*** gyee has quit IRC | 00:03 | |
*** itlinux has quit IRC | 00:13 | |
*** threestrands has quit IRC | 00:38 | |
notmyname | I worked with timur on a couple of swiftclient patches today. they're both in the gate now. I'll update the authors/changelog patch, and then do the tag request | 00:51 |
---|---|---|
*** rcernin_ has joined #openstack-swift | 00:56 | |
*** rcernin has quit IRC | 00:58 | |
openstackgerrit | John Dickinson proposed openstack/python-swiftclient master: authors/changelog update for next release https://review.openstack.org/584081 | 00:59 |
timur | yay! | 01:02 |
timur | thanks notmyname! | 01:02 |
*** links has joined #openstack-swift | 01:17 | |
*** ^andrea^ has quit IRC | 01:27 | |
mattoliverau | nice | 02:02 |
*** armaan has quit IRC | 02:19 | |
*** armaan has joined #openstack-swift | 02:20 | |
ktsuyuzaki | good morning | 02:27 |
ktsuyuzaki | I'm back from long summer vacation | 02:28 |
ktsuyuzaki | and it seems like i missed my nick at my irc client? | 02:28 |
ktsuyuzaki | let me check... | 02:28 |
*** ktsuyuzaki is now known as kota_ | 02:28 | |
kota_ | hello, testing. | 02:28 |
kota_ | it looks to work. | 02:28 |
*** psachin` has joined #openstack-swift | 02:50 | |
mattoliverau | kota_: welcome back | 03:13 |
*** ukaynar_ has joined #openstack-swift | 03:23 | |
*** ukaynar_ has quit IRC | 03:29 | |
*** ukaynar_ has joined #openstack-swift | 03:29 | |
*** ukaynar_ has quit IRC | 03:30 | |
*** ukaynar_ has joined #openstack-swift | 03:30 | |
*** ukaynar_ has quit IRC | 03:35 | |
kota_ | mattoliverau: o/ | 03:41 |
openstackgerrit | Merged openstack/python-swiftclient master: Add close() to _RetryBody. https://review.openstack.org/549901 | 03:48 |
*** armaan has quit IRC | 04:15 | |
*** psachin` has quit IRC | 04:15 | |
*** armaan has joined #openstack-swift | 04:18 | |
*** psachin` has joined #openstack-swift | 04:19 | |
*** mahatic has quit IRC | 04:23 | |
*** itlinux has joined #openstack-swift | 04:28 | |
*** mahatic has joined #openstack-swift | 04:28 | |
*** kei_yama has quit IRC | 04:43 | |
*** armaan has quit IRC | 05:05 | |
*** armaan has joined #openstack-swift | 05:06 | |
*** drewn3ss has quit IRC | 05:07 | |
*** armaan has quit IRC | 05:09 | |
*** armaan has joined #openstack-swift | 05:10 | |
*** itlinux has quit IRC | 05:30 | |
*** viks_ has joined #openstack-swift | 05:30 | |
*** cshastri has joined #openstack-swift | 05:31 | |
*** armaan has quit IRC | 05:43 | |
*** armaan has joined #openstack-swift | 05:44 | |
*** ccamacho has joined #openstack-swift | 05:44 | |
*** drewn3ss has joined #openstack-swift | 05:53 | |
timburke_ | notmyname: idk where you are in the swiftclient tagging yet, but i think https://review.openstack.org/#/c/568914/ might still be worth considering | 06:15 |
patchbot | patch 568914 - python-swiftclient - Back out some version bumps | 06:15 |
*** armaan_ has joined #openstack-swift | 06:26 | |
timburke_ | cleaning up and landing https://review.openstack.org/#/c/581921/ might also be nice. seems like an easy win (that may also generate some good will with -infra) | 06:28 |
patchbot | patch 581921 - python-swiftclient - Log exceptions received during retrying | 06:29 |
*** armaan has quit IRC | 06:29 | |
*** pcaruana has joined #openstack-swift | 06:34 | |
mattoliverau | timburke_: do it! | 06:35 |
timburke_ | mattoliverau: i might need to sleep first ;-) | 06:35 |
mattoliverau | pfft :P | 06:35 |
timburke_ | just got back from a trip, trying to catch up a little before bed | 06:35 |
mattoliverau | We appreciate that, I've been meaning to review a bunch today, but a work thing that was only suppose to take the morning has taken over my day. But on the plus side I've now pushed 2 monasca swift check plugins. So not completely wasted time ;) | 06:39 |
timburke_ | yay! | 06:40 |
mattoliverau | for more reference: https://oliver.net.au/?p=358 | 06:45 |
mattoliverau | Though I'm sure SwiftStack already have much better monitoring/metrics. but hopefully this will open the doors for others :) | 06:46 |
openstackgerrit | Merged openstack/python-swiftclient master: Properly handle unicode headers. https://review.openstack.org/562059 | 06:52 |
*** tesseract has joined #openstack-swift | 07:18 | |
kota_ | mattoliverau: nice work on the monasca plugins | 07:34 |
*** neonpastor has joined #openstack-swift | 07:36 | |
mattoliverau | ta :) | 07:41 |
*** rcernin_ has quit IRC | 07:56 | |
*** geaaru has joined #openstack-swift | 08:07 | |
*** rcernin_ has joined #openstack-swift | 08:09 | |
*** gkadam has joined #openstack-swift | 08:13 | |
*** armaan_ has quit IRC | 08:34 | |
*** armaan has joined #openstack-swift | 08:34 | |
*** kei_yama has joined #openstack-swift | 08:50 | |
*** armaan has quit IRC | 09:07 | |
*** armaan has joined #openstack-swift | 09:08 | |
*** kei_yama has quit IRC | 10:14 | |
*** mvk_ has quit IRC | 10:23 | |
*** rcernin_ has quit IRC | 10:52 | |
*** mvk_ has joined #openstack-swift | 10:53 | |
*** mvk_ has quit IRC | 11:33 | |
*** stephen_m has joined #openstack-swift | 11:38 | |
*** mvk_ has joined #openstack-swift | 11:47 | |
openstackgerrit | Alexander Bolshakov proposed openstack/python-swiftclient master: Fix update when PUT on containers not allowed https://review.openstack.org/585325 | 11:51 |
*** linkmark has joined #openstack-swift | 11:54 | |
*** mikecmpbll has joined #openstack-swift | 12:00 | |
*** armaan has quit IRC | 12:26 | |
*** armaan has joined #openstack-swift | 12:26 | |
*** psachin` has quit IRC | 12:30 | |
*** mikecmpbll has quit IRC | 12:38 | |
*** mikecmpbll has joined #openstack-swift | 12:46 | |
*** ianychoi has quit IRC | 13:08 | |
*** ianychoi has joined #openstack-swift | 13:08 | |
*** ccamacho1 has joined #openstack-swift | 13:43 | |
*** ccamacho has quit IRC | 13:43 | |
*** neonpastor has quit IRC | 13:58 | |
*** links has quit IRC | 14:31 | |
*** armaan has quit IRC | 14:37 | |
*** armaan has joined #openstack-swift | 14:38 | |
timburke_ | notmyname: we really ought to land https://review.openstack.org/#/c/541817/ | 14:43 |
patchbot | patch 541817 - swift (stable/queens) - Update UPPER_CONSTRAINTS_FILE for stable/queens | 14:43 |
timburke_ | also, if you wanted https://review.openstack.org/#/c/568470/ to land, it'll need a +2 vote | 14:45 |
patchbot | patch 568470 - swift (stable/queens) - Deprecate auth_uri option | 14:45 |
*** stephen_m has quit IRC | 14:47 | |
*** stephen_m has joined #openstack-swift | 14:50 | |
*** stephen_m has quit IRC | 14:50 | |
notmyname | timburke_: welcome back :-) | 14:54 |
timburke_ | NOW GO MERGE THINGS! | 14:54 |
timburke_ | ;-) | 14:54 |
notmyname | THINGS ARE LANDING NOW! | 14:57 |
timburke_ | yay! | 14:58 |
timburke_ | ty | 14:58 |
timburke_ | oh yeah, and someone besides me probably ought to look at https://review.openstack.org/#/c/578075/ -- otherwise i'll get antsy and just land it anyway | 14:59 |
patchbot | patch 578075 - swift - Add keymaster to fetch root secret from KMIP service | 14:59 |
notmyname | timburke_: any reson not to land the authors/changelog for swiftclient? I'd like to get that started in the gate asap. but you mentioned a couple of things on patch 562059 that may need to be conidered | 15:01 |
patchbot | https://review.openstack.org/#/c/562059/ - python-swiftclient - Properly handle unicode headers. (MERGED) | 15:01 |
notmyname | timburke_: you're clearly still on eastern time zone ;-) | 15:01 |
timburke_ | i clearly don't have a kid to get to daycare :P | 15:01 |
notmyname | lol | 15:02 |
timburke_ | *shrug* i don't think those are blockers, but i'll try to get a patch up to address the latter two comments today | 15:03 |
*** cshastri has quit IRC | 15:03 | |
timburke_ | want me to give the a/cl patch a once over? | 15:03 |
timburke_ | oh! right! the dep bumps. that patch *is* viable now -- but if we do a release, i'll probably just abandon it | 15:04 |
timburke_ | https://review.openstack.org/#/c/568914/ | 15:04 |
patchbot | patch 568914 - python-swiftclient - Back out some version bumps | 15:04 |
notmyname | timburke_: may be better to look at https://review.openstack.org/#/c/585325/ instead of the a/cl one, if you have to choose (although a/cl is very short) | 15:04 |
patchbot | patch 585325 - python-swiftclient - Fix update when PUT on containers not allowed | 15:04 |
*** rcernin_ has joined #openstack-swift | 15:05 | |
notmyname | however, that one doesn't seem to have tests... | 15:05 |
timburke_ | yeah, i'd taken a brief look... might be better handled in shell.py... i'll see how ugly that gets | 15:05 |
notmyname | ok, I gotta pull myself away. YOU CAN'T GET ME SUCKED INTO EVERYTHING RIGHT NOW, timburke_! | 15:05 |
notmyname | i need to get ready and go to the office, then I can get pulled into outstanding patches and stuff :-) | 15:06 |
timburke_ | or maybe we could just override success for those responses? i'm just nervous about dropping responses on the floor instead of passing them back to the caller... | 15:06 |
*** ccamacho1 has quit IRC | 15:18 | |
*** rcernin_ has quit IRC | 15:30 | |
*** pcaruana has quit IRC | 15:32 | |
*** gyee has joined #openstack-swift | 15:37 | |
*** itlinux has joined #openstack-swift | 15:49 | |
*** ccamacho has joined #openstack-swift | 16:24 | |
*** ccamacho has quit IRC | 16:25 | |
*** ccamacho has joined #openstack-swift | 16:25 | |
notmyname | good morning again. this time for real | 16:34 |
openstackgerrit | John Dickinson proposed openstack/python-swiftclient master: authors/changelog update for 3.6.0 https://review.openstack.org/584081 | 16:41 |
*** mikecmpbll has quit IRC | 16:41 | |
notmyname | timburke_: ^ | 16:43 |
*** mikecmpbll has joined #openstack-swift | 17:03 | |
*** itlinux_ has joined #openstack-swift | 17:07 | |
timburke_ | ugh, i gotta get https://review.openstack.org/#/c/577874/ into shape... took longer than i'd like to admit to figure out that the 500s i'm seeing are the result of some leftover test data while reviewing that | 17:08 |
patchbot | patch 577874 - swift - Add support for multiple root encryption secrets | 17:08 |
*** itlinux has quit IRC | 17:09 | |
*** mvk_ has quit IRC | 17:09 | |
timur | timburke_: I'll try to look into the keystone interaction that you mentioned. Unless you're way ahead of me on that already | 17:13 |
timburke_ | haven't actually started on that follow-up patch yet; thanks for looking at it | 17:14 |
*** tesseract has quit IRC | 17:17 | |
*** ccamacho has quit IRC | 17:20 | |
*** armaan has quit IRC | 17:25 | |
*** itlinux_ has quit IRC | 17:42 | |
*** itlinux has joined #openstack-swift | 17:42 | |
*** mvk_ has joined #openstack-swift | 17:50 | |
itlinux | hi guys, what's the best way to set a project quota on swift | 17:50 |
itlinux | I remember there is some changes I need to make | 17:50 |
notmyname | itlinux: what sort of quota are you looking for? bytes used? objects used? requests? | 17:50 |
itlinux | and since I have not made some changes to swift for at least a year.. I want to make sure that I do have a quota set to bytes.. | 17:51 |
itlinux | not opbject | 17:51 |
itlinux | just want to make sure they do not go over the disk usage allocatated to that project | 17:52 |
notmyname | is a project mapped to a single swift account? | 17:52 |
itlinux | no I have several accounts on a project | 17:52 |
notmyname | ah | 17:53 |
itlinux | this is on LDAP and using groups for the projects | 17:53 |
notmyname | you can set a soft max bytes quota on an account. but you can't do something like "allow a total of XXX bytes on these N accounts" | 17:54 |
notmyname | that would require running your own utilization reports and then attaching it to the auth system to reject access for write reuqests | 17:54 |
itlinux | ahh ok.. how is the soft quota set? | 17:54 |
notmyname | https://docs.openstack.org/swift/latest/middleware.html#module-swift.common.middleware.account_quotas | 17:55 |
itlinux | ok thanks notmyname: | 17:55 |
notmyname | it references some info included in https://docs.openstack.org/swift/latest/middleware.html#module-swift.common.middleware.container_quotas too | 17:55 |
itlinux | ok.. looks like the quota there is account is per user? | 17:56 |
itlinux | uhm.. or account is project? | 17:56 |
notmyname | there is a swift account (a place where you put stuff, like a bank account). a project is an auth thing that can be mapped to swift accounts). and a user is a thing that has access to particular resources | 17:57 |
itlinux | ok I will dig into that.. | 17:57 |
notmyname | in swift, you can set quotas on either the swift account or a container in a swift account | 17:58 |
itlinux | which means that the tenant can create new containers.. | 17:58 |
itlinux | I am not sure I want them to do that more than x number so that the total will be xMB | 17:59 |
itlinux | is there a way to set max number of containers they can create? | 18:00 |
notmyname | no. but you can limit the max number of containers in an account at the cluster level. but not per account | 18:05 |
notmyname | could be an interesting thing to add to the account quota middleware, though | 18:06 |
*** geaaru has quit IRC | 18:09 | |
timur | timburke_: while trying to setup keystone to test these swiftclient changes, it appears that the openstack client doesn't handle non-ASCII characters in projects names correctly when trying to create a user for such a project... grrr | 18:12 |
timburke_ | that seems... utterly unsurprising. | 18:13 |
*** timburke_ is now known as timburke | 18:14 | |
timburke | so this is kinda weird: i'm seeing some unit test failures locally on stable/queens and stable/pike | 18:14 |
timburke | from queens: http://paste.openstack.org/show/726545/ | 18:15 |
timburke | and pike: http://paste.openstack.org/show/726546/ | 18:16 |
*** gkadam has quit IRC | 18:17 | |
timburke | for the socket error, i guess i oughta backport https://review.openstack.org/#/c/577313/ -- apparently bionic's py27 includes https://github.com/python/cpython/commit/3ef3bcb | 18:27 |
patchbot | patch 577313 - swift - py36: Fix test_get_logger_sysloghandler_plumbing (MERGED) | 18:27 |
notmyname | timur: instead of doing more of the keystone checking (if existing tools don't let us set up the unicode accounts), we could instead leave it until we get a bug report about it | 18:30 |
timburke | huh. the proposal bot seems to be down. or just can't get the feed of proposals | 18:33 |
timburke | p 585506 p 585507 p 585508 | 18:34 |
patchbot | https://review.openstack.org/#/c/585506/ - swift (stable/queens) - py36: Fix test_get_logger_sysloghandler_plumbing | 18:34 |
patchbot | https://review.openstack.org/#/c/585507/ - swift (stable/pike) - Clean up a couple hand-rolled mocks. | 18:34 |
patchbot | https://review.openstack.org/#/c/585508/ - swift (stable/pike) - py36: Fix test_get_logger_sysloghandler_plumbing | 18:34 |
timur | notmyname: well, turns out I can't use openstack client to add a user with a non-ASCII name to the admin role so I can actually test this. I'll address timburke's suggestion of applying this fix only for v1 Auth and punt on keystone | 18:37 |
notmyname | timur: ok | 18:37 |
timur | err, "I can't test this" is what I meant to say | 18:37 |
timur | I did learn how to setup keystone so it's not all wasted effort :) | 18:37 |
timburke | timur: fwiw, i still use https://github.com/openstack/swift3/blob/master/swift3/test/functional/setup_keystone as my how-to-keystone reference... | 18:39 |
timur | timburke: thanks for the pointer! I found this: https://wiki.openstack.org/wiki/Swift/DevstackSetupForKeystoneV3 | 18:42 |
*** itlinux has quit IRC | 18:54 | |
timur | timburke: actually, we don't even need to call parse_header_string() at all, because getheader() is the thing that we fixed! I somehow missed that entirely before | 18:54 |
timburke | even better! | 18:55 |
openstackgerrit | Timur Alperovich proposed openstack/python-swiftclient master: Remove unnecessary calls to parse_header_string(). https://review.openstack.org/585515 | 18:55 |
timur | notmyname timburke ^^^ that's the fixup. I only needed to remove some of the existing calls. Tests passed and I did some functional testing of non-ASCII account names with v1 auth... and gave up on keystone v3 | 18:56 |
*** armaan has joined #openstack-swift | 18:57 | |
*** itlinux has joined #openstack-swift | 19:02 | |
*** itlinux has quit IRC | 19:03 | |
timburke | timur: i'm starting to wonder how much (if at all) we really need to have parse_header_string now... | 19:05 |
timur | good question! I don't want this patch to get any bigger, though | 19:05 |
timur | I think the two spots you found can be removed, though | 19:06 |
timburke | fair enough | 19:06 |
timur | maybe a follow on change after notmyname makes the release? | 19:07 |
openstackgerrit | Merged openstack/swift master: py3: fix up some NameErrors https://review.openstack.org/570478 | 19:22 |
*** armaan has quit IRC | 19:24 | |
*** armaan has joined #openstack-swift | 19:25 | |
*** armaan has quit IRC | 19:29 | |
*** itlinux has joined #openstack-swift | 19:58 | |
notmyname | timburke: what are your thoughts on https://review.openstack.org/#/c/585515/? should I keep waiting on a release for it? | 20:10 |
patchbot | patch 585515 - python-swiftclient - Remove unnecessary calls to parse_header_string(). | 20:10 |
timburke | seems better than what we've got now, and pretty close to what we had before. might still be able to take out more calls to parse_header_string | 20:13 |
timburke | but in general, it shouldn't really be much of an issue following https://review.openstack.org/#/c/571904/ *anyway*... auth tokens and storage urls shouldn't really have non-ascii data... | 20:14 |
patchbot | patch 571904 - swift - tempauth: Send back url-encoded account names (MERGED) | 20:14 |
timburke | calling parse_header_string too many times will definitely cause problems... if you've got a (quoted) url with a %2525 in it, we need to get to *exactly* '%25' and not double-unquote to just '%'... | 20:17 |
*** rcernin_ has joined #openstack-swift | 20:18 | |
*** d0ugal has quit IRC | 20:44 | |
timur | agreed. timburke do we have tests for the quoting behavior? | 21:07 |
timur | if it's broken right now, none of the tests catch this :\ | 21:07 |
*** rcernin_ has quit IRC | 21:30 | |
*** patchbot has quit IRC | 21:32 | |
*** patchbot has joined #openstack-swift | 21:33 | |
*** patchbot has quit IRC | 21:35 | |
*** patchbot has joined #openstack-swift | 21:35 | |
*** itlinux has quit IRC | 21:38 | |
*** mikecmpbll has quit IRC | 22:03 | |
notmyname | did anyone follow up with the probetests fail on master question that came up a week or so ago? | 22:04 |
*** rcernin has joined #openstack-swift | 22:30 | |
timburke | i'm gonna go out on a limb here and say that despite installing ipaddress 1.0.22 (see http://logs.openstack.org/11/427911/32/check/swift-probetests-centos-7/fef8d27/job-output.txt.gz#_2018-07-18_22_53_35_967709) we're still getting system-installed ipaddress 1.0.16 | 22:48 |
timburke | so, most likely introduced with https://review.openstack.org/#/c/569404/ (sorry, that was on my watch) | 22:48 |
patchbot | patch 569404 - swift - IP Range restrictions in temp urls (MERGED) | 22:48 |
timburke | but i'm pretty sure we're fine to lower the required version | 22:49 |
openstackgerrit | Tim Burke proposed openstack/swift master: Lower the required version of ipaddress https://review.openstack.org/585568 | 22:51 |
*** kei_yama has joined #openstack-swift | 23:11 | |
openstackgerrit | Merged openstack/python-swiftclient master: Back out some version bumps https://review.openstack.org/568914 | 23:12 |
mattoliverau | morning | 23:14 |
openstackgerrit | Merged openstack/python-swiftclient master: authors/changelog update for 3.6.0 https://review.openstack.org/584081 | 23:27 |
*** linkmark has quit IRC | 23:57 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!