| *** recyclehero has quit IRC | 00:26 | |
| *** recyclehero has joined #openstack-swift | 00:27 | |
| *** mikecmpbll has quit IRC | 00:28 | |
| *** mikecmpbll has joined #openstack-swift | 00:33 | |
| *** gyee has quit IRC | 01:26 | |
| *** rcernin has quit IRC | 01:28 | |
| *** rcernin has joined #openstack-swift | 01:43 | |
| *** rcernin has quit IRC | 01:43 | |
| *** rcernin has joined #openstack-swift | 01:44 | |
| *** psachin has joined #openstack-swift | 03:35 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-swift | 04:33 | |
| *** m75abrams has joined #openstack-swift | 04:53 | |
| *** Hamidreza has joined #openstack-swift | 05:26 | |
| *** mikecmpbll has quit IRC | 05:39 | |
| *** mikecmpbll has joined #openstack-swift | 05:40 | |
| *** mikecmpbll has quit IRC | 06:58 | |
| *** mikecmpbll has joined #openstack-swift | 07:01 | |
| *** zigo has joined #openstack-swift | 07:12 | |
| *** rcernin has quit IRC | 07:19 | |
| *** persia has quit IRC | 08:32 | |
| *** persia has joined #openstack-swift | 08:32 | |
| *** tkajinam has quit IRC | 08:47 | |
| *** mikecmpbll has quit IRC | 10:02 | |
| *** mikecmpbll has joined #openstack-swift | 10:05 | |
| *** rcernin has joined #openstack-swift | 10:20 | |
| *** rcernin has quit IRC | 10:26 | |
| *** rcernin has joined #openstack-swift | 10:26 | |
| *** godog has quit IRC | 11:23 | |
| *** godog has joined #openstack-swift | 11:23 | |
| *** ozzzo has quit IRC | 12:10 | |
| *** psachin has quit IRC | 12:35 | |
| *** rcernin has quit IRC | 13:38 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Test proxy-server.conf-sample https://review.opendev.org/755087 | 14:18 |
|---|---|---|
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Add staticweb to default pipeline https://review.opendev.org/755132 | 14:18 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Log error processing manifest as ServerError https://review.opendev.org/752770 | 14:18 |
| *** mikecmpbll has quit IRC | 14:29 | |
| *** mikecmpbll has joined #openstack-swift | 14:32 | |
| *** Hamidreza has quit IRC | 14:47 | |
| *** m75abrams has quit IRC | 14:55 | |
| clayg | grepping through my logs on my saio for " SW " i'm realizing that all my s3api requests are getting hit by that env hijacking | 15:13 |
| clayg | that's some bs `%20StaticWeb` | 15:14 |
| *** djhankb has quit IRC | 15:54 | |
| *** djhankb has joined #openstack-swift | 15:54 | |
| *** gyee has joined #openstack-swift | 16:12 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Log storage policy with s3api requests https://review.opendev.org/755587 | 16:25 |
| *** mikecmpbll has quit IRC | 16:33 | |
| *** mikecmpbll has joined #openstack-swift | 16:36 | |
| *** renich has joined #openstack-swift | 17:21 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Update master for stable/victoria https://review.opendev.org/752752 | 17:29 |
| timburke | clayg, i'm increasingly convinced that when s3_acl is enabled, s3api needs to set REMOTE_USER to some special value | 17:31 |
| clayg | seems reasonable to me - mostly in the context of "static web breaks everything again" https://bugs.launchpad.net/swift/+bug/1833287 | 17:33 |
| openstack | Launchpad bug 1833287 in OpenStack Object Storage (swift) "staticweb redirects cause 500s in s3api" [Undecided,Confirmed] | 17:33 |
| clayg | I mean... what is REMOTE_USER supposed to indicate exactly? why does staticweb not apply for ... what? "authenticated" requests? why does it get screw up everything else? | 17:33 |
| timburke | yeah, so my understanding is that auth middlewares will populate it when they see a user that they recognize. often, it'll be populated with both users and groups/roles, though the format varies: tempauth sets it to a csv, keystoneauth sets it to a tuple, swiftstackauth sets it to a NUL-delimited string | 17:39 |
| timburke | in theory, auth middlewares could use it to do some cross-IDP magic, but i think generally each just thinks about itself | 17:40 |
| clayg | well, is s3_acl handling an authenticated request - can it put some kind of identity in REMOTE_USER? maybe static web is just terrible? | 17:41 |
| timburke | s3_acl almost becomes its own independent auth system that's mostly divorced from anything you can do on the swift side -- it's part of why i don't much like the option :-( | 17:45 |
| timburke | yeah, it probably *could* do something more useful than just some "i'm s3api doing my own thing" sentinel... hmm... | 17:46 |
| timburke | (fwiw, timur has a patch to try to unify the REMOTE_USER formats: https://review.opendev.org/#/c/630415/ -- i should dust that off) | 17:47 |
| patchbot | patch 630415 - swift - Fix the handling of keystone groups in tempauth. - 5 patch sets | 17:47 |
| *** renich has quit IRC | 18:10 | |
| *** renich has joined #openstack-swift | 18:10 | |
| *** fingo has joined #openstack-swift | 18:17 | |
| *** renich has quit IRC | 18:21 | |
| *** renich has joined #openstack-swift | 18:38 | |
| *** renich has quit IRC | 18:47 | |
| timburke | clayg, the changes in https://review.opendev.org/#/c/755087/4/swift/common/wsgi.py seem weird -- why hasn't this been an issue before?? we use ConfigString for container-sync if there isn't an internal_client_conf_path configured -- is it some difference between using the InternalClient constructor and calling loadapp() directly? or is that busted too, and nobody noticed? i'm so confused... | 19:05 |
| patchbot | patch 755087 - swift - Test proxy-server.conf-sample - 4 patch sets | 19:05 |
| *** renich has joined #openstack-swift | 19:06 | |
| clayg | i don't think we ever tried to load a full pipeline from a ConfigString or we would have seen the problem with middlewares like dlo that re-parse the config for legacy settings | 19:06 |
| clayg | it's possible that whole mess is a distraction - rather than trying to test that we keep env's through the pipeline we could just kick out stupid broken nonsense like staticweb | 19:07 |
| clayg | also I could make a new test-case that validates the default pipeline - but not ALSO validate it works with a ConfigString pipeline | 19:08 |
| timburke | oh, yeah... so container-sync uses the default allow_modify_pipeline=False... so we won't auto-insert any of the middlewares that used to be in the proxy app. got it | 19:15 |
| renich | I'm having issues pushing my corrections for the apache deployment. Here's a screenshot: https://paste.pics/28ca044d8112f2067a9e61f180cdd236 | 19:16 |
| renich | I dunno what I'm missing... | 19:16 |
| clayg | maybe it has something to do with https://review.opendev.org/#/c/748043/ being closed? | 19:19 |
| patchbot | patch 748043 - swift - docs: Clean up some formatting around using server... (MERGED) - 1 patch set | 19:19 |
| clayg | like if you've rebased or something - you probably can't push up a new version of an already merged change 🤷♂️ | 19:19 |
| renich | clayg: rebase? OK, I gotta find out how to do this properly | 19:20 |
| clayg | maybe rebase the whole stack on on eefe35755c8b9f53ed9f018cac9886879b68b0b6 or just diff d659d84ea and master to see what changed? | 19:22 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: New proxy logging field for wire status https://review.opendev.org/755639 | 19:54 |
| clayg | @timburke ^ what do you think of adding catch_errors in the pipeline before logging instead of messing with env message passing p 752770 | 19:59 |
| patchbot | https://review.opendev.org/#/c/752770/ - swift - Log error processing manifest as ServerError - 4 patch sets | 19:59 |
| clayg | i think YOU had a patch that tried to add bye counting to proxy logging somewhere too? | 20:00 |
| renich | OK, I figured out the rebase. Now, it says I am not allowed. https://paste.pics/2aaf5082dad042a31e282aebfdb04934 | 20:43 |
| renich | I thought I signed the CLA and stuff. | 20:43 |
| renich | ICLA, that is. | 20:47 |
| timburke | renich, have you double-checked that your ssh key is set up? https://review.opendev.org/#/settings/ssh-keys | 20:50 |
| renich | timburke: I just did that. It is. | 20:50 |
| renich | I can share a screenshot, hehe. | 20:51 |
| renich | is it being rejected because of the titles? | 20:51 |
| timburke | oh, the merge! i see now. try a `git rebase origin/master`; i think that might square it? you may also want to squash them all into a single commit with a title like "Clean up apache deployment docs" | 20:55 |
| renich | OK, the rebase is done. But, the push, isn't working. I'll try squashing. | 20:56 |
| timburke | the long message lines really are just warnings -- i push up commits with long messages a decent bit | 20:56 |
| timburke | if you do something like `git log --oneline origin/master..` are there still any merges? | 20:57 |
| renich | timburke: https://paste.centos.org/view/04624beb | 20:59 |
| timburke | looks good... is it the same "you are not allowed to upload merges" error as before? | 21:04 |
| renich | I'm in the middle of squashing, hehe | 21:05 |
| openstackgerrit | Renich Bon Ciric proposed openstack/swift master: docs: Switching to /srv/www from /var/www to be more FHS 3.0 conformat https://review.opendev.org/755651 | 21:06 |
| openstackgerrit | Renich Bon Ciric proposed openstack/swift master: Clean up apache deployment docs https://review.opendev.org/755652 | 21:06 |
| renich | timburke: OK, that worked, hehe. The squash thingy | 21:06 |
| timburke | thanks, renich! | 21:40 |
| renich | timburke: I am glad to help. You've helped me much! | 21:55 |
| renich | I see the comments... trying to figure out how to comment back, hehe. | 22:00 |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift master: Fix a race condition in case of cross-replication https://review.opendev.org/754242 | 22:06 |
| *** rcernin has joined #openstack-swift | 22:24 | |
| *** renich has quit IRC | 22:40 | |
| *** tkajinam has joined #openstack-swift | 23:00 | |
| openstackgerrit | Merged openstack/swift master: docs: Switching to /srv/www from /var/www to be more FHS 3.0 conformat https://review.opendev.org/755651 | 23:06 |
| openstackgerrit | Merged openstack/swift master: Clean up apache deployment docs https://review.opendev.org/755652 | 23:06 |
| *** djhankb has quit IRC | 23:15 | |
| *** djhankb has joined #openstack-swift | 23:15 | |
| openstackgerrit | Merged openstack/swift master: Update master for stable/victoria https://review.opendev.org/752752 | 23:54 |
| *** mikecmpbll has quit IRC | 23:57 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!