| *** NM has quit IRC | 00:14 | |
| *** links has joined #openstack-swift | 02:17 | |
| kota_ | hello | 02:30 |
|---|---|---|
| kota_ | oic | 02:32 |
| kota_ | i wonder why the patch 566421 can modify swift golang code at the master but i noticed that depends on the patch 560480 adding golang code for losf | 02:34 |
| patchbot | https://review.openstack.org/#/c/566421/ - swift - Cleanups for LOSF RPC server | 02:34 |
| patchbot | https://review.openstack.org/#/c/560480/ - swift - Initial LOSF RPC commit | 02:34 |
| kota_ | that makes me sense. | 02:34 |
| *** psachin has joined #openstack-swift | 02:42 | |
| *** mingyu has joined #openstack-swift | 03:28 | |
| *** mingyu has quit IRC | 03:48 | |
| *** cshastri has joined #openstack-swift | 03:57 | |
| *** threestrands has joined #openstack-swift | 04:04 | |
| *** gyankum has joined #openstack-swift | 04:10 | |
| *** SkyRocknRoll has joined #openstack-swift | 04:37 | |
| *** SkyRocknRoll has quit IRC | 04:38 | |
| *** silor has joined #openstack-swift | 04:47 | |
| *** hoonetorg has quit IRC | 05:07 | |
| *** hoonetorg has joined #openstack-swift | 05:20 | |
| *** mingyu has joined #openstack-swift | 05:57 | |
| *** pcaruana has joined #openstack-swift | 06:02 | |
| *** geaaru has quit IRC | 06:06 | |
| *** armaan has quit IRC | 06:34 | |
| *** armaan has joined #openstack-swift | 06:35 | |
| *** armaan has quit IRC | 06:42 | |
| *** armaan has joined #openstack-swift | 06:43 | |
| *** armaan has quit IRC | 06:58 | |
| *** armaan has joined #openstack-swift | 06:58 | |
| *** hseipp has joined #openstack-swift | 07:16 | |
| *** thegrif has quit IRC | 07:23 | |
| *** silor has quit IRC | 07:28 | |
| *** silor has joined #openstack-swift | 07:34 | |
| *** threestrands has quit IRC | 07:39 | |
| *** ntt has quit IRC | 07:43 | |
| *** ntt has joined #openstack-swift | 07:45 | |
| *** geaaru has joined #openstack-swift | 08:03 | |
| *** timss- is now known as timss | 08:29 | |
| *** timss has quit IRC | 08:32 | |
| *** timss has joined #openstack-swift | 08:33 | |
| *** armaan has quit IRC | 09:38 | |
| *** armaan has joined #openstack-swift | 09:40 | |
| *** silor1 has joined #openstack-swift | 09:57 | |
| *** silor has quit IRC | 09:59 | |
| *** silor1 is now known as silor | 09:59 | |
| *** afazekas is now known as afazekas|pto | 09:59 | |
| kota_ | swift-cores: sorry, i couldn't manage my time to review at the deep-review, probably I will have that tomorrow. | 10:08 |
| *** armaan has quit IRC | 10:45 | |
| *** armaan has joined #openstack-swift | 10:45 | |
| openstackgerrit | Hajime Kondo proposed openstack/slogging master: First commit of devstack directory https://review.openstack.org/564708 | 10:54 |
| *** guimaluf has quit IRC | 11:15 | |
| *** kei_yama has quit IRC | 12:01 | |
| *** hseipp has quit IRC | 12:02 | |
| *** NM has joined #openstack-swift | 12:30 | |
| *** cshastri has quit IRC | 12:46 | |
| *** vinsh has joined #openstack-swift | 12:57 | |
| *** psachin` has joined #openstack-swift | 13:04 | |
| *** links has quit IRC | 13:05 | |
| *** remix_tj has joined #openstack-swift | 13:06 | |
| remix_tj | hello! I'm looking for some information on how to write a middleware for filtering some operations on some files. Where is the most appropriate place to insert the middleware? Proxy Server? | 13:07 |
| *** zigo_ has joined #openstack-swift | 13:07 | |
| remix_tj | And the other question is: are operations swift.proxy.controllers.base.get_container_info and swift.proxy.controllers.base.get_object_info the best functions to get the metadata set on a container and an object respectively? | 13:08 |
| *** ktsuyuzaki has joined #openstack-swift | 13:09 | |
| *** ChanServ sets mode: +v ktsuyuzaki | 13:09 | |
| *** armaan has quit IRC | 13:09 | |
| *** armaan has joined #openstack-swift | 13:10 | |
| *** psachin has quit IRC | 13:11 | |
| *** openstackgerrit has quit IRC | 13:11 | |
| *** zigo has quit IRC | 13:11 | |
| *** kota_ has quit IRC | 13:11 | |
| *** vinsh has quit IRC | 13:12 | |
| acoles | remix_tj: the proxy is appropriate for most middlewares, there's some docs here https://docs.openstack.org/swift/latest/development_middleware.html | 13:12 |
| remix_tj | yes, it's exactly where i was looking at. | 13:12 |
| *** vinsh has joined #openstack-swift | 13:12 | |
| *** ktsuyuzaki has quit IRC | 13:13 | |
| acoles | get_object_info does an object backend HEAD, so may or may not be useful depending on whether you need to fetch the object anyway, or if some other middleware gas already fetched info during the current request. get_container_info may return cached info (from memcache) or do a backend HEAD. | 13:15 |
| acoles | s/gas/has/ | 13:15 |
| *** kota_ has joined #openstack-swift | 13:16 | |
| *** ChanServ sets mode: +v kota_ | 13:16 | |
| remix_tj | i want to check if a given metadata is set on object and given the value to decide if an operation can be performed | 13:16 |
| remix_tj | if the metadata is not set on object, i want to lookup for "global setting" looking into container's metadata. | 13:18 |
| remix_tj | acoles: last stupid questions: if i try to access the data through S3api, can i invoke the middleware? how? | 13:20 |
| acoles | remix_tj: I was just looking for an example of how get_object_info is used but only see it in tests. should do the job though. be aware of the docstring re. internal info requests being pre-authed. | 13:20 |
| acoles | remix_tj: you mean invoke your own middleware? | 13:21 |
| remix_tj | yes | 13:21 |
| acoles | I'm not too familiar with the s3api but as far as I know it is compatible with other middlewares | 13:23 |
| *** psachin` has quit IRC | 13:24 | |
| remix_tj | acoles: maybe is enough to add my middleware in the pipeline after the s3api middleware? | 13:27 |
| *** neonpastor has quit IRC | 13:27 | |
| acoles | yes I'd guess it needs to be after (to the right of) in the pipeline | 13:27 |
| remix_tj | so i'm expected to have the s3 calls converted into swift calls? | 13:30 |
| acoles | internally the proxy server only handles swift requests so that is what I'd expect to be handling to the right of s3api. there may be some nuances but you'd need to check with others - try asking later in US daytime or even later in Japan daytime. | 13:33 |
| *** armaan has quit IRC | 13:34 | |
| *** NM has quit IRC | 13:34 | |
| *** armaan has joined #openstack-swift | 13:34 | |
| remix_tj | acoles: i start to develop my middleware with only swift support and then, if is the case, i'll try to expand with s3 support | 13:35 |
| *** NM has joined #openstack-swift | 13:36 | |
| *** NM has quit IRC | 13:40 | |
| *** gyankum has quit IRC | 14:05 | |
| *** Supun has joined #openstack-swift | 14:10 | |
| *** NM has joined #openstack-swift | 14:10 | |
| *** NM has quit IRC | 14:11 | |
| *** armaan has quit IRC | 14:16 | |
| *** armaan has joined #openstack-swift | 14:17 | |
| *** armaan has quit IRC | 14:20 | |
| *** armaan has joined #openstack-swift | 14:21 | |
| *** NM has joined #openstack-swift | 14:25 | |
| *** NM has quit IRC | 14:27 | |
| *** hseipp has joined #openstack-swift | 14:30 | |
| *** NM has joined #openstack-swift | 14:33 | |
| *** NM has joined #openstack-swift | 14:39 | |
| *** NM has quit IRC | 14:39 | |
| *** NM has joined #openstack-swift | 14:43 | |
| *** NM has joined #openstack-swift | 14:45 | |
| *** NM has quit IRC | 15:07 | |
| *** pcaruana has quit IRC | 15:09 | |
| *** Supun has quit IRC | 15:17 | |
| *** ukaynar has joined #openstack-swift | 15:28 | |
| *** Supun has joined #openstack-swift | 15:41 | |
| *** armaan has quit IRC | 15:49 | |
| *** gyee has joined #openstack-swift | 16:03 | |
| notmyname | good morning | 16:17 |
| *** Supun has quit IRC | 16:21 | |
| *** openstackgerrit has joined #openstack-swift | 16:21 | |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Add ShardRange class https://review.openstack.org/565743 | 16:21 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Add support for sharding in ContainerBroker https://review.openstack.org/565744 | 16:21 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Add shard range support to container server https://review.openstack.org/565745 | 16:21 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Enable proxy to build listings from shards https://review.openstack.org/565746 | 16:21 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Redirect object updates to shard containers https://review.openstack.org/565747 | 16:21 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Add sharder daemon, manage_shard_ranges tool and probe tests https://review.openstack.org/565748 | 16:22 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Add container sharding documentation https://review.openstack.org/565749 | 16:22 |
| acoles | ^^ first round of revisions | 16:22 |
| notmyname | acoles: thanks! | 16:22 |
| *** hseipp has quit IRC | 16:23 | |
| acoles | This is a better link for seeing the deep-review chain (excludes abandoned patches) https://review.openstack.org/#/q/branch:feature/deep-review+status:open | 16:25 |
| *** NM has joined #openstack-swift | 16:52 | |
| *** cshastri has joined #openstack-swift | 17:07 | |
| *** silor has quit IRC | 17:21 | |
| *** ccamacho has quit IRC | 17:27 | |
| notmyname | I've been thinking about who's around in the swift community and how we measure "sufficient" reviews on feature/deep-review | 17:49 |
| notmyname | with the limited number of people we have, how do we get enough reviews on the feature branch, especially when considering who's been contributing code to it | 17:50 |
| notmyname | for that matter, what is "enough reviews"? | 17:51 |
| notmyname | here's what I've thought through | 17:51 |
| notmyname | a reviewer needs to answer a few questions: (1) do I understand the overall design and think it's a good way to solve the stated problem? (2) is the proposed code maintainable? is it tested? does it avoid obviously bad code smells? (3) is there a single area of the review chain you've looked at in more depth and you've done a full review on? | 17:53 |
| notmyname | I think that if we have that sort of mindset, it will allow us to make more progress than the mindset of "I must start at the beginning and go to the end and leave either a -1 or +2" | 17:53 |
| notmyname | and if we're all doing that, then collectively we've done reviews on the full proposed chain | 17:54 |
| notmyname | I think it's a lot to ask of everyone to load several years of work into your head and understand all of the nuance and give approval or not, all in the next 2 weeks | 17:55 |
| notmyname | instead, look at the high level design and goals, look for code maintainability, look for obvious wrong things, and answer the question "is this code that solves problems for users and that I'm willing to maintain?" | 17:55 |
| notmyname | the hard part, of course, is the interactions between different parts. acoles has done a great job breaking the feature into a patch chain of logical parts | 17:56 |
| notmyname | but it's still a lot easier to look at a class or daemon and say it looks good without seeing the issues that arise from interactions with other parts of the system | 17:57 |
| *** geaaru has quit IRC | 17:57 | |
| notmyname | mattoliverau wrote a lot of the initial work, acoles wrote the largest part of the proposed patch, clayg and timburke made significant code contributions | 18:00 |
| notmyname | of the 8 patches in the chain, if I reviewed 1 and 6-8 and tdasilva reviewed 1-4 and kota_ reviewed 3-7 and mattoliverau reviewed 4, 6, and 8 and clayg review 1-2 and 5-7 and etc etc for everyone else, then we've got a way to make reasonable progress on reviewing a really big feature | 18:01 |
| notmyname | sorry for the big wall of text. I think I'm just about done :-) | 18:02 |
| notmyname | mostly, in summary, I don't want people to feel overwhelmed by the scope and therefore feel like they can't make any progress at all | 18:03 |
| notmyname | I want to know your thoughts on this, too. that's what I've been thinking | 18:05 |
| *** Supun has joined #openstack-swift | 18:08 | |
| *** NM has quit IRC | 18:17 | |
| *** ukaynar has quit IRC | 18:18 | |
| *** mvenesio has joined #openstack-swift | 18:19 | |
| *** ukaynar has joined #openstack-swift | 18:20 | |
| mvenesio | Hi Guys, how are you, is there any script to synchronize accounts between two swift clusters ( keystone based authenticated ) ? | 18:21 |
| notmyname | timur: ^^ | 18:21 |
| notmyname | mvenesio: yes. there's a project at https://github.com/swiftstack/swift-s3-sync that can do exactly that | 18:23 |
| notmyname | (the name comes from the original scope of that project. it's expanded in scope since then, and will be renamed soon [this week, if I can get to it]) | 18:24 |
| mvenesio | Hi John, Thanks man | 18:25 |
| mvenesio | notmyname: i'll try it | 18:25 |
| notmyname | mvenesio: please feel free to ask questions in here about it. either of myself or timur | 18:25 |
| mvenesio | notmyname: i'll do it for sure, thanks again | 18:26 |
| timur | mvenesio: the keystone auth will require merging the keystone auth PR. It's not merged yet, as I haven't had time to do the final review of it | 18:28 |
| notmyname | https://github.com/swiftstack/swift-s3-sync/pull/23 | 18:28 |
| timur | patch: https://github.com/swiftstack/swift-s3-sync/pull/16 | 18:28 |
| notmyname | thanks timur :-) | 18:28 |
| mvenesio | timur: great, i'll need that | 18:28 |
| timur | yea, we should merge one of those -- we were going back and forth on formatting, hence the two PRs. I think it may need to be rebased and merge conflicts fixed up :\ | 18:29 |
| mvenesio | timur: when do you think could be merged ? | 18:32 |
| *** Supun has quit IRC | 18:37 | |
| notmyname | mvenesio: it might be tricky to get it merged this week. | 18:39 |
| mvenesio | notmyname: ok not problem | 18:40 |
| *** ukaynar has quit IRC | 18:40 | |
| notmyname | mvenesio: what is it you're trying to do? wht's the overall goal/problem? | 18:40 |
| mvenesio | notmyname: basically we have a client with a swift redhat cluster installed, and we recently installed a brand new cluster , and we have to migrate all the data since the old one is getting out of space. | 18:42 |
| mvenesio | notmyname: both uses keystone to authenticate | 18:42 |
| notmyname | are you moving away from the redhat cluster? is there a reason you can't add capacity to the existing cluster? | 18:42 |
| mvenesio | notmyname: i think its something about pricing | 18:43 |
| mvenesio | notmyname: and support | 18:43 |
| notmyname | ok. I wanted to make sure you weren't under the impression that swift clusters had a static size and couldn't be expanded :-) | 18:44 |
| mvenesio | notmyname: no, this isn't that case, but thanks for asking | 18:46 |
| *** cshastri has quit IRC | 18:49 | |
| *** Supun has joined #openstack-swift | 18:58 | |
| *** ukaynar has joined #openstack-swift | 19:03 | |
| *** Supun has quit IRC | 19:08 | |
| *** ukaynar has quit IRC | 19:28 | |
| *** NM has joined #openstack-swift | 19:29 | |
| *** ukaynar_ has joined #openstack-swift | 19:35 | |
| *** NM has quit IRC | 19:39 | |
| *** NM has joined #openstack-swift | 19:43 | |
| *** ukaynar_ has quit IRC | 20:06 | |
| *** ukaynar has joined #openstack-swift | 20:10 | |
| *** mvenesio has quit IRC | 20:38 | |
| *** mvenesio has joined #openstack-swift | 20:39 | |
| *** geaaru has joined #openstack-swift | 20:40 | |
| *** lxkong_ has joined #openstack-swift | 20:45 | |
| notmyname | latest eventlet, 0.23.0 release yesterday, breaks swift. https://bugs.launchpad.net/swift/+bug/1769749 | 20:46 |
| openstack | Launchpad bug 1769749 in OpenStack Object Storage (swift) "eventlet 0.23.0 breaks swift's unicode handling" [Undecided,New] | 20:46 |
| notmyname | note that the gate is currently capped at 0.20.0 | 20:46 |
| *** szaher_ has joined #openstack-swift | 20:48 | |
| *** mvenesio has quit IRC | 20:53 | |
| *** lxkong has quit IRC | 20:53 | |
| *** mvk has quit IRC | 20:53 | |
| *** zaitcev has quit IRC | 20:53 | |
| *** szaher has quit IRC | 20:53 | |
| *** lxkong_ is now known as lxkong | 20:53 | |
| *** mvk has joined #openstack-swift | 20:54 | |
| *** kota_ has quit IRC | 20:55 | |
| *** kota_ has joined #openstack-swift | 20:58 | |
| *** ChanServ sets mode: +v kota_ | 20:58 | |
| *** NM has quit IRC | 21:30 | |
| *** gholt has joined #openstack-swift | 21:43 | |
| gholt | Anybody noticed any oddities recently with Unicode decoding? We're getting this in our Travis builds today. https://pastebin.com/gxrNwPxP | 21:47 |
| gholt | I figure it's some weird combo of a bad library version or Python version or something. But I figured I'd ping you guys in case it was "known" before I spent a bunch of time trying to figure it out. :) | 21:48 |
| gholt | That trace is from the proxy processing a request from the test: test_dlo_post_with_manifest_header (test.functional.test_dlo.TestDloUTF8) | 21:51 |
| gholt | But it has a similar issue with a lot of tests today. | 21:51 |
| gholt | In case the URL matters, it was ResponseError: 500: 'Internal Error' ('PUT' '/v1/AUTH_test/794551b2041240ba94db630bb0df022d/%E0%A4%82%EB%8C%8B%E0%A4%83%E8%94%88%ED%88%90%EB%8C%8D%E1%A0%83%EB%8C%8B%E1%A0%82%E0%A4%83%E0%A4%80%ED%88%8E%EF%84%91%E9%90%89%E0%A4%80') txid=tx01f5b124a6f34273a717a-005af0c5c3 | 21:52 |
| notmyname | gholt: we just found out last night's eventlet broke some unicode stuff | 22:00 |
| notmyname | 0.23.0 | 22:00 |
| notmyname | https://bugs.launchpad.net/swift/+bug/1769749 | 22:00 |
| gholt | Ah okay! | 22:00 |
| openstack | Launchpad bug 1769749 in OpenStack Object Storage (swift) "eventlet 0.23.0 breaks swift's unicode handling" [Undecided,Confirmed] | 22:00 |
| notmyname | openstack gate is pinned at 0.20.0 currently, so haven't seen issues there | 22:01 |
| gholt | I'm happy I asked then! | 22:01 |
| gholt | I'll check into that for us. | 22:01 |
| gholt | Should the requirements.txt be updated? | 22:05 |
| notmyname | yes, but no not yet | 22:06 |
| gholt | Okay, I'll try a sed -i for our Travis builds | 22:07 |
| clayg | well, you know how it is - everyone is constantly trying to figure out the most efficient way to maintain requirements.txt over the years. pinning works as long as you update it, looser restrictions work as long to keep you apprised of problematic releases. | 22:07 |
| notmyname | it may be that there's a relatively simple fix for it. someone here suggested perhaps an update to swob may fix it. but I don't know yet | 22:08 |
| *** ukaynar has quit IRC | 22:08 | |
| clayg | in the case of eventlet OpenStack has this storied history of having to blacklist eventlet releases - but I think we all finally agreed to work with temoto and try to make OpenStack and eventlet releases cohabitate more - i'm not sure what broke down this time exactly. | 22:09 |
| gholt | Hehe, np. | 22:10 |
| gholt | I'm just really glad I asked because who knows how long I'd have spun my wheels researching areas I simply do not know anymore. :) | 22:10 |
| clayg | go team! | 22:10 |
| *** ukaynar has joined #openstack-swift | 22:11 | |
| gholt | It's funny that at conferences Swift is still my only claim to fame. ;^) And then I get asked all sorts of questions I can't answer anymore. | 22:12 |
| gholt | Dangit, == versus =. Time to wait for Travis again. | 22:15 |
| *** ukaynar has quit IRC | 22:15 | |
| clayg | 🤣 You're very famous gholt, the kids all ask me if I really knew THE gholt 😉 | 22:17 |
| notmyname | 50 years from now, an old man sits in a rocking chair. sunlight coming in through the windows. The peacefulness of the afternoon is suddenly broken as 5 small children rush in, shouting. "Grandpa Clay! Grandpa Clay! Tell us the stories of the olden times. The before times. Did you really know gholt? THE gholt?!" | 22:19 |
| clayg | ☝️ | 22:20 |
| clayg | Of course they know I knew notmyname - EVERYONE knows notmyname | 22:20 |
| notmyname | "We don't talk of that subject" | 22:21 |
| gholt | not.mn | 22:21 |
| gholt | Yay, green builds! | 22:22 |
| gholt | ty guys | 22:22 |
| notmyname | gholt: \o/ | 22:23 |
| *** rcernin has joined #openstack-swift | 22:25 | |
| openstackgerrit | Gage Hugo proposed openstack/python-swiftclient master: WIP - Migrate to using keystoneauth session https://review.openstack.org/560138 | 22:34 |
| *** klamath_ has joined #openstack-swift | 22:49 | |
| *** klamath has quit IRC | 22:49 | |
| -openstackstatus- NOTICE: Any devstack job failure due to rsync errors related to tripleo-incubator can safely be rechecked now | 22:58 | |
| openstackgerrit | Merged openstack/slogging master: First commit of devstack directory https://review.openstack.org/564708 | 22:58 |
| mattoliverau | morning | 23:08 |
| *** kei_yama has joined #openstack-swift | 23:14 | |
| *** geaaru has quit IRC | 23:18 | |
| *** cloudnull is now known as cloudkiller | 23:20 | |
| *** cloudkiller is now known as cloudnull | 23:21 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift feature/deep-review: Add config option shard_replication_quorum https://review.openstack.org/566362 | 23:51 |
| openstackgerrit | Keiichi Hikita proposed openstack/slogging master: This commit remove dead link in Readme https://review.openstack.org/566762 | 23:55 |
| openstackgerrit | Clay Gerrard proposed openstack/swift feature/deep-review: Add config option shard_replication_quorum https://review.openstack.org/566362 | 23:56 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!