*** hoonetorg has quit IRC | 01:38 | |
*** hoonetorg has joined #openstack-swift | 01:51 | |
*** openstackgerrit has quit IRC | 02:04 | |
*** openstackgerrit has joined #openstack-swift | 02:08 | |
openstackgerrit | Thiago da Silva proposed openstack/swift master: s3api: Implement object versioning API https://review.opendev.org/673682 | 02:08 |
---|---|---|
*** psachin has joined #openstack-swift | 03:35 | |
*** tkajinam has quit IRC | 07:02 | |
*** tkajinam has joined #openstack-swift | 07:04 | |
openstackgerrit | Merged openstack/swift master: container-sync: Sync static links similar to how we sync SLOs https://review.opendev.org/698139 | 07:24 |
*** tkajinam_ has joined #openstack-swift | 07:52 | |
*** tkajinam has quit IRC | 07:55 | |
*** rdejoux has joined #openstack-swift | 08:13 | |
*** tesseract has joined #openstack-swift | 08:13 | |
*** cjloader has quit IRC | 08:16 | |
*** tkajinam_ has quit IRC | 08:18 | |
*** rpittau|afk is now known as rpittau | 08:18 | |
*** pcaruana has joined #openstack-swift | 08:25 | |
openstackgerrit | Merged openstack/swift master: New Object Versioning mode https://review.opendev.org/682382 | 09:16 |
*** ccamacho has joined #openstack-swift | 09:25 | |
*** rpittau is now known as rpittau|bbl | 11:21 | |
*** tomha has joined #openstack-swift | 12:07 | |
*** rpittau|bbl is now known as rpittau | 12:55 | |
*** psachin has quit IRC | 13:29 | |
*** jvisser has joined #openstack-swift | 14:41 | |
*** paladox is now known as paladox|UKInEU | 14:56 | |
*** cjloader has joined #openstack-swift | 15:08 | |
*** theintern_ has joined #openstack-swift | 15:50 | |
*** theintern_ has quit IRC | 16:07 | |
*** theintern_ has joined #openstack-swift | 16:19 | |
*** gyee has joined #openstack-swift | 16:29 | |
*** zaitcev has joined #openstack-swift | 16:29 | |
*** ChanServ sets mode: +v zaitcev | 16:29 | |
*** ccamacho has quit IRC | 16:34 | |
*** theintern_ has quit IRC | 16:37 | |
*** theintern_ has joined #openstack-swift | 16:49 | |
openstackgerrit | Tim Burke proposed openstack/swift master: Add unittest2 back to lower-constraints.txt https://review.opendev.org/704356 | 16:50 |
timburke | thanks for fixing up p 673682, tdasilva! | 16:51 |
patchbot | https://review.opendev.org/#/c/673682/ - swift - s3api: Implement object versioning API - 51 patch sets | 16:51 |
*** rdejoux has quit IRC | 16:59 | |
timburke | hmm... based on my read of the code, i'm pretty sure https://github.com/openstack/swift/blob/2.23.0/swift/common/middleware/s3api/s3request.py#L535-L536 are currently wsgi strings... should probably be native, though... | 17:05 |
*** paladox|UKInEU is now known as paladox | 17:09 | |
*** gregwork has joined #openstack-swift | 17:15 | |
*** theintern_ has quit IRC | 17:23 | |
*** tesseract has quit IRC | 17:38 | |
*** rpittau is now known as rpittau|afk | 18:18 | |
*** gregwork has quit IRC | 19:25 | |
openstackgerrit | Tim Burke proposed openstack/swift master: s3api: use native strings in s3api.auth_details access_key https://review.opendev.org/704413 | 20:13 |
*** gyee has quit IRC | 20:27 | |
*** gyee has joined #openstack-swift | 20:28 | |
*** openstackstatus has joined #openstack-swift | 20:28 | |
*** ChanServ sets mode: +v openstackstatus | 20:28 | |
openstackgerrit | Tim Burke proposed openstack/swift master: Middleware that allows a user to have quoted Etags https://review.opendev.org/700056 | 20:53 |
openstackgerrit | Romain LE DISEZ proposed openstack/swift master: proxy-logging: add fields ttfb and pid https://review.opendev.org/704424 | 20:56 |
openstackgerrit | Romain LE DISEZ proposed openstack/swift master: SSYNC: log body of errors in SSYNC subrequests https://review.opendev.org/704430 | 21:14 |
openstackgerrit | Romain LE DISEZ proposed openstack/swift master: Mark a container reported if account was reclaimed https://review.opendev.org/704435 | 21:29 |
clayg | rledisez: with the patches! 👍 | 21:32 |
rledisez | clayg: just going through the patches we run in production but that are not proposed yet | 21:34 |
rledisez | really small stuff, but useful I think | 21:34 |
clayg | undoubtly!!! | 21:42 |
timburke | love 'em! | 21:57 |
timburke | i'm a little nervous about the possibility of containers *with data* in accounts that have been reclaimed though... | 21:58 |
timburke | idk that there's an obvious right thing to do in that case though... | 21:59 |
openstackgerrit | Tim Burke proposed openstack/swift master: s3api: use native strings in s3api.auth_details access_key https://review.opendev.org/704413 | 22:01 |
rledisez | timburke: actually it does not delete the data, just the container db (as the account db was already deleted). but the datafile stays there (I admit it might be harder to find the objects) | 22:02 |
timburke | rledisez, but if the container db has objects in it, it won't get reclaimed, right? so you've not only got the dark .data files lurking on your object nodes but also these dark .db files... | 22:06 |
timburke | but since the account db's *actually* gone, there's nothing to signal that any of this should've been deleted :-/ | 22:07 |
timburke | like, the 500 in account/server.py's no good. but there are two "reasonable" things we could do to fix it -- 404 (as in this patch) or pop the db into existence (like we would if it started with auto_account_prefix) | 22:08 |
timburke | er, auto_create_account_prefix -- you know what i mean | 22:09 |
*** rcernin has joined #openstack-swift | 22:10 | |
mattoliverau | morning | 22:16 |
timburke | maybe it'd make sense to quarantine if we've done all those PUTs, nobody has an account DB, and the container db is not marked deleted? | 22:37 |
timburke | idk | 22:37 |
*** diablo_rojo has quit IRC | 22:42 | |
rledisez | i think the right thing to do is to delete these objects. dark data is bad, we should fight it :) | 22:53 |
rledisez | what about re-creating the account in a deleted state so that the reaper will get another chance to clean it? | 22:53 |
*** tkajinam has joined #openstack-swift | 22:55 | |
timburke | seems reasonable -- if we do it in the account-server as part of the container-update flow, we'll want to be careful about choosing timestamps in case one of the other replicas has a db | 23:19 |
tdasilva | rledisez, timburke: was wondering if it would make sense to add a "deprecation flag" to /info requests. I'm trying to think of effective ways of communicating deprecation plans to API users and operators and was thinking that highlighting current deprecations (or plans) for a given option could be a good way | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!