| opendevreview | Merged openstack/manila stable/zed: [RBAC] Return 404 if share is inaccessible https://review.opendev.org/c/openstack/manila/+/859235 | 01:38 |
|---|---|---|
| opendevreview | GABRIEL MOGAJI proposed openstack/manila master: Fixed a documentation bug https://review.opendev.org/c/openstack/manila/+/861965 | 10:41 |
| opendevreview | OpenStack Release Bot proposed openstack/puppet-manila stable/zed: Update .gitreview for stable/zed https://review.opendev.org/c/openstack/puppet-manila/+/862017 | 11:06 |
| opendevreview | OpenStack Release Bot proposed openstack/puppet-manila stable/zed: Update TOX_CONSTRAINTS_FILE for stable/zed https://review.opendev.org/c/openstack/puppet-manila/+/862018 | 11:06 |
| opendevreview | OpenStack Release Bot proposed openstack/puppet-manila master: Update master for stable/zed https://review.opendev.org/c/openstack/puppet-manila/+/862019 | 11:06 |
| *** dviroel|out is now known as dviroel | 11:25 | |
| opendevreview | Alexander Deiter proposed openstack/manila master: [Infinidat] fixed host assisted migration https://review.opendev.org/c/openstack/manila/+/860788 | 14:05 |
| *** dviroel is now known as dviroel|ptg | 14:08 | |
| opendevreview | Merged openstack/puppet-manila master: Update master for stable/zed https://review.opendev.org/c/openstack/puppet-manila/+/862019 | 14:24 |
| opendevreview | Merged openstack/puppet-manila stable/zed: Update .gitreview for stable/zed https://review.opendev.org/c/openstack/puppet-manila/+/862017 | 14:29 |
| opendevreview | Merged openstack/puppet-manila stable/zed: Update TOX_CONSTRAINTS_FILE for stable/zed https://review.opendev.org/c/openstack/puppet-manila/+/862018 | 14:29 |
| *** dviroel|ptg is now known as dviroel|lunch\ | 16:06 | |
| opendevreview | GABRIEL MOGAJI proposed openstack/manila master: Fixed a documentation bug https://review.opendev.org/c/openstack/manila/+/861965 | 16:18 |
| *** dviroel|lunch\ is now known as dviroel | 16:53 | |
| gouthamr | vkmc fpantano rraja: https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/KSHR4QS5X6EMC5V4DTJUJJ4464LX5BPV/ | 16:57 |
| gouthamr | ken dryer says quincy packages for jammy will come soon | 16:57 |
| gouthamr | no ETA unfortunately | 16:57 |
| gouthamr | but, he notes that no plans to release octopus/pacific for jammy | 16:58 |
| rraja | weren't we concerned about only about quincy packages | 16:58 |
| rraja | I see he mentioned "soon" back in Aug. But I think a follow up email on the same thread will be good. | 16:59 |
| rraja | Ken may not be aware of the manila CI needs | 17:00 |
| gouthamr | ack rraja | 17:00 |
| gouthamr | i'll respond and copy you folks | 17:00 |
| gouthamr | you're right, we don't need octopus/pacific packages | 17:00 |
| gouthamr | we *may* be interested in reef/master -- but only for experiments, and we consume master from shaman anyway | 17:01 |
| gouthamr | gmann kopecmartin ^ | 17:01 |
| rraja | gouthamr, the cephfs native driver start up issue should go away when "share_service_inithost_offload" is set to True | 17:01 |
| gouthamr | fyi :) | 17:01 |
| rraja | ? | 17:02 |
| rraja | with regards to this issue "big startup time due to calculation of paths for all shares in particular backend" | 17:03 |
| vkmc | gouthamr++ | 17:03 |
| vkmc | it's good that they were already looking into it, maybe they another reason to release it | 17:04 |
| vkmc | we can be that another reason | 17:04 |
| vkmc | :D | 17:04 |
| vkmc | rraja, how is that? | 17:05 |
| rraja | vkmc, that's what I got from the operator's meeting that the issue with driver startup taking a long time was related to how manila share service issued ensure_share calls to the driver | 17:06 |
| rraja | gouthamr, ^^ | 17:07 |
| gouthamr | rraja: yes; will check with josecastroleon after he's able to set that configuration | 17:08 |
| gouthamr | haixin is probably interested in making that option default to True in antelope | 17:08 |
| rraja | great! | 17:08 |
| rraja | I was wondering why it's not already default | 17:08 |
| gouthamr | ++ | 17:09 |
| rraja | gouthamr, vkmc in which manila release did the cephfs driver switch to using mgr/volumes instead of ceph_volume_client? | 17:18 |
| rraja | carloss, I'm looking at https://drive.google.com/file/d/1iqhPH3a2Fdqm0FsbJpyC1dCHopEpnQsH/view . would you know the number of users who responded to the survey? | 17:22 |
| rraja | e.g., it will let me estimate, among the respondents, the number of users of a particular driver or number of users of particular driver in production? | 17:29 |
| rraja | next time would it possible to split CephFS driver category into CephFS native and CephFS NFS? | 17:30 |
| gouthamr | ^ good point | 17:37 |
| rraja | also, I see folks use CephFS drivers in DHSS=True. so operators are using custom drivers to do this | 17:38 |
| carloss | rraja: it was around 166 answers, but there were some that mentioned drivers and didn't mention if they are production or not | 17:39 |
| gouthamr | rraja: i think those folks are interested in a 'generic' share servers solution for cephfs | 17:39 |
| rraja | or did they misunderstand what DHSS is? at least to me the acronym has always been cryptic | 17:39 |
| gouthamr | also; we're aware from prior surveys that there are deployments with the generic driver where rbd is the cinder backend | 17:40 |
| carloss | yes, I agree it would be interesting to split cephfs from cephfs nfs | 17:40 |
| gouthamr | that would mean using a nova server as an NFS server with rbd cinder volumes as shares | 17:41 |
| rraja | very interesting! that could also be added accidentally into the CephFS driver usage? | 17:42 |
| gouthamr | these folks like the ceph/nfs gateway solution - i'm not sure if all have migrated over though... | 17:42 |
| rraja | maybe not, I see a separate LVM/generic driver | 17:42 |
| gouthamr | rraja: unlikely, unless it was misconstrued.. | 17:42 |
| rraja | got it | 17:43 |
| rraja | carloss, thanks! | 17:44 |
| *** dviroel is now known as dviroel|out | 22:47 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!