| *** mvk has joined #openstack-swift | 00:02 | |
| *** hoonetorg has quit IRC | 00:10 | |
| timburke | acoles: we found a bug in zuul! https://storyboard.openstack.org/#!/story/2001496 | 00:27 |
|---|---|---|
| timburke | hmm... i wonder what kind of API storyboard exposes, and how hard it would be to get the openstack bot to write a similar little summary... | 00:29 |
| *** hoonetorg has joined #openstack-swift | 00:30 | |
| *** saint_ has quit IRC | 00:42 | |
| *** tovin07_ has joined #openstack-swift | 00:45 | |
| openstackgerrit | Pete Zaitcev proposed openstack/swift master: PUT+POST and its development test https://review.openstack.org/427911 | 00:59 |
| zaitcev | Oh goodie. I created Content-Length with literally a string value of "None". No wonder nothing worked :-) | 01:02 |
| zaitcev | This should make PUT+POST past functional tests _while disabled_. | 01:03 |
| *** zhurong has joined #openstack-swift | 01:05 | |
| torgomatic | test_failed_delete_keeps_entry in test_expirer.py confuses me. It mocks out a method with "raise Exception('This should not have been called')", then asserts that "This should not have been called" appears in the logs | 01:09 |
| torgomatic | but the only way for it to appear in the logs is if That Which Must Not Be Called was called. | 01:09 |
| zaitcev | a proprietary programmer would've replaced the string with "test" and moved on | 01:10 |
| *** two_tired has joined #openstack-swift | 01:14 | |
| *** hoonetorg has quit IRC | 01:17 | |
| timburke | yeah, that's ... weird ... | 01:18 |
| *** hoonetorg has joined #openstack-swift | 01:21 | |
| *** yashmurty has joined #openstack-swift | 01:53 | |
| yashmurty | @acoles thanks for the directions. @notmyname thank you for your reply. *slogging* still works, I was curios to see it's not in development. | 01:55 |
| yashmurty | I am currently using *slogging* to view the object-usage, etc. for my customers accounts. Is there a better alternative that you'd recommend? @notmyname | 01:58 |
| kota_ | good morning | 01:58 |
| kota_ | acoles: thanks for the reminder. actually I was asleep :P | 01:59 |
| yashmurty | Good morning! :) | 02:00 |
| *** yashmurty has quit IRC | 02:07 | |
| *** yashmurty has joined #openstack-swift | 02:09 | |
| *** zhurong has quit IRC | 02:14 | |
| *** two_tired has quit IRC | 02:14 | |
| *** gkadam has joined #openstack-swift | 02:15 | |
| *** briancline has quit IRC | 02:17 | |
| *** briancline has joined #openstack-swift | 02:18 | |
| openstackgerrit | Gao Fei proposed openstack/swift master: Replace Chinese punctuation with English punctuation https://review.openstack.org/536718 | 02:20 |
| kota_ | yashmurty: o/ | 02:26 |
| *** two_tired has joined #openstack-swift | 02:31 | |
| yashmurty | @+kota_ I noticed you are from Tokyo? Me too... | 02:39 |
| notmyname | yashmurty: TBH, I think the techniques used in slogging are the best way to get usage information from a swift cluster, and no I don't know any other projects that get the same information in the same way | 02:39 |
| notmyname | the only alternatives I know are stuff like openstack's gnocchi. but I don't know much about it | 02:40 |
| yashmurty | @notmyname thanks. As I mentioned, we are using slogging for our billing needs. Do you know how other public cloud providers do it? | 02:48 |
| notmyname | several of them use slogging :-) | 02:48 |
| *** ianychoi has quit IRC | 02:49 | |
| yashmurty | I am quite surprised that slogging is not a part of the official openstack repos! | 02:49 |
| notmyname | it used to be part of swift | 02:49 |
| *** ianychoi has joined #openstack-swift | 02:49 | |
| notmyname | we removed it because we felt it was outside of the scope of an object storage system (and at the time, we wanted more rapid releases of slogging that we had in swift) | 02:50 |
| yashmurty | Oh I see! | 02:51 |
| *** zhurong has joined #openstack-swift | 03:08 | |
| kota_ | yashmurty: yup, I'm at Tokyo. | 03:25 |
| kota_ | (and base on) | 03:25 |
| *** openstackgerrit has quit IRC | 03:33 | |
| yashmurty | @notmyname I see. Thanks for the information. | 03:48 |
| *** kei_yama has quit IRC | 03:52 | |
| yashmurty | I was wondering since Swift is widely used as a Standalone deployment, would it not be better to have *slogging* as part of the main repo again? | 03:53 |
| yashmurty | Also since you mentioned several providers use it in production. | 03:54 |
| *** kei_yama has joined #openstack-swift | 04:02 | |
| zaitcev | We're currently trying to absorb swift3 back. | 04:09 |
| *** links has joined #openstack-swift | 04:09 | |
| zaitcev | Depending how well that works, maybe pulling slogging back would not be far fetched. | 04:09 |
| yashmurty | @zaitcev Thanks! I heard about the swift3 integration in the Sydney summit project update video. | 04:17 |
| kota_ | yashmurty: fyi, that's in progress at https://github.com/openstack/swift/tree/feature/s3api, https://trello.com/b/ZloaZ23t/s3api | 04:21 |
| yashmurty | @kota_ thanks for sharing the URL. I also saw the Trello boards other than s3api. | 04:27 |
| yashmurty | I am wondering how to contribute to those tasks! I have very less experience/knowledge of Swift :'( | 04:30 |
| kota_ | yashmurty: great. basically I mainly control the trello board so any tasks in backlog is available for you :D | 04:31 |
| yashmurty | @kota_ They looks so overwhelming to me! I am clueless about where to begin... :bow: | 04:39 |
| kota_ | yashmurty: hmm... exactly, the easiest tasks has been done. Perhaps, trying the current repo (feature/s3api) and report other remaining tasks may be helpful for us. | 04:42 |
| *** psachin has joined #openstack-swift | 04:42 | |
| kota_ | note that i noticed multi-upload may be broken due to the recent patches in the feature repo. (I'm trying to resolve them with functests) | 04:43 |
| *** gkadam_ has joined #openstack-swift | 04:49 | |
| *** gkadam has quit IRC | 04:50 | |
| *** SkyRocknRoll has joined #openstack-swift | 04:57 | |
| *** two_tired has quit IRC | 05:02 | |
| notmyname | yashmurty: you can do it! we would love to have you contribute to swift | 05:11 |
| *** gkadam_ has quit IRC | 05:18 | |
| *** gkadam has joined #openstack-swift | 05:21 | |
| yashmurty | Thank you for the motivation! @notmyname | 05:27 |
| yashmurty | @kota_ I will try the current repo and get back to you. :) | 05:27 |
| kota_ | yashmurty: thx! | 05:29 |
| *** gkadam has quit IRC | 05:29 | |
| *** gkadam has joined #openstack-swift | 05:30 | |
| yashmurty | @kota_ would you be attending the event : OpenStack OPS Meetup - Tokyo? | 05:32 |
| yashmurty | https://www.eventbrite.com/e/openstack-ops-meetup-tokyo-tickets-39089912982 | 05:32 |
| *** openstackgerrit has joined #openstack-swift | 05:33 | |
| openstackgerrit | Gao Fei proposed openstack/swift master: Replace Chinese punctuation with English punctuation https://review.openstack.org/536718 | 05:33 |
| kota_ | yashmurty: maybe, that's a TBD thing to me. Are you? | 05:34 |
| yashmurty | Yes, I work in the same building as the venue for the event. I would definitely attend it. :) | 05:36 |
| yashmurty | Though it will be my first time to attend Openstack related event. | 05:36 |
| yashmurty | I would also like to attend the Summit's in the future (hopefully Vancouver) ;) | 05:37 |
| *** cshastri has joined #openstack-swift | 05:48 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 06:01 | |
| *** ianychoi has quit IRC | 06:08 | |
| *** ianychoi has joined #openstack-swift | 06:09 | |
| *** zhurong has quit IRC | 06:13 | |
| *** gkadam has quit IRC | 06:13 | |
| *** rcernin has quit IRC | 06:14 | |
| openstackgerrit | Merged openstack/swift master: Reset state every iteration to avoid stale reference https://review.openstack.org/534957 | 06:28 |
| *** threestrands_ has quit IRC | 07:02 | |
| *** hoonetorg has quit IRC | 07:17 | |
| *** hoonetorg has joined #openstack-swift | 07:28 | |
| *** yashmurty has quit IRC | 07:31 | |
| *** yashmurty has joined #openstack-swift | 07:32 | |
| *** hseipp has joined #openstack-swift | 07:41 | |
| *** SkyRocknRoll_ has quit IRC | 07:44 | |
| *** armaan has joined #openstack-swift | 07:47 | |
| *** hseipp has quit IRC | 07:50 | |
| *** pcaruana has joined #openstack-swift | 07:55 | |
| *** YanWei has quit IRC | 08:06 | |
| *** mvk has quit IRC | 08:08 | |
| *** tesseract has joined #openstack-swift | 08:20 | |
| *** gkadam has joined #openstack-swift | 08:28 | |
| *** yashmurty has quit IRC | 08:44 | |
| *** yashmurty has joined #openstack-swift | 08:44 | |
| *** yashmurty has quit IRC | 08:46 | |
| *** geaaru has joined #openstack-swift | 08:47 | |
| *** yashmurty has joined #openstack-swift | 08:48 | |
| *** cbartz has joined #openstack-swift | 08:56 | |
| *** yashmurty has quit IRC | 09:01 | |
| acoles | good morning | 09:11 |
| acoles | timburke: OIC, it was a bug - perhaps the job specs were cached and not getting reset? | 09:12 |
| acoles | yashmurty: if you are interested in getting involved with swift development there are some useful docs here https://docs.openstack.org/swift/latest/#developer-documentation | 09:13 |
| acoles | and many of us will be at this event https://www.openstack.org/ptg/ | 09:14 |
| *** hseipp has joined #openstack-swift | 09:17 | |
| *** itlinux has joined #openstack-swift | 09:31 | |
| *** zhurong_ has joined #openstack-swift | 09:57 | |
| *** itlinux has quit IRC | 09:58 | |
| *** itlinux has joined #openstack-swift | 10:01 | |
| *** bkopilov has quit IRC | 10:05 | |
| *** bkopilov has joined #openstack-swift | 10:06 | |
| *** itlinux has quit IRC | 10:09 | |
| *** hoonetorg has quit IRC | 10:09 | |
| *** tovin07_ has quit IRC | 10:12 | |
| *** hoonetorg has joined #openstack-swift | 10:22 | |
| *** cshastri has quit IRC | 10:34 | |
| *** cshastri has joined #openstack-swift | 10:35 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift feature/s3api: Fix s3api middleware to work with functests https://review.openstack.org/537838 | 10:42 |
| kota_ | tdasilva: could you please check out p 537838 ? | 10:44 |
| patchbot | https://review.openstack.org/#/c/537838/ - swift (feature/s3api) - Fix s3api middleware to work with functests | 10:44 |
| kota_ | tdasilva: that works at least in my local. I have to figure out how the functests gate works yet but i think it's close to fix. | 10:45 |
| kota_ | timburke: ^^^ you may be interested in. | 10:45 |
| *** jistr is now known as jistr|mtg | 10:49 | |
| * kota_ is waiting zuul runs the functional tests. | 10:50 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: Refactor expirer unit tests https://review.openstack.org/537841 | 10:51 |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: Refactor expirer unit tests https://review.openstack.org/537841 | 10:58 |
| *** SkyRocknRoll has quit IRC | 12:09 | |
| *** kei_yama has quit IRC | 12:09 | |
| *** SkyRocknRoll has joined #openstack-swift | 12:13 | |
| *** tesseract has quit IRC | 12:24 | |
| *** jistr|mtg is now known as jistr | 12:38 | |
| *** Supun has joined #openstack-swift | 12:45 | |
| *** Supun has quit IRC | 12:51 | |
| *** Supun has joined #openstack-swift | 12:56 | |
| *** Supun has quit IRC | 13:01 | |
| *** Supun has joined #openstack-swift | 13:02 | |
| *** Supun has quit IRC | 13:08 | |
| *** Supun has joined #openstack-swift | 13:09 | |
| *** hseipp has quit IRC | 13:10 | |
| *** Supun has quit IRC | 13:14 | |
| *** Supun has joined #openstack-swift | 13:15 | |
| *** armaan has quit IRC | 13:24 | |
| *** tesseract has joined #openstack-swift | 13:27 | |
| *** armaan has joined #openstack-swift | 13:39 | |
| *** tesseract has quit IRC | 13:42 | |
| *** tesseract has joined #openstack-swift | 13:42 | |
| *** zhurong_ has quit IRC | 13:44 | |
| *** psachin has quit IRC | 14:02 | |
| *** tesseract has quit IRC | 14:17 | |
| *** tesseract has joined #openstack-swift | 14:18 | |
| -openstackstatus- NOTICE: We're currently experiencing issues with the logs.openstack.org server which will result in POST_FAILURE for jobs, please stand by and don't needlessly recheck jobs while we troubleshoot the problem. | 14:26 | |
| *** psachin has joined #openstack-swift | 14:27 | |
| *** psachin has quit IRC | 14:32 | |
| *** psachin has joined #openstack-swift | 14:33 | |
| *** cshastri has quit IRC | 14:36 | |
| *** cshastri has joined #openstack-swift | 14:37 | |
| *** pcaruana has quit IRC | 15:20 | |
| *** itlinux has joined #openstack-swift | 15:27 | |
| *** d0ugal has quit IRC | 15:31 | |
| *** geaaru has quit IRC | 15:37 | |
| *** d0ugal has joined #openstack-swift | 15:41 | |
| -openstackstatus- NOTICE: logs.openstack.org is stabilized and there should no longer be *new* POST_FAILURE errors. Logs for jobs that ran in the past weeks until earlier today are currently unavailable pending FSCK completion. We're going to temporarily disable *successful* jobs from uploading their logs to reduce strain on our current limited capacity. Thanks for your patience ! | 16:02 | |
| *** silor has joined #openstack-swift | 16:31 | |
| *** openstackstatus has quit IRC | 16:41 | |
| *** linkmark has joined #openstack-swift | 16:43 | |
| *** openstackstatus has joined #openstack-swift | 16:43 | |
| *** ChanServ sets mode: +v openstackstatus | 16:43 | |
| *** links has quit IRC | 16:57 | |
| *** tesseract has quit IRC | 16:58 | |
| *** gkadam has quit IRC | 17:06 | |
| *** cbartz has quit IRC | 17:07 | |
| *** jistr is now known as jistr|conf | 17:13 | |
| *** saint_ has joined #openstack-swift | 17:24 | |
| *** silor has quit IRC | 17:26 | |
| timburke | good morning | 17:34 |
| *** saint_ has quit IRC | 17:36 | |
| *** itlinux has quit IRC | 17:38 | |
| *** cshastri has quit IRC | 17:42 | |
| *** jistr|conf is now known as jistr | 18:14 | |
| clayg | morning | 18:26 |
| clayg | I don't really know what to think about lp bug #1745245 | 18:46 |
| openstack | Launchpad bug 1745245 in OpenStack Object Storage (swift) "Objects stored on PUT even when etags don't match on object-servers" [High,New] https://launchpad.net/bugs/1745245 | 18:46 |
| timburke | yeah, i don't know that the replicators can resolve that situation... | 18:47 |
| clayg | notmyname set the priority to "high" - but it's hardly a new behavior... and I don't have any new ideas to address it? Objects with the same timestamp and different contents are not great for sure... | 18:47 |
| timburke | there are some solutions -- "run with encryption" for example ;-) | 18:48 |
| clayg | I guess it'd be pretty swift if we proxy could send an etag in footers with every PUT to all the object servers | 18:48 |
| clayg | zaitcev's PUT+POST might have a path to a generic solution | 18:49 |
| clayg | i.e. EC probably already has to do this correctly | 18:50 |
| clayg | but of course if the client doesn't send an ETag there's only so much we can do | 18:50 |
| timburke | or if it doesn't, it *should* | 18:51 |
| *** armaan has quit IRC | 18:51 | |
| zaitcev | I don't understand. I thought the object server computed the etag independently and sent it in the headers of the response. In case of PUT+POST(+POST) it's sent in the reply to the POST, but it's okay, because the object is not in the namespace after just PUT. | 18:51 |
| zaitcev | I think currently I made so if proxy _supplies_ Etag, it gets overridden by the computed Etag. | 18:52 |
| zaitcev | Hmm. Actually I thought I copied the old implementation, but maybe it's trickier. | 18:52 |
| clayg | it's only unencrypted replicated where the object server would even be *able* to compute the ETag - in all other cases only the proxy has the raw bytes from the client - so we should at least not be getting two answers | 18:53 |
| clayg | and for the unencrypted replicated case if we let the proxy do the md5 and send it after the body - what else could the object-server do if it notices a mis-match besides quarantine/unlink? | 18:54 |
| timburke | good! we *do* handle proxy <-> object data validation for EC! https://github.com/openstack/swift/blob/2.16.0/swift/proxy/controllers/obj.py#L2573-L2574 | 18:55 |
| clayg | timburke: yeah makes sense | 18:56 |
| *** psachin has quit IRC | 19:02 | |
| timburke | i think only for 2.9.0+ though -- pretty sure it came in with encryption, in https://review.openstack.org/#/c/328204/ | 19:03 |
| patchbot | patch 328204 - swift (feature/crypto-review) - Support for http footers - Replication and EC (MERGED) | 19:03 |
| timburke | oh, maybe not... maybe it was earlier... | 19:03 |
| timburke | all right, cool. it was there from the beginning for EC, in https://review.openstack.org/#/c/169989/ | 19:06 |
| patchbot | patch 169989 - swift (feature/ec_review) - Foundational support for PUT and GET of erasure-co... (MERGED) | 19:06 |
| *** hoonetorg has quit IRC | 19:07 | |
| timburke | know what i'd love to do? write an itty-bitty dns server so we can use p 435768 to add (all in-process!?) func tests to cover p 435771 | 19:17 |
| patchbot | https://review.openstack.org/#/c/435768/ - swift - Allow to configure the nameservers in cname_lookup (MERGED) | 19:17 |
| patchbot | https://review.openstack.org/#/c/435771/ - swift - Rewrite redirection in cname_lookup & domain_remap (MERGED) | 19:17 |
| clayg | *ew* | 19:18 |
| timburke | (i always feel like i'm winning when i elicit that reaction from clayg. like with p 533028 :P) | 19:19 |
| patchbot | https://review.openstack.org/#/c/533028/ - swift - Add some functional CORS tests | 19:19 |
| clayg | #winningatlife | 19:20 |
| openstackgerrit | Merged openstack/swift master: Add a small sleep when trying to predict X-Timestamp https://review.openstack.org/528865 | 19:26 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Don't lose objects when db is rsync replicated during sharding https://review.openstack.org/536943 | 19:31 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Remove special case for mismatched last cleaved pointer https://review.openstack.org/538009 | 19:31 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: make get_own_shard_range() always return something https://review.openstack.org/538010 | 19:31 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep: Make find_shard_ranges cope with misplaced objects https://review.openstack.org/538011 | 19:31 |
| *** onovy has quit IRC | 19:45 | |
| *** peterlisak has quit IRC | 19:45 | |
| *** peterlisak has joined #openstack-swift | 19:51 | |
| *** onovy has joined #openstack-swift | 19:58 | |
| *** Supun has quit IRC | 19:58 | |
| openstackgerrit | Samuel Merritt proposed openstack/swift master: Augment object-expirer's StatsD buckets. https://review.openstack.org/538016 | 20:06 |
| torgomatic | Is there an openstack release coming up soon? The gate seems more clogged than usual. | 20:11 |
| timburke | torgomatic: yup. i think today's the freeze for client libs? | 20:12 |
| torgomatic | That explains it, then. | 20:12 |
| timburke | so it kinda sucks that https://review.openstack.org/#/c/536633/ hasn't landed yet, despite being approved two days ago | 20:13 |
| patchbot | patch 536633 - python-swiftclient - authors/changelog updates for 3.5.0 release | 20:13 |
| *** hoonetorg has joined #openstack-swift | 20:14 | |
| *** Jeffrey4l has quit IRC | 20:19 | |
| *** flwang has quit IRC | 20:23 | |
| *** Jeffrey4l has joined #openstack-swift | 20:24 | |
| notmyname | good morning | 20:30 |
| *** flwang has joined #openstack-swift | 20:36 | |
| notmyname | I put https://bugs.launchpad.net/swift/+bug/1745245 as high because it has the ability to corrupt data in the cluster. thanks for the comment on it clayg. I updated the bug fields | 20:40 |
| openstack | Launchpad bug 1745245 in OpenStack Object Storage (swift) "Objects stored on PUT even when etags don't match on object-servers" [Medium,Confirmed] | 20:40 |
| *** mvk has joined #openstack-swift | 20:55 | |
| *** Jeffrey4l has quit IRC | 21:02 | |
| *** Jeffrey4l has joined #openstack-swift | 21:14 | |
| *** armaan has joined #openstack-swift | 21:25 | |
| torgomatic | apparently the gate is broken because of someone logging from a non-main thread in an Eventlet context: http://lists.openstack.org/pipermail/openstack-dev/2018-January/126586.html | 21:36 |
| torgomatic | would anyone smarter than I like to take a shot at https://github.com/eventlet/eventlet/issues/432 ? | 21:37 |
| notmyname | torgomatic: because another project is? or because swiftclient is? or because someone else has configured swiftclient to log in the wrong way? | 21:52 |
| torgomatic | notmyname: someone was calling swiftclient from a background thread, and swiftclient was logging, and log handlers have mutexes, and those mutexes break when used from background threads | 21:53 |
| torgomatic | not swiftclient's fault, just another sad story in the book of sad stories from that bug | 21:53 |
| notmyname | anything spring to mind that we can do to help the situation? | 21:53 |
| notmyname | FWIW, we havne't done a swift client release because it failed the post jobs after being in the gate for 24 hours. timburke rechecked it again this morning to start over... | 21:54 |
| torgomatic | I'm trying to recruit someone who could fix it. I took a swing at it, and wound up with some kind of awful code soup that was maliciously nondeterministic | 21:54 |
| torgomatic | I thought the code change I made would work, but it was like things were intelligent and actively fighting my attempts at fixing them | 21:56 |
| torgomatic | and then I had to go do other things instead, because company-internal reasons | 21:56 |
| notmyname | torgomatic: thanks | 21:59 |
| *** linkmark has quit IRC | 22:00 | |
| notmyname | I have to step out for a while. i'll check back in a few hours | 22:02 |
| openstackgerrit | Merged openstack/swift feature/deep: Remove func-post-as-copy tox env (again) https://review.openstack.org/536517 | 22:20 |
| *** itlinux has joined #openstack-swift | 22:37 | |
| *** kei_yama has joined #openstack-swift | 22:47 | |
| *** mvk has quit IRC | 23:00 | |
| *** itlinux has quit IRC | 23:04 | |
| *** itlinux has joined #openstack-swift | 23:06 | |
| *** itlinux has quit IRC | 23:11 | |
| *** ianychoi has quit IRC | 23:26 | |
| *** ianychoi has joined #openstack-swift | 23:27 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!