*** sorrison has quit IRC | 01:42 | |
*** sorrison has joined #openstack-swift | 01:44 | |
*** paladox has quit IRC | 02:01 | |
*** paladox has joined #openstack-swift | 02:02 | |
*** diablo_rojo has quit IRC | 02:14 | |
*** gyee has quit IRC | 02:25 | |
*** dasp has quit IRC | 04:41 | |
*** dasp has joined #openstack-swift | 04:41 | |
*** pcaruana has joined #openstack-swift | 06:01 | |
*** sorrison has quit IRC | 06:39 | |
*** sorrison has joined #openstack-swift | 06:52 | |
*** sorrison has quit IRC | 07:06 | |
*** rcernin has quit IRC | 07:09 | |
*** sorrison has joined #openstack-swift | 07:09 | |
*** sorrison has quit IRC | 07:14 | |
*** sorrison has joined #openstack-swift | 07:27 | |
*** sorrison has quit IRC | 07:56 | |
*** sorrison has joined #openstack-swift | 07:59 | |
*** renich has quit IRC | 07:59 | |
*** tesseract has joined #openstack-swift | 08:00 | |
*** tkajinam has quit IRC | 08:17 | |
*** spotz has quit IRC | 08:23 | |
*** dabukalam has quit IRC | 08:23 | |
*** timur has quit IRC | 08:23 | |
*** tesseract has quit IRC | 08:24 | |
*** pcaruana has quit IRC | 08:24 | |
*** Jeffrey4l has quit IRC | 08:24 | |
*** hoonetorg has quit IRC | 08:24 | |
*** irclogbot_3 has quit IRC | 08:24 | |
*** benj_ has quit IRC | 08:24 | |
*** corvus has quit IRC | 08:24 | |
*** edausq has quit IRC | 08:24 | |
*** dosaboy has quit IRC | 08:24 | |
*** szaher has quit IRC | 08:24 | |
*** MooingLemur has quit IRC | 08:24 | |
*** sorrison has quit IRC | 08:24 | |
*** dasp has quit IRC | 08:24 | |
*** baojg has quit IRC | 08:24 | |
*** mugsie has quit IRC | 08:24 | |
*** donnyd has quit IRC | 08:24 | |
*** ccamel has quit IRC | 08:24 | |
*** ormandj has quit IRC | 08:24 | |
*** beisner has quit IRC | 08:24 | |
*** rpittau|afk has quit IRC | 08:24 | |
*** gregwork has quit IRC | 08:24 | |
*** toanster has quit IRC | 08:24 | |
*** rledisez has quit IRC | 08:24 | |
*** godog has quit IRC | 08:24 | |
*** seongsoocho has quit IRC | 08:24 | |
*** abelur has quit IRC | 08:24 | |
*** d34dh0r53 has quit IRC | 08:24 | |
*** openstackgerrit has quit IRC | 08:24 | |
*** hugokuo has quit IRC | 08:24 | |
*** evrardjp has quit IRC | 08:24 | |
*** StevenK has quit IRC | 08:24 | |
*** manuvakery has quit IRC | 08:24 | |
*** onovy has quit IRC | 08:24 | |
*** kota_ has quit IRC | 08:24 | |
*** Anticimex has quit IRC | 08:24 | |
*** coreycb has quit IRC | 08:24 | |
*** clayg has quit IRC | 08:24 | |
*** nottrobin has quit IRC | 08:24 | |
*** persia_ has quit IRC | 08:24 | |
*** joeljwright has quit IRC | 08:24 | |
*** fungi has quit IRC | 08:24 | |
*** gmann has quit IRC | 08:24 | |
*** paladox has quit IRC | 08:24 | |
*** jistr has quit IRC | 08:24 | |
*** mvkr has quit IRC | 08:24 | |
*** patchbot has quit IRC | 08:24 | |
*** tonyb has quit IRC | 08:24 | |
*** kukacz_ has quit IRC | 08:24 | |
*** aluria has quit IRC | 08:24 | |
*** rickflare has quit IRC | 08:24 | |
*** dcourtoi has quit IRC | 08:24 | |
*** cschwede has quit IRC | 08:24 | |
*** timburke has quit IRC | 08:24 | |
*** tristanC has quit IRC | 08:24 | |
*** mathiasb has quit IRC | 08:24 | |
*** mattoliverau has quit IRC | 08:24 | |
*** noonedeadpunk has quit IRC | 08:24 | |
*** baffle has quit IRC | 08:24 | |
*** viks___ has quit IRC | 08:24 | |
*** csmart has quit IRC | 08:24 | |
*** f0o has quit IRC | 08:24 | |
*** jrosser has quit IRC | 08:24 | |
*** tdasilva has quit IRC | 08:24 | |
*** mgagne has quit IRC | 08:24 | |
*** lifeless has quit IRC | 08:24 | |
*** mahatic has quit IRC | 08:24 | |
*** early has quit IRC | 08:24 | |
*** ab-a has quit IRC | 08:24 | |
*** DHE has quit IRC | 08:24 | |
*** alecuyer has quit IRC | 08:24 | |
*** cjloader has quit IRC | 08:24 | |
*** ChanServ has quit IRC | 08:24 | |
*** tesseract has joined #openstack-swift | 08:24 | |
*** sorrison has joined #openstack-swift | 08:24 | |
*** pcaruana has joined #openstack-swift | 08:24 | |
*** dasp has joined #openstack-swift | 08:24 | |
*** paladox has joined #openstack-swift | 08:24 | |
*** ormandj has joined #openstack-swift | 08:24 | |
*** baojg has joined #openstack-swift | 08:24 | |
*** jistr has joined #openstack-swift | 08:24 | |
*** f0o has joined #openstack-swift | 08:24 | |
*** Jeffrey4l has joined #openstack-swift | 08:24 | |
*** StevenK has joined #openstack-swift | 08:24 | |
*** mugsie has joined #openstack-swift | 08:24 | |
*** persia_ has joined #openstack-swift | 08:24 | |
*** jrosser has joined #openstack-swift | 08:24 | |
*** donnyd has joined #openstack-swift | 08:24 | |
*** ccamel has joined #openstack-swift | 08:24 | |
*** abelur has joined #openstack-swift | 08:24 | |
*** godog has joined #openstack-swift | 08:24 | |
*** rledisez has joined #openstack-swift | 08:24 | |
*** seongsoocho has joined #openstack-swift | 08:24 | |
*** toanster has joined #openstack-swift | 08:24 | |
*** gregwork has joined #openstack-swift | 08:24 | |
*** rpittau has joined #openstack-swift | 08:24 | |
*** beisner has joined #openstack-swift | 08:24 | |
*** timur has joined #openstack-swift | 08:24 | |
*** spotz has joined #openstack-swift | 08:24 | |
*** dabukalam has joined #openstack-swift | 08:24 | |
*** mvkr has joined #openstack-swift | 08:24 | |
*** hoonetorg has joined #openstack-swift | 08:24 | |
*** baffle has joined #openstack-swift | 08:24 | |
*** csmart has joined #openstack-swift | 08:24 | |
*** joeljwright has joined #openstack-swift | 08:24 | |
*** openstackgerrit has joined #openstack-swift | 08:24 | |
*** d34dh0r53 has joined #openstack-swift | 08:24 | |
*** ab-a has joined #openstack-swift | 08:24 | |
*** patchbot has joined #openstack-swift | 08:24 | |
*** tonyb has joined #openstack-swift | 08:24 | |
*** irclogbot_3 has joined #openstack-swift | 08:24 | |
*** benj_ has joined #openstack-swift | 08:24 | |
*** corvus has joined #openstack-swift | 08:24 | |
*** edausq has joined #openstack-swift | 08:24 | |
*** tdasilva has joined #openstack-swift | 08:24 | |
*** mgagne has joined #openstack-swift | 08:24 | |
*** orwell.freenode.net sets mode: +vvv rledisez joeljwright tdasilva | 08:24 | |
*** kukacz_ has joined #openstack-swift | 08:24 | |
*** manuvakery has joined #openstack-swift | 08:24 | |
*** lifeless has joined #openstack-swift | 08:24 | |
*** mahatic has joined #openstack-swift | 08:24 | |
*** onovy has joined #openstack-swift | 08:24 | |
*** early has joined #openstack-swift | 08:24 | |
*** rickflare has joined #openstack-swift | 08:24 | |
*** evrardjp has joined #openstack-swift | 08:24 | |
*** hugokuo has joined #openstack-swift | 08:24 | |
*** nottrobin has joined #openstack-swift | 08:24 | |
*** clayg has joined #openstack-swift | 08:24 | |
*** coreycb has joined #openstack-swift | 08:24 | |
*** Anticimex has joined #openstack-swift | 08:24 | |
*** kota_ has joined #openstack-swift | 08:24 | |
*** aluria has joined #openstack-swift | 08:24 | |
*** dosaboy has joined #openstack-swift | 08:24 | |
*** viks___ has joined #openstack-swift | 08:24 | |
*** fungi has joined #openstack-swift | 08:24 | |
*** dcourtoi has joined #openstack-swift | 08:24 | |
*** szaher has joined #openstack-swift | 08:24 | |
*** DHE has joined #openstack-swift | 08:24 | |
*** MooingLemur has joined #openstack-swift | 08:24 | |
*** mathiasb has joined #openstack-swift | 08:24 | |
*** cschwede has joined #openstack-swift | 08:24 | |
*** orwell.freenode.net sets mode: +vvvv mahatic clayg kota_ cschwede | 08:24 | |
*** alecuyer has joined #openstack-swift | 08:24 | |
*** timburke has joined #openstack-swift | 08:24 | |
*** tristanC has joined #openstack-swift | 08:24 | |
*** noonedeadpunk has joined #openstack-swift | 08:24 | |
*** cjloader has joined #openstack-swift | 08:24 | |
*** gmann has joined #openstack-swift | 08:24 | |
*** ChanServ has joined #openstack-swift | 08:24 | |
*** mattoliverau has joined #openstack-swift | 08:24 | |
*** orwell.freenode.net sets mode: +vov timburke ChanServ mattoliverau | 08:24 | |
*** rcernin has joined #openstack-swift | 08:51 | |
*** rdejoux has joined #openstack-swift | 08:54 | |
*** ccamacho has joined #openstack-swift | 09:08 | |
*** rcernin has quit IRC | 09:53 | |
*** rpittau is now known as rpittau|bbl | 11:27 | |
*** rpittau|bbl is now known as rpittau | 13:36 | |
*** spotz has quit IRC | 14:04 | |
*** dabukalam has quit IRC | 14:04 | |
*** timur has quit IRC | 14:04 | |
*** dabukalam has joined #openstack-swift | 14:09 | |
*** spotz has joined #openstack-swift | 14:09 | |
*** timur has joined #openstack-swift | 14:09 | |
*** efried has joined #openstack-swift | 14:13 | |
efried | o/ swift! | 14:13 |
---|---|---|
efried | qq: if $user creates an object, is it possible (to set policy) for admin to delete it? | 14:13 |
*** spotz has quit IRC | 14:13 | |
*** dabukalam has quit IRC | 14:13 | |
*** timur has quit IRC | 14:13 | |
efried | on the theory that, if $user vanishes from existence, admin needs to be able to clean up. | 14:14 |
*** dabukalam has joined #openstack-swift | 14:19 | |
*** spotz has joined #openstack-swift | 14:19 | |
*** timur has joined #openstack-swift | 14:19 | |
*** dasp has quit IRC | 15:02 | |
*** dasp has joined #openstack-swift | 15:03 | |
clayg | tdasilva: yes, I think 400 on put=null is very reasonable! | 15:34 |
clayg | yeah "disable versioning for this PUT" would be pretty wild | 15:34 |
clayg | efried: I think the "policy" is that the admin would have all rights always | 15:35 |
efried | clayg: So they would even be able to retrieve the object? | 15:36 |
clayg | yes | 15:36 |
efried | and there's not a way to customize that to a more granular level? | 15:37 |
efried | So e.g. the admin can't GET, but can DELETE? | 15:37 |
clayg | efried: no, not really, not for the admin | 15:42 |
clayg | efried: you could use container acl's to grant another user write access w/o read access | 15:43 |
efried | "another user" including the admin? | 15:44 |
efried | sorry, I think you already maybe answered that. | 15:44 |
efried | For my use case, it's not crucial, but would be nice. | 15:44 |
clayg | sorry, admin is intended to have elevated access levels | 15:45 |
clayg | it sounds like it'd be easier in your situation to de-escalate access rather than grant access - but the existing authorization schema doesn't work like that, sorry | 15:46 |
efried | okay, thanks for the help clayg. | 15:49 |
clayg | there was some work a LONG time ago that tried to integrate more granualar keystone RBAC https://review.opendev.org/#/c/149930/ | 15:50 |
patchbot | patch 149930 - swift - WIP: Enable Role-based access control using oslo.p... (ABANDONED) - 15 patch sets | 15:50 |
efried | mm, noted. | 15:51 |
ormandj | so on the ratelimit question from yesterday, i fudged ratelimit.py to force it to process even if the container size was returned as 0 and it appears to function - our real mystery is now why ontainer size is always being returned as 0 | 16:03 |
ormandj | that doesn't seem... right | 16:03 |
*** gyee has joined #openstack-swift | 16:09 | |
clayg | yeah that's strange - if you head the container it's not zero? | 16:14 |
ormandj | clayg: checking. sorry adding a ton of debug code to the base proxy controller | 16:29 |
ormandj | clayg: yes, sure: | 16:58 |
ormandj | HEAD call to https://LOCATION/v1/AUTH_USERID/ormandj used request id txBLAHBLAH-005de7e588 | 16:59 |
ormandj | +--------------+---------------------------------------+ | 16:59 |
ormandj | | Field | Value | | 16:59 |
ormandj | +--------------+---------------------------------------+ | 16:59 |
ormandj | | account | AUTH_USERID | | 16:59 |
ormandj | | bytes_used | 5000 | | 16:59 |
ormandj | | container | ormandj | | 16:59 |
ormandj | | object_count | 1000 | | 16:59 |
ormandj | +--------------+---------------------------------------+ | 16:59 |
ormandj | let me make a snippit showing some of the log changes we've made and the output | 17:00 |
ormandj | clayg: https://privatebin.net/?b36956193401b654#Dip2amPeE4g43aBLKWQfQWm1W4DvmFvCxogeTXNohRik | 17:07 |
ormandj | you'll see where i modified get_maxrate to accept 0 as valid, which makes it 'work', but previously, did not have that or | 17:07 |
ormandj | but you can see from the log output the container size is being returned as 0, which is incorrect | 17:08 |
ormandj | (stock file is "if size:" | 17:09 |
ormandj | ) | 17:09 |
*** rdejoux has quit IRC | 17:10 | |
*** persia_ is now known as persia | 17:16 | |
ormandj | clayg: something else interesting, if I issue a DELETE, rate limiting works/i see the correct container size | 17:19 |
ormandj | when i issue PUTs, not so much | 17:20 |
*** diablo_rojo has joined #openstack-swift | 17:31 | |
ormandj | clayg: even more digging, it looks like the account name that's passed in (we're using s3api) is the ec2 access key with PUTs, but the account name that's passed in for DELETE/etc is the actual account name AUTH_blah | 17:32 |
clayg | ormandj: oh dude! I remmeber something about reatelimit and s3api being not working 🤔 | 17:39 |
openstackgerrit | Tim Burke proposed openstack/swift stable/train: Fix BadResponseLength error when creating symlink https://review.opendev.org/697333 | 17:41 |
clayg | ormandj: i found some internal refernece to https://bugs.launchpad.net/swift/+bug/1669888 | 17:42 |
openstack | Launchpad bug 1669888 in OpenStack Object Storage (swift) "ratelimit doesn't filter requests very well" [Low,Fix released] | 17:42 |
clayg | ormandj: so you're probably on the right track with with the /path leading to wrong stats - no idea why DELETE seems to work? | 17:43 |
clayg | ormandj: if you haven't already you might add loggig in get_container_size that looks at the env and the returned container_info | 17:45 |
*** rpittau is now known as rpittau|afk | 17:50 | |
timburke | clayg, ormandj: ratelimit's trouble with s3api is that RL sees the s3-style request, not the swift one. so it doesn't even look like a valid swift path; the get_container_info call almost certainly 404s | 17:56 |
*** tesseract has quit IRC | 18:04 | |
ormandj | timburke: let me show you some more info re: debug output | 18:38 |
ormandj | DELETE/etc seem to work fine, it's only PUTs where we see the container info fail | 18:38 |
ormandj | we're on Stein so I don't think this one applies to us (the earlier bug report) - we have 2.21 and fix is in 2.20 | 18:39 |
ormandj | clayg: timburke: https://privatebin.net/?8ff8d0e831fe4ceb#GzHuaSJEHX4iRWANbprPb5HWszpBC7jfmQPMfv2JMX15 | 18:45 |
ormandj | the same debug stuff from the previous paste re: where debug output is generated should cover that | 18:45 |
ormandj | you'll see in the PUT the account_name that gets parsed is the ec2 account key | 18:46 |
ormandj | but in the DELETE the account_name that gets parsed is correct | 18:46 |
ormandj | (our dev environment so i quit bothering to obfuscate account info hah) | 18:46 |
ormandj | we can't put ratelimit before s3api in the pipeline because of the s3 style vhost stuff goofing up the handling of requests | 18:48 |
timburke | so RL is currently between s3api and auth? could try putting it after auth... by then, all the path translations should be done... but if you're using some external auth service like keystone it'll likely see an uptick in traffic | 19:07 |
*** baojg has quit IRC | 19:07 | |
ormandj | pipeline = catch_errors gatekeeper healthcheck proxy-logging cache listing_formats authtoken s3api ratelimit container_sync tempurl s3token keystoneauth copy container-quotas account-quotas slo dlo versioned_writes symlink bulk keymaster encryption proxy-logging ceilometer proxy-server | 19:11 |
ormandj | timburke: i'm just not understanding why a DELETE/etc work, but not a PUT :) | 19:11 |
ormandj | we can try moving ratelimit after s3token and see if that works, but not sure why DELETE would work with this but not PUT in this case. | 19:17 |
openstackgerrit | Tim Burke proposed openstack/swift stable/train: Update known-failures and config for up-rev'ed ceph/s3tests https://review.opendev.org/697349 | 19:20 |
openstackgerrit | Tim Burke proposed openstack/swift stable/stein: Update known-failures and config for up-rev'ed ceph/s3tests https://review.opendev.org/697350 | 19:20 |
*** diablo_rojo has quit IRC | 19:30 | |
*** peluse has joined #openstack-swift | 19:55 | |
kota_ | good morning | 20:58 |
seongsoocho | good morning o/ | 20:58 |
kota_ | seongsoocho: ¥o/ | 20:58 |
kota_ | seongsoocho: is it early morning for you, too? | 20:59 |
mattoliverau | o/ | 20:59 |
kota_ | hi mattoliverau | 20:59 |
timburke | meeting time! | 21:00 |
seongsoocho | yes. you and I are same timezone. It's 6am here too kota_ | 21:00 |
kota_ | :D | 21:00 |
clayg | 🎉 | 21:03 |
openstackgerrit | Merged openstack/swift master: Turn off dns_compliant_bucket_names for ceph tests https://review.opendev.org/695744 | 21:04 |
openstackgerrit | Merged openstack/swift stable/train: Switch py2 DSVM jobs to only run swift under py2 https://review.opendev.org/695596 | 21:04 |
openstackgerrit | Merged openstack/swift stable/stein: Switch py2 DSVM jobs to only run swift under py2 https://review.opendev.org/695597 | 21:04 |
*** diablo_rojo has joined #openstack-swift | 21:11 | |
*** pcaruana has quit IRC | 21:32 | |
seongsoocho | yey , It's time to get ready to go to work. | 21:42 |
timburke | thanks for coming seongsoocho! sorry it's so early... | 21:42 |
seongsoocho | you're welcome. It is not too early to me. that's fine. | 21:43 |
*** rcernin has joined #openstack-swift | 22:19 | |
*** tkajinam has joined #openstack-swift | 23:05 | |
openstackgerrit | Tim Burke proposed openstack/swift stable/train: py3: Fix s3api header casing https://review.opendev.org/697384 | 23:10 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!