very helpful to get it merged. Thanks!14:39
jbernardLucas_: ack, added it to tehh review etherpad14:40
Lucas_thanks a lot! That's very helpful.14:41
Anoop_Shukla@jbernard: I have a question wrt pool management in Cinder. Is there a reason we do not have APIs to manage pools in Cinder? It happens via looped call to driver via ssc update. But if a user wants to manually manage the pools without cinder volume service restart, there is no way.14:41
harshjayaanand: Could you please look into this, thanks :)  [SVf] lsportip needs to fetch IPs with "host" flag | https://review.opendev.org/c/openstack/cinder/+/86930514:42
Anoop_ShuklaMost of the other cloud solutions - including cloudstack have these APIs to manage pool.14:42
jbernardAnoop_Shukla: ohh, i don't think i was around when those choices were made, what operations are wanted?14:43
Anoop_ShuklaI would recommend to have a CRUD API support which can manage all operations - adding a backend pool, deleting a backend pool, extending a backend pool etc..14:43
zigoSince everyone is doing it, my turn... :)14:43
zigoI also would appreciate reviews on this patch: https://review.opendev.org/c/openstack/cinder/+/97964114:43
zigoIt's been stuck for a few weeks, even though it's almost a one-liner, and it's addressing a real bug.14:43
jbernardzigo: ack14:44
Anoop_ShuklaDriver can manage the respective configs for pool management..cinder orchestrates it14:44
Anoop_ShuklaToday pool management relies on name pattern matching or fixed pool design to workaround it14:44
jbernardAnoop_Shukla: im guessing we didn't forsee wanting to manage those dynamically at runtime14:44
Anoop_ShuklaBut the ssc update only happens hourly and a cloud backend can suddenly go out of space if bulk provisioning is done14:45
Anoop_ShuklaIs there a room for discussion on these APIs..as a proposal (may be during mid cycle) to see if we can gather the feasibility for such APIs?14:45
agalicaI personally think that would be a pretty radical change since at this time the configuration file is read only14:46
jbernardAnoop_Shukla: midcycle would be the place for that topic14:46
Anoop_ShuklaSure..let me find a slot to discuss this. @agalica - I did not understand what you meant by radical change?14:47
jbernardAnoop_Shukla: preping folks for the dicsussion would be good, so we're not looking at git history and api calls mid-discussion14:47
Anoop_ShuklaI have seen complaints that Cinder can only handle it on an hourly basis..and they are pretty dynamically looking to automate the pool management14:48
agalicaAnnop_Shukla: I just mean that openstack configuration files are all read-only in terms of openstack opepration.  So to add new pools would require that they become read/write, which to me is a big deal.  14:48
agalicathat's all14:48
Anoop_ShuklaAlso, if I want to do it immediately - ssc updates are non-configurable or other option is restart driver14:48
Anoop_ShuklaI am not proposing adding a new backend..14:49
agalicaI know14:49
Anoop_ShuklaIts about managing the pools within a backend14:49
agalicaI don't know what SSC is though, so maybe I can just be ignored14:49
Anoop_ShuklaWhere customers rely on patten matching and ssc update updates backed pool config every hour14:49
Anoop_ShuklaHmm..trying to find some solutions to limitations on netapp driver..where we configure FlexVolumes (containers) as pools for a backend.14:51
Anoop_ShuklaNetApp driver dynamically manages this limitation by allowing users to configure a naming pattern that periodically polls the backend to find if new FlexVolumes are added on storage system..if there are - we add the pool and perform capability discovery hourly..14:53
jbernardi remember something related to ssc pool management coming up before, i think there was a spec, or at least meeting logs, will try to find14:55
Anoop_ShuklaThanks @jbernard. I will prep an ether pad to discuss it in mid cycles14:56
jbernardAnoop_Shukla: I will add content for second midcycle here: https://etherpad.opendev.org/p/cinder-hibiscus-midcycles14:57
jbernardAnoop_Shukla: if you can get your info there, ill move it around14:57
jbernard1 minute, last call for the things and the stuff14:59
fqiAlso, We found two issues related to volume retype with migration between an LVM backend and a generic NFS backend https://bugs.launchpad.net/cinder/+bug/216020614:59
fqiThe patches to fix the issue are awaiting review. Any help would be greatly appreciated.14:59
fqihttps://review.opendev.org/c/openstack/cinder/+/99672814:59
fqihttps://review.opendev.org/c/openstack/cinder/+/99672714:59
jbernardfqi: im sure eric will want to be a part of those reviews, as he's interested in nfs, but i know he's busy, ill mention it when i get a chance15:00
fqiGot it. Thank you very much.15:01
Anoop_ShuklaAdded the topic at the top. We can work with the slot.15:01
jbernardAnoop_Shukla: cool15:02
jbernardok, we're at time15:02
jbernardthank you everyone15:02
jbernard#endmeeting15:02
opendevmeetMeeting ended Wed Jul 29 15:02:52 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)15:02
opendevmeetMinutes:        https://meetings.opendev.org/meetings/cinder/2026/cinder.2026-07-29-14.00.html15:02
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/cinder/2026/cinder.2026-07-29-14.00.txt15:02
opendevmeetLog:            https://meetings.opendev.org/meetings/cinder/2026/cinder.2026-07-29-14.00.log.html15:02
agalicaGLAD IT WAS A SHORT ONE!15:04
harsh:D15:05
jbernardagalica: lol15:08
opendevreviewMerged openstack/cinder master: Quobyte: adapt to image_utils fixes  https://review.opendev.org/c/openstack/cinder/+/95396615:33
sailaja=15:35
opendevreviewMerged openstack/cinder master: Quobyte: Fix metadata for volume from snapshot #2142477  https://review.opendev.org/c/openstack/cinder/+/97856415:47
opendevreviewMerged openstack/cinder master: Remove unnecessary direct dependency on Pygments  https://review.opendev.org/c/openstack/cinder/+/98257015:47
opendevreviewanthony gamboa proposed openstack/cinder master: [Hitachi] Remove eventlet  https://review.opendev.org/c/openstack/cinder/+/99918319:51
opendevreviewAnthony Galica proposed openstack/cinder master: [Hitachi] Add support for NVMe-TCP.  https://review.opendev.org/c/openstack/cinder/+/99746923:29

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