| opendevreview | Atsushi Kawai proposed openstack/cinder master: Hitachi: Make the parameters name variable for supporting OEM storages https://review.opendev.org/c/openstack/cinder/+/815461 | 00:53 |
|---|---|---|
| opendevreview | Atsushi Kawai proposed openstack/cinder master: Hitachi: Add maintenance parameters https://review.opendev.org/c/openstack/cinder/+/786873 | 00:54 |
| opendevreview | Atsushi Kawai proposed openstack/cinder master: Hitachi: Make the parameters name variable for supporting OEM storages https://review.opendev.org/c/openstack/cinder/+/815461 | 04:44 |
| opendevreview | Atsushi Kawai proposed openstack/cinder master: HPE: Cinder driver for HPE XP storage FC and iSCSI arrays https://review.opendev.org/c/openstack/cinder/+/815582 | 05:00 |
| opendevreview | Atsushi Kawai proposed openstack/cinder master: HPE: Cinder driver for HPE XP storage FC and iSCSI arrays https://review.opendev.org/c/openstack/cinder/+/815582 | 05:06 |
| crohmann | Good morning. My collegue Simon and me did send in a changeset (https://review.opendev.org/c/openstack/openstacksdk/+/814232) to add the updated_at field to the volume ressource. Now the unit tests are unhappy. Could anybody kindly point us to the right place to extend the mocked ressource or give a hint on what needs changing here? | 08:27 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Remove leftover nested quota DB fields https://review.opendev.org/c/openstack/cinder/+/815145 | 09:19 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Clean old temporary tracking https://review.opendev.org/c/openstack/cinder/+/815146 | 09:19 |
| opendevreview | Zohar Mamedov proposed openstack/cinder-specs master: Os-brick NVMeoF connector multipathing spec https://review.opendev.org/c/openstack/cinder-specs/+/815613 | 10:09 |
| opendevreview | Naoki Saito proposed openstack/cinder master: Add cinder volume drivers for NEC V series Storage https://review.opendev.org/c/openstack/cinder/+/815614 | 10:09 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Clean old temporary tracking https://review.opendev.org/c/openstack/cinder/+/815146 | 11:00 |
| *** dviroel|rover|afk is now known as dviroel|rover | 11:15 | |
| rosmaita | crohmann: thanks for working on making openstackSDK more compatible with the volume API ... left a comment for you on your patch | 11:41 |
| hemna | good morning | 12:15 |
| crohmann | rosmaita: Thanks a bunch! | 12:27 |
| crohmann | rosmaita: I am still a little puzzled. The get_mock_url call does use volumev2 but this is then compared to a volumev3 (the fake volume)? I would need to add the updated_at to volumev2 as well to make this work. But I did not want to change the "old" volumev2 actually. | 12:37 |
| jungleboyj | hemna: Morning stranger. :-) | 12:41 |
| jungleboyj | rosmaita: I have a customer meeting that conflicts with the Cinder meeting so I probably won't be able to make the Cinder meeting today. | 12:41 |
| rosmaita | crohmann: you'll need to ask someone on the SDK team about how to handle that, i think you can ask in #openstack-sdks channel | 12:51 |
| crohmann | rosmaita: We implemented the changes now and pushed a new patchset. could I name someone to assign as reviewers or should I better hop onto #openstack-sdks and ask? | 12:54 |
| rosmaita | crohmann: probably better to ask (gives you 2 ways to get attention to your patch) | 12:54 |
| rosmaita | jungleboyj: ack, enjoy your meeting | 12:54 |
| jungleboyj | rosmaita: Thanks. I am hoping to. | 12:56 |
| crohmann | rosmaita: While we are chatting ... another thing we are looking into implementing is this blueprint: https://blueprints.launchpad.net/cinder/+spec/volume-status-for-server-migration. We have a few basic questions on how to handle the volume state (overwriting the backing-up with in-use on successful migration, ...). Should we come to a cinder meeting and place this as a topic in the Etherpad? | 12:57 |
| crohmann | Or just ask here and start implemeting? | 12:57 |
| rosmaita | probably better to come to the meeting, it's 1400 UTC today | 12:58 |
| crohmann | k. No need to place this on the agenta prior or just showing up? | 12:58 |
| crohmann | "agenda" | 12:58 |
| rosmaita | crohmann: add it to the agenda: https://etherpad.opendev.org/p/cinder-yoga-meetings | 13:03 |
| rosmaita | meeting is in irc + video this week | 13:04 |
| rosmaita | just realized the connection info isn't on the agenda yet | 13:04 |
| rosmaita | ok, updated | 13:07 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Remove attach and detach volume driver methods https://review.opendev.org/c/openstack/cinder/+/814898 | 13:14 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Join do_setup and check_for_setup_error https://review.opendev.org/c/openstack/cinder/+/815659 | 13:14 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Docs: Document clone_image driver method https://review.opendev.org/c/openstack/cinder/+/815660 | 13:14 |
| opendevreview | Sofia Enriquez proposed openstack/cinder stable/victoria: DNM upgrade ddt https://review.opendev.org/c/openstack/cinder/+/815663 | 13:19 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Docs: Document clone_image driver method https://review.opendev.org/c/openstack/cinder/+/815660 | 13:20 |
| rosmaita | reminder: today's cinder meeting is on video + #openstack-meeting-alt at 1400 UTC | 13:58 |
| rosmaita | connection info: https://bluejeans.com/3228528973 | 13:58 |
| rosmaita | jungleboyj rosmaita smcginnis tosky whoami-rajat m5z e0ne geguileo eharney walshh_ jbernard sfernand enriquetaso hemna fabiooliveira ^^ | 13:58 |
| opendevreview | Eric Harney proposed openstack/cinder stable/victoria: Tests: Fix compatibility with ddt 1.2.x https://review.opendev.org/c/openstack/cinder/+/815673 | 14:17 |
| opendevreview | Eric Harney proposed openstack/cinder master: Remove various unused code https://review.opendev.org/c/openstack/cinder/+/800703 | 15:01 |
| enriquetaso | #startmeeting cinder_bs | 15:02 |
| enriquetaso | #startmeeting cinder_bs | 15:02 |
| opendevmeet | Meeting started Wed Oct 27 15:02:13 2021 UTC and is due to finish in 60 minutes. The chair is enriquetaso. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:02 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:02 |
| opendevmeet | The meeting name has been set to 'cinder_bs' | 15:02 |
| enriquetaso | #link http://lists.openstack.org/pipermail/openstack-discuss/2021-October/025561.html | 15:02 |
| enriquetaso | #topic bug_1 Cinder started requiring write access to glance images RBD pool | 15:02 |
| enriquetaso | #link https://bugs.launchpad.net/cinder/+bug/1947518 | 15:02 |
| enriquetaso | Summary: Cinder started requiring write access to glance images RBD pool (on creating volume from image). This problem was introduced by: "RBD: use correct stripe unit in clone operation" | 15:03 |
| enriquetaso | #link https://review.opendev.org/c/openstack/cinder/+/786260 | 15:03 |
| enriquetaso | Unfortunately, this patch has been backported back to Train and broke stable deployments. | 15:03 |
| enriquetaso | The reporter added a test for this to catch early issues like this on the kolla ansible project: | 15:04 |
| enriquetaso | #link https://review.opendev.org/c/openstack/kolla-ansible/+/814324 | 15:04 |
| eharney | https://review.opendev.org/c/openstack/cinder/+/783643 | 15:04 |
| enriquetaso | #topic bug_2 Inconsistency for resource column between quota_usages and reservations table | 15:05 |
| enriquetaso | #link https://bugs.launchpad.net/cinder/+bug/1948916 | 15:06 |
| enriquetaso | Summary: This problem was not visible before Stein release because cinder db schema changed a bit and resource size inside quota_usages increased from 255 to 300 when the same column remained unchanged in reservations table. That was due to | 15:07 |
| enriquetaso | https://review.opendev.org/c/openstack/cinder/+/611530/ | 15:07 |
| enriquetaso | #topic open discussion | 15:09 |
| enriquetaso | #link https://review.opendev.org/c/openstack/cinder/+/815673 | 15:10 |
| enriquetaso | we don't do lower contains any more | 15:11 |
| opendevreview | Eric Harney proposed openstack/cinder master: RBD: Open RBD images read-only where possible https://review.opendev.org/c/openstack/cinder/+/783643 | 15:12 |
| enriquetaso | #endmeeting | 15:24 |
| opendevmeet | Meeting ended Wed Oct 27 15:24:37 2021 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:24 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/cinder_bs/2021/cinder_bs.2021-10-27-15.02.html | 15:24 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/cinder_bs/2021/cinder_bs.2021-10-27-15.02.txt | 15:24 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/cinder_bs/2021/cinder_bs.2021-10-27-15.02.log.html | 15:24 |
| rosmaita | crohmann: meeting recording -- https://youtu.be/R-Q-sSKH4k8 | 15:56 |
| opendevreview | Javier Cacheiro proposed openstack/cinder master: Fix: _get_stripe_unit should open glance image read-only https://review.opendev.org/c/openstack/cinder/+/815701 | 16:02 |
| geguileo | enriquetaso: I think the quota_usages bug we discussed is not really a cinder bug https://bugs.launchpad.net/cinder/+bug/1948916 | 16:24 |
| geguileo | enriquetaso: I mean, we have an inconsistency, but the bug they mention is unrelated to the inconsistency and related to the field size increase | 16:24 |
| enriquetaso | :o | 16:24 |
| geguileo | I added some comments to the bug | 16:25 |
| geguileo | I'll create a new bug for the inconsistencies (there are more) and fix them | 16:25 |
| enriquetaso | so the bug is invalid then... | 16:26 |
| enriquetaso | OK, thanks geguileo, i'll close 1948916 | 16:27 |
| geguileo | enriquetaso: maybe we should wait to hear back from the reporter... | 16:44 |
| enriquetaso | sorry geguileo, OK, let's wait to the reporter :) | 17:00 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: DB: Set quota resource to 300 https://review.opendev.org/c/openstack/cinder/+/815712 | 17:01 |
| geguileo | enriquetaso: ^ that's the fix for the inconsistencies | 17:01 |
| enriquetaso | thanks! | 17:37 |
| qwebirc17101 | rosmaita kindly explained to me yesterday that the encrypted volumes' LUKS key that Cinder generates and gives to Barbican, is stored as a binary file. For example, the plaintext LUKS passphrase "FAFAFA..." is stored as a binary file with contents "0xFA 0xFA 0xFA ...". | 17:50 |
| qwebirc17101 | It's not complex to work with, but why is it this way? Barbican can easily store plaintext secrets, and the passphrase is plaintext, and I can't figure out a way to use a binary file with luks anyway | 17:51 |
| rosmaita | qwebirc17101: it's because we let barbican do a random fill of the number of bytes requested | 17:53 |
| rosmaita | that way the randomness is pushed off to barbican, which should have a suitable source | 17:53 |
| qwebirc17101 | And Barbican can't simply generate a random string of length x? | 17:54 |
| rosmaita | qwebirc17101: not sure, tbh, we've done it this way since before i worked on the project | 17:55 |
| qwebirc17101 | the default bit length for the key is 256, so bytes or hex/ascii text shouldn't matter (I guess) | 17:55 |
| qwebirc17101 | ok, honestly just curious. thanks! | 17:55 |
| rosmaita | np | 17:55 |
| qwebirc17101 | Another thing to add/ask about, the key is generated in Barbican with no name. I think it would make sense to add some identifier to the secret, to show that it's a luks key used by cinder. Where could I bring up that recommendation/feature request? | 17:58 |
| qwebirc17101 | At the moment, I'm doing something like this to filter to all luks keys. In our situation it works fine, but it's not bullet proof: | 17:59 |
| qwebirc17101 | openstack secret list -f json --limit 100 -c "Secret href" -c "Name" -c "Created" --sort-column "Created" --algorithm "aes" --bit-length 256 | \ jq '.[] | select(.Name==null)' | 17:59 |
| opendevreview | Merged openstack/cinder stable/rocky: Correct group:reset_group_snapshot_status policy https://review.opendev.org/c/openstack/cinder/+/809657 | 18:30 |
| opendevreview | Merged openstack/cinder stable/train: Fix issue of getting detail backups list info https://review.opendev.org/c/openstack/cinder/+/810200 | 18:30 |
| opendevreview | Merged openstack/os-brick stable/wallaby: multipath/iscsi: iSCSI connections are not reinitiated after reboot https://review.opendev.org/c/openstack/os-brick/+/814754 | 18:55 |
| opendevreview | Simon Dodsley proposed openstack/os-brick stable/victoria: multipath/iscsi: iSCSI connections are not reinitiated after reboot https://review.opendev.org/c/openstack/os-brick/+/815641 | 18:56 |
| Zer0Byte | hey | 19:07 |
| Zer0Byte | question | 19:07 |
| Zer0Byte | im usign cinder with a qos frontend option to limit the iops of a volume | 19:07 |
| Zer0Byte | but when i perform cinder extend the volume grow up but i dont see that the iops per gb update the total iops | 19:08 |
| Zer0Byte | im using this spec | 19:08 |
| Zer0Byte | total_iops_sec_per_gb=3 | 19:08 |
| geguileo | Zer0Byte: I belive that would be a Nova bug | 19:44 |
| Zer0Byte | exist a bug report | 19:45 |
| geguileo | Zer0Byte: I understand this is a volume that is attached to an instance, so when cinder notifies nova that it has extended the volume they should update it | 19:45 |
| Zer0Byte | yes is a attached volume | 19:46 |
| geguileo | mmmm, may not be so easy... | 19:51 |
| geguileo | specs seem to be provided on the connection, and I don't think there is a mechanism for Cinder to update Nova with values | 19:51 |
| geguileo | so new code may be necessary in Nova & Cinder | 19:52 |
| * geguileo keeps digging | 19:52 | |
| geguileo | Zer0Byte: confirmed, it's a Nova-Cinder bug that would require changes in both projects | 19:55 |
| geguileo | right now it's not dynamic and only takes effect on attach | 19:56 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: DB: Set quota resource to 300 https://review.opendev.org/c/openstack/cinder/+/815712 | 20:40 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Remove leftover nested quota DB fields https://review.opendev.org/c/openstack/cinder/+/815145 | 20:40 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Clean old temporary tracking https://review.opendev.org/c/openstack/cinder/+/815146 | 20:40 |
| opendevreview | Gorka Eguileor proposed openstack/cinder master: Prevent table and column alter and drop https://review.opendev.org/c/openstack/cinder/+/815746 | 20:40 |
| geguileo | rosmaita: eharney We'll need to merge that latest patch before any new alembic migration, because we don't have the rolling upgrade checks we had for the sqlalchemy-migrate migrations | 20:42 |
| rosmaita | geguileo: ack | 20:44 |
| simondodsley | rosmaita, geguileo, could you review the Victoria iteration of the os_brick iscsi patch? https://review.opendev.org/c/openstack/os-brick/+/815641/ | 20:54 |
| geguileo | simondodsley: done | 20:56 |
| simondodsley | thank you rosmaita and geguileo | 21:14 |
| opendevreview | Merged openstack/cinder master: mypy: coordination.py https://review.opendev.org/c/openstack/cinder/+/788364 | 21:16 |
| rosmaita | qwebirc17101: cinder uses castellan (a generic keymanager interface) to interact with barbican, and i don't think it gives us the option to set a name on a secret even though barbican would allow it. | 21:30 |
| rosmaita | qwebirc17101: there used to be opposition to requiring barbican to be deployed in openstack clouds, but i'm not sure that's a concern any more | 21:31 |
| rosmaita | qwebirc17101: plus, we are going to want to use the "Secret Consumers API" once it's implemented in Barbican: https://specs.openstack.org/openstack/barbican-specs/specs/train/secret-consumers.html | 21:32 |
| rosmaita | so maybe we need to transition to Barbican instead of generic key manager | 21:33 |
| rosmaita | simondodsley: thanks for keeping an eye on those brick backports! | 21:34 |
| *** dviroel|rover is now known as dviroel|rover|afk | 22:08 | |
| opendevreview | Brian Rosmaita proposed openstack/cinder stable/queens: Correct group:reset_group_snapshot_status policy https://review.opendev.org/c/openstack/cinder/+/815646 | 22:46 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!