semantic | Hi guys! | 04:48 |
---|---|---|
semantic | We somehow missed the moment when glance-scrubber was deprecated. We've been actively used glance scrubber and delayed_delete feature in our production cloud for last five years. We use swift as our backend storage, so I believe image deletion time is actually a function of data size. Moreover, we tend to delete large number of large images (backups/snapshots) of instances on a schedule. It seems that glance-scrubber | 04:48 |
semantic | removal will negatively affect us, or maybe we're missing something here, that you may advice?.. Does nobody really use scrubber? | 04:48 |
opendevreview | Abhishek Kekane proposed openstack/glance master: Remove S3 credentials from debug log https://review.opendev.org/c/openstack/glance/+/940358 | 05:29 |
*** tkajinam is now known as Guest7334 | 09:05 | |
*** whoami-rajat_ is now known as whoami-rajat | 14:02 | |
*** Guest7381 is now known as bbezak | 14:56 | |
bbezak | Hi - I'm wondering about back-portability of this patch (and glance_store counterpart) - https://review.opendev.org/c/openstack/glance/+/934716 | 14:59 |
rosmaita | semantic: I put an item on the weekly meeting agenda to discuss glance-scrubber: https://etherpad.opendev.org/p/glance-team-meeting-agenda | 14:59 |
bbezak | bug is pretty nasty. slowing down significantly image list operation on bigger clouds | 14:59 |
bbezak | in caracal | 15:00 |
rosmaita | bbezak: seems like a reasonable backport, i think the main thing to do is to check to make sure there's a tempest test or something that is using the store weight, so we know it won't cause a regression | 15:04 |
rosmaita | (since the code is currently only in master) | 15:04 |
rosmaita | or if you are using the patch locally with success, you could report that | 15:05 |
bbezak | thx. I'll test that in test env | 15:21 |
bbezak | (and report back) | 15:37 |
rosmaita | sounds good! | 15:58 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!