| opendevreview | Merged openstack/openstack-ansible stable/xena: Add internal_lb_vip_address to no_proxy https://review.opendev.org/c/openstack/openstack-ansible/+/883010 | 00:30 |
|---|---|---|
| jrosser | good morning | 07:45 |
| jrosser | noonedeadpunk: do we have tracking (etherpad?) of what we need to get done for release? | 07:45 |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible-os_magnum master: Ensure service is restarted on unit file changes https://review.opendev.org/c/openstack/openstack-ansible-os_magnum/+/879970 | 15:04 |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible-os_magnum master: Add TLS support to magnum backends https://review.opendev.org/c/openstack/openstack-ansible-os_magnum/+/879918 | 15:04 |
| damiandabrowski | #startmeeting openstack_ansible_meeting | 15:06 |
| opendevmeet | Meeting started Tue May 16 15:06:50 2023 UTC and is due to finish in 60 minutes. The chair is damiandabrowski. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:06 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:06 |
| opendevmeet | The meeting name has been set to 'openstack_ansible_meeting' | 15:06 |
| jrosser | o/ hello | 15:08 |
| damiandabrowski | noonedeadpunk: had some emergency today and he probably won't join so i spontaneously decided to take care of the meeting | 15:08 |
| damiandabrowski | (however i'm completely not prepared :|) | 15:08 |
| damiandabrowski | #rollcall | 15:08 |
| jrosser | hello | 15:09 |
| damiandabrowski | #topic rollcall | 15:09 |
| NeilHanlon | heyo | 15:09 |
| jrosser | feels like it will be quiet today :) | 15:10 |
| damiandabrowski | #topic office hours | 15:10 |
| NeilHanlon | yeah i'm busy this morning with releasing rocky 9.2 | 15:10 |
| damiandabrowski | jrosser: your question about a list of things that need to be done before release is super valid, however I can't answer it :D | 15:11 |
| damiandabrowski | for sure we need to fix gating for magnum and zun | 15:11 |
| jrosser | should we start an etherpad? | 15:11 |
| jrosser | i think we need to be all over reviews/recheck etc this week and time is slipping by | 15:11 |
| damiandabrowski | good idea | 15:12 |
| damiandabrowski | #link https://etherpad.opendev.org/p/openstack-ansible-antelope | 15:12 |
| NeilHanlon | not sure if useful but | 15:13 |
| NeilHanlon | #link https://etherpad.opendev.org/p/osa-bobcat-ptg | 15:13 |
| NeilHanlon | if we want, we should be able to upgrade rocky to ovn3.1 now. i will introduce a patch for that soon, once dib builds us a new image | 15:14 |
| damiandabrowski | thanks Neil! | 15:14 |
| jrosser | so apart from the ehterpad there is a ton of stuff to review | 15:18 |
| damiandabrowski | yeah, i'll spend some time on reviews this week | 15:20 |
| damiandabrowski | I also need to work on your suggestions related to tls backend in ironic | 15:20 |
| jrosser | some of that might just be documentation i think | 15:21 |
| damiandabrowski | that's right, like this ipa-insecure=1 parameter | 15:22 |
| noonedeadpunk | Sry, I'm in the middle of nowhere now( | 15:23 |
| NeilHanlon | noonedeadpunk: hope all is okay! | 15:52 |
| damiandabrowski | #endmeeting | 15:58 |
| opendevmeet | Meeting ended Tue May 16 15:58:45 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:58 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/openstack_ansible_meeting/2023/openstack_ansible_meeting.2023-05-16-15.06.html | 15:58 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/openstack_ansible_meeting/2023/openstack_ansible_meeting.2023-05-16-15.06.txt | 15:58 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/openstack_ansible_meeting/2023/openstack_ansible_meeting.2023-05-16-15.06.log.html | 15:58 |
| spatel | Does cinder support incremental backup? | 16:14 |
| spatel | currently if i take backup of any volume for example 150G then its creating backup of 150G | 16:14 |
| spatel | Look like by default cinder use default | 16:19 |
| spatel | full backup | 16:19 |
| damiandabrowski | spatel: according to the documentation you're right, you need to explicitly specify `--incremental` flag | 16:29 |
| damiandabrowski | https://docs.openstack.org/cinder/latest/admin/volume-backups.html#:~:text=Without%20the%20incremental%20flag%2C%20a%20full%20backup%20is%20created%20by%20default.%20With%20the%20incremental%20flag%2C%20an%20incremental%20backup%20is%20created. | 16:29 |
| spatel | Do i need to add backup_incremental = True in cinder.conf? | 16:30 |
| damiandabrowski | hmm, where did you find backup_incremental option? I don't see it anywhere | 16:38 |
| spatel | damiandabrowski hmm! i found that option in my google. try to find same link again | 16:41 |
| spatel | You are saying just pass --incremental flag should be enough correct? does Horizon support that flag? | 16:42 |
| damiandabrowski | so: I'm not using cinder backup :D I just went through docs and code | 16:45 |
| damiandabrowski | so as long as your driver supports incremental backups, then it should be just a matter of passing --incremental flag | 16:46 |
| spatel | what do you use for backup of volume? | 16:46 |
| damiandabrowski | horizon should support incremental backups but it's a fairly new feature: https://github.com/openstack/horizon/commit/61df006f4ab41aed23e614ac11874a92b1d10256 | 16:46 |
| damiandabrowski | some important docs w.r.t cinder backup: | 16:46 |
| damiandabrowski | https://docs.openstack.org/cinder/latest/configuration/block-storage/backup/ceph-backup-driver.html | 16:46 |
| damiandabrowski | https://docs.openstack.org/openstack-ansible-os_cinder/latest/configure-cinder.html#using-ceph-for-cinder-backups | 16:46 |
| spatel | when i pass --incremental flag then it endup in error state | 16:47 |
| spatel | may be need to play with backup_ceph_chunk_size = 134217728 | 16:47 |
| damiandabrowski | we have a custom "disaster recovery" service: https://docs.cleura.cloud/background/disaster-recovery/ | 16:48 |
| spatel | damiandabrowski so it do volume snapshot ? | 16:49 |
| damiandabrowski | yeah | 16:50 |
| jrosser | cinder backup is manual isnt it, which is really not very compatible with human behaviours | 16:51 |
| spatel | Hmm! so i should use cinder volume snapshot in that case if backup is not great. | 16:52 |
| jrosser | perhaps? depends why you want to do it | 16:53 |
| jrosser | if you want an autonomous thing creating backups of your instances continually then cinder backup has a feature gap there which i don't really understand | 16:53 |
| jrosser | and maybe you do/dont want those backups to be on the same storage backend | 16:54 |
| jrosser | cinder backup snapshotting inside the same ceph cluster might cover some use cases of backups, but not really one where the ceph fails | 16:55 |
| spatel | Thanks jrosser But what is the big difference between cinder backup and cinder snapshot of volume? | 16:56 |
| jrosser | well, cinder backup can have a different storage backend, if you want | 16:57 |
| spatel | I am trying to find backup solution and not sure what i should use and why. I would like to implement some kind of scheduler who do it for me. | 16:57 |
| jrosser | lime ceph -> NFS | 16:57 |
| spatel | jrosser ah!! | 16:57 |
| jrosser | *like | 16:57 |
| spatel | So snapshot only for single backend like ceph only | 16:57 |
| jrosser | so really depends on what/why you want to do it, and what you are trrying to protext against | 16:57 |
| spatel | I have only ceph storage | 16:57 |
| jrosser | well i mnaser has already shared this with us before https://github.com/vexxhost/staffeln | 16:58 |
| spatel | Yes, I remembered it now. I need to play and understand how I can integrate it with but I need to find out what I should use cinder-backup or volume snapshot | 17:02 |
| noonedeadpunk | jrosser: what we wanted t oland for the release was stated in ptg etherpad https://etherpad.opendev.org/p/osa-bobcat-ptg | 17:17 |
| noonedeadpunk | NeilHanlon: Yeah, was sorting out visa stuff, now I'm 1 step closer, despite that required another travel abroad... | 17:18 |
| NeilHanlon | ouch | 17:18 |
| noonedeadpunk | Now if post service won't fail delivery, I have some chance to get it in time | 17:20 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Replace etcd usage with zookeeper for zun https://review.opendev.org/c/openstack/openstack-ansible/+/883194 | 17:21 |
| jrosser | noonedeadpunk: yeah, I meant actually what we need to concentrate on this week because a bunch of ptg stuff won’t get done | 17:34 |
| jrosser | mainly just to track the in flight patches to help review | 17:34 |
| noonedeadpunk | yeah, but we kinda got all done I think | 17:34 |
| noonedeadpunk | I will create one though | 17:34 |
| noonedeadpunk | btw, magnum seems happy now https://review.opendev.org/c/openstack/openstack-ansible-os_magnum/+/880484 | 17:38 |
| noonedeadpunk | oh.... do we have neutron issue now? | 17:40 |
| noonedeadpunk | `nothing provides openvswitch2.17 needed by rdo-openvswitch-2:2.17-4.el9s.noarch` | 17:40 |
| noonedeadpunk | NeilHanlon: was you discussing that already? I guess we should just revert workaround as new ovs is ready? | 17:41 |
| NeilHanlon | that's my fault.. i did not tag 2.17 into the release repo fast enough | 17:42 |
| NeilHanlon | we can revert, yes, but also 2.17 will be available from NFV sig in like 1 day | 17:42 |
| NeilHanlon | let me just unpatch that | 17:42 |
| noonedeadpunk | Well, I think for master we wanna it to be 3.1 | 17:42 |
| noonedeadpunk | rest can wait for couple of days for sure | 17:43 |
| noonedeadpunk | btw, the link with patches - just started working on that - https://etherpad.opendev.org/p/osa-antelope-leftover-patches | 17:43 |
| noonedeadpunk | but I think it's pretty much it | 17:45 |
| opendevreview | Neil Hanlon proposed openstack/openstack-ansible-openstack_hosts master: Revert pinning rdo-deps for openvswitch 2.17 for rocky https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/883294 | 17:50 |
| NeilHanlon | this was my weekend https://bugzilla.redhat.com/show_bug.cgi?id=2203919 | 17:56 |
| NeilHanlon | only affects powerpc, thankfully | 17:56 |
| * noonedeadpunk wonders who cares abot powerpc | 18:07 | |
| spatel | damiandabrowski jrosser what is wrong here? Am i missing something - https://paste.opendev.org/show/bu7koZzXApDyKlTcLU08/ | 18:09 |
| spatel | Do i need to take first default backup and then incremental ? | 18:10 |
| noonedeadpunk | yes, furst must be full | 18:10 |
| NeilHanlon | noonedeadpunk: me too.... | 18:10 |
| noonedeadpunk | incremental arfe created out of full and should be merged with nearest full on recovery | 18:10 |
| spatel | noonedeadpunk thanks!!! it works but why incremental also has same size 10G ? | 18:15 |
| spatel | https://paste.opendev.org/show/bJ5pUbctqJ2nRRgfWMiD/ | 18:15 |
| spatel | Am i missing anything in command? | 18:15 |
| noonedeadpunk | spatel: there's a bug in some mariadb version, maybe damiandabrowski does recall version | 18:16 |
| noonedeadpunk | but idea is that they broke incremental ones | 18:16 |
| noonedeadpunk | https://opendev.org/openstack/openstack-ansible-galera_server/commit/d65197c8696ca8dc1cbfd6af376494df56fea8ee | 18:16 |
| noonedeadpunk | spatel: ^ | 18:16 |
| spatel | noonedeadpunk oh wait... This is cinder backup and you are saying related to mariaDB? | 18:17 |
| noonedeadpunk | oh.... | 18:19 |
| noonedeadpunk | ok, then I don't know :) | 18:20 |
| spatel | :) | 18:20 |
| noonedeadpunk | yeah, I was reffering to galera, sorry | 18:20 |
| spatel | no problem | 18:20 |
| spatel | I am trying to take cinder-backup of volume and trying to test how incremental backup work.. but look like I am going something stupid https://paste.opendev.org/show/bJ5pUbctqJ2nRRgfWMiD/ | 18:21 |
| spatel | Even for incremental backup its showing same size. Is that normal? | 18:21 |
| spatel | Feels like ceph doesn't support incremental back | 18:38 |
| spatel | Feels like ceph doesn't support incremental backup | 18:38 |
| spatel | https://opendev.org/openstack/cinder/commit/5018727f8e18992e64c55b84f0d5b39ae7c70b32 | 18:40 |
| noonedeadpunk | I depends on the backend I think | 18:43 |
| noonedeadpunk | as if you store backups in S3 (rgw) - it should be supported I think | 18:44 |
| noonedeadpunk | but not if it's ceph native | 18:44 |
| noonedeadpunk | though I don't have really much experience with that | 18:44 |
| spatel | Feels like ceph isn't supported option for incremental | 18:45 |
| spatel | I will ask in mailing list and see | 18:45 |
| opendevreview | Merged openstack/openstack-ansible-os_manila master: Don't install zlibc, it was removed from Debian https://review.opendev.org/c/openstack/openstack-ansible-os_manila/+/883067 | 19:49 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!