| *** germs_ has quit IRC | 00:48 | |
| *** gyee has quit IRC | 01:38 | |
| spsurya | Hello all | 01:46 |
|---|---|---|
| spsurya | is there any automated tool to install SAIO ? | 01:47 |
| spsurya | rather than following SAIO deployment guide | 01:47 |
| *** gyankum has joined #openstack-swift | 01:55 | |
| *** bkopilov has quit IRC | 02:00 | |
| notmyname | spsurya: are you wanting to build a developer einvironment or for a testing target? | 02:06 |
| spsurya | notmyname: for POC and testing | 02:07 |
| spsurya | but quick one | 02:07 |
| spsurya | I know SAIO is more about dev purpose | 02:07 |
| notmyname | what do you mean by POC? are you testing that an app works when talking to swift or are you testing a deployment to see if it will work at your scale? | 02:07 |
| spsurya | before that i have a query | 02:10 |
| notmyname | you could be wanting to contribute code to swift. you could be wanting to practice with swift from a sysadmin perspective. you could be wanting to test an app to see if it works against swift. depending on which of these three categories you fall in, I can point you in different directions | 02:10 |
| spsurya | can i want to run swift existing account,container and object into container | 02:12 |
| notmyname | I'm sorry, but I don't understand what you mean | 02:13 |
| spsurya | i want to migrate swift services into container | 02:13 |
| spsurya | i.e operational swift into container | 02:13 |
| spsurya | thats is my goal | 02:14 |
| spsurya | but to reach there i am doing small tests of swift | 02:14 |
| spsurya | notmyname: hope you get my point | 02:15 |
| notmyname | not completely, yet. by container you mean docker containers? lxc? | 02:15 |
| spsurya | notmyname: docker container | 02:15 |
| spsurya | notmyname: lemme explain | 02:16 |
| spsurya | suppose you have installed swift on two node and each one running the application account, contqainer and object and proxy too | 02:17 |
| notmyname | ok | 02:17 |
| spsurya | as any operator add new node to swift cluster then after rebalancing new node would be part of that cluster | 02:18 |
| notmyname | right | 02:18 |
| spsurya | now rather than adding node to swift cluster i want to add container in swift cluster | 02:18 |
| spsurya | and now that container would be serving all the purpose of swift services | 02:19 |
| spsurya | so container would be running account, container and object service with old data of swift | 02:20 |
| spsurya | this is what i want to do at the moment | 02:20 |
| spsurya | but before that also wanted to install SAIO for dev just to few things | 02:21 |
| spsurya | Hope you got my point ^^ | 02:21 |
| spsurya | notmyname: | 02:21 |
| notmyname | what have you found or tried so far? | 02:22 |
| spsurya | notmyname: i want to find the configuration point where we configure Swift to add new node | 02:24 |
| spsurya | can you show me the point in swift | 02:24 |
| notmyname | that doesn't have anything to do with docker... | 02:25 |
| notmyname | new nodes are added by adding drives to a ring | 02:25 |
| notmyname | normally via `swift-ring-builder` | 02:25 |
| spsurya | notmyname: yes, we add drive to ring | 02:25 |
| spsurya | so wanted to add swift container docker/volume to swift-ring file | 02:26 |
| spsurya | we can check if more configuration or support required for thisd | 02:26 |
| notmyname | devices in a ring are identified by IP, port, and mount point. (as an operator, you also give it a zone and region based on your cluster layout) | 02:27 |
| spsurya | so notmyname i am testing all these point | 02:27 |
| notmyname | that doesn't have anything to do with how you choose to package the code (docker or otherwise) | 02:27 |
| spsurya | yes in docker by some tooling | 02:28 |
| spsurya | just trying to do this | 02:29 |
| notmyname | I'm not sure exactly what you're wanting, but I'm sure you're getting frustrated with me not understanding. here's a project that runs some swift processes in a docker container: https://github.com/bouncestorage/docker-swift | 02:29 |
| spsurya | notmyname: | 02:29 |
| notmyname | I know of zero people who run swift in production in this way | 02:29 |
| notmyname | that project is only one way to do things. and it's not designed to do *any* clustering at all | 02:29 |
| notmyname | how you package and deploy the linux processes does't matter. use docker. use distro packages. build your own from source. for any way, you'll end up with a set of hard drives across all your machines that are each identified by an (IP, port, mountpoint). add that info to your ring files, deploy your rings, and the running swift processes will take care of the rest | 02:31 |
| notmyname | that's not to say there aren't any hard problems with docker-izing swift. that's still a thing to do. you can look at the project above to help you figure that out. | 02:33 |
| spsurya | notmyname: thanks for info | 02:34 |
| *** gkadam has joined #openstack-swift | 02:36 | |
| *** klamath has quit IRC | 02:36 | |
| spsurya | notmyname: will you be at Summit ? | 02:38 |
| notmyname | I will, but not the whole week. I'll be there tuesday afternoon through thursday | 02:39 |
| spsurya | notmyname: nice | 02:45 |
| *** psachin has joined #openstack-swift | 02:50 | |
| spsurya | https://www.irccloud.com/pastebin/55liTRcJ/ | 02:57 |
| spsurya | notmyname: | 02:57 |
| spsurya | failing on $$$$ cd $HOME/swift; sudo pip install --no-binary cryptography -r requirements.txt; sudo python setup.py develop; cd - | 03:03 |
| *** gkadam has quit IRC | 03:07 | |
| *** gyankum has quit IRC | 03:20 | |
| *** bkopilov has joined #openstack-swift | 03:23 | |
| *** links has joined #openstack-swift | 03:41 | |
| *** klamath has joined #openstack-swift | 03:51 | |
| *** klamath has quit IRC | 04:07 | |
| *** gyankum has joined #openstack-swift | 04:10 | |
| *** cshastri has joined #openstack-swift | 04:59 | |
| kota_ | good afternoon | 05:36 |
| *** armaan has joined #openstack-swift | 05:47 | |
| *** psachin has quit IRC | 06:03 | |
| *** armaan has quit IRC | 06:10 | |
| *** armaan has joined #openstack-swift | 06:10 | |
| *** psachin has joined #openstack-swift | 06:11 | |
| *** silor has joined #openstack-swift | 06:19 | |
| *** klamath has joined #openstack-swift | 06:39 | |
| *** klamath has quit IRC | 06:44 | |
| *** ccamacho has joined #openstack-swift | 07:08 | |
| *** rcernin has quit IRC | 07:10 | |
| *** geaaru has joined #openstack-swift | 07:17 | |
| *** tesseract has joined #openstack-swift | 07:17 | |
| *** mikecmpbll has joined #openstack-swift | 07:54 | |
| *** armaan has quit IRC | 08:06 | |
| *** armaan has joined #openstack-swift | 08:07 | |
| *** armaan has quit IRC | 08:11 | |
| *** armaan has joined #openstack-swift | 08:11 | |
| *** zaitcev has quit IRC | 08:12 | |
| acoles | good morning | 08:17 |
| *** cbartz has joined #openstack-swift | 08:22 | |
| *** kei_yama has quit IRC | 08:44 | |
| *** gkadam has joined #openstack-swift | 08:53 | |
| *** links has quit IRC | 09:07 | |
| *** jistr has quit IRC | 09:09 | |
| *** jistr has joined #openstack-swift | 09:12 | |
| *** links has joined #openstack-swift | 09:23 | |
| yuxin_ | o/ | 09:28 |
| yuxin_ | good afternoon | 09:28 |
| yuxin_ | I am writing unit tests for swift3, and I need to patch swift3.request.get_container_info. But there are two places calling get_container_info in swift3/requet.py with different args (with[1] and without[2] swift_source kwarg). | 09:28 |
| yuxin_ | [1]https://github.com/openstack/swift3/blob/master/swift3/request.py#L1103 | 09:29 |
| yuxin_ | [2]https://github.com/openstack/swift3/blob/master/swift3/request.py#L1254 | 09:29 |
| yuxin_ | When it patched with `lambda x, y: {}`, it raised an error said `lambda() got an unexpected keyword argument 'swift_source'`. can anyone help me out? | 09:29 |
| acoles | yuxin_: try `lambda x, y, **kwargs: {}`, or if you don't care about the args in the lambda, `lambda *args, **kwargs: {}` | 09:40 |
| yuxin_ | acoles: thanks, acloes! it works. | 09:43 |
| *** silor1 has joined #openstack-swift | 09:58 | |
| *** silor has quit IRC | 10:00 | |
| *** silor1 is now known as silor | 10:00 | |
| *** bkopilov has quit IRC | 10:05 | |
| kota_ | yuxin_: note that swift3 is now a part of swift upstream as "s3api". if you're thinking of using the newest master, you can consider to fix the Swift's one. | 10:20 |
| kota_ | yuxin_: around https://github.com/openstack/swift/blob/master/swift/common/middleware/s3api/s3request.py | 10:20 |
| * kota_ has been tied up to fix my presentation slides for OpenStack Summit in the next week to get approval from my boss :/ | 10:21 | |
| kota_ | i hope i can work on reviewing deep container since tomorrow, it might be... too late??? | 10:22 |
| *** corvus has quit IRC | 10:29 | |
| *** corvus has joined #openstack-swift | 10:30 | |
| *** notmyname has quit IRC | 10:33 | |
| *** notmyname has joined #openstack-swift | 10:34 | |
| *** ChanServ sets mode: +v notmyname | 10:34 | |
| acoles | kota_: the first few patches in the chain have had a lot of review but it would be great if you get chance to look at the later ones | 10:41 |
| *** klamath has joined #openstack-swift | 10:41 | |
| kota_ | acoles: ok, thanks for the info! | 10:43 |
| *** links has quit IRC | 10:44 | |
| *** klamath has quit IRC | 10:46 | |
| *** links has joined #openstack-swift | 10:57 | |
| remix_tj | just for curiosity... how does internal swift communication happens? how does proxy communicates with account/container/object server? Using http queries? | 11:07 |
| yuxin_ | kota_: cool. | 11:08 |
| yuxin_ | By the way, I am working on swift3 object versioning and wonder if anyone still focusing on this feature. I've got some ideas on this and would like to get my works submitted. | 11:08 |
| acoles | remix_tj: yes, http is used between internal services | 11:10 |
| remix_tj | so i expect to see queries to the object server requesting to do something on the object? | 11:10 |
| acoles | remix_tj: yes, proxy sends http requests to object server(s) that host the particular object | 11:12 |
| remix_tj | i'm trying to design a WORM middleware for trying to replace retention features that EMC Centera provides (file cannot be updated or deleted before the expiration of a given retention time, SEC Rule 17a-4) | 11:14 |
| remix_tj | It's just for playing with some python code | 11:14 |
| remix_tj | i think the best option is to write a filter that denies some operations | 11:16 |
| *** armaan has quit IRC | 11:26 | |
| *** armaan has joined #openstack-swift | 11:27 | |
| openstackgerrit | Nguyen Hai proposed openstack/python-swiftclient master: Change python-swiftclient webpage to openstackdocstheme https://review.openstack.org/553433 | 11:37 |
| *** gkadam has quit IRC | 11:43 | |
| *** gkadam has joined #openstack-swift | 11:43 | |
| *** vinsh_ has joined #openstack-swift | 12:32 | |
| *** vinsh has quit IRC | 12:32 | |
| *** armaan has quit IRC | 12:41 | |
| *** klamath has joined #openstack-swift | 12:42 | |
| *** armaan has joined #openstack-swift | 12:42 | |
| *** klamath has quit IRC | 12:46 | |
| *** armaan has quit IRC | 12:50 | |
| *** armaan has joined #openstack-swift | 12:50 | |
| openstackgerrit | Nguyen Hai proposed openstack/python-swiftclient master: Switch from oslosphinx to openstackdocstheme https://review.openstack.org/553433 | 12:57 |
| *** mikecmpbll has quit IRC | 13:30 | |
| *** cshastri has quit IRC | 13:38 | |
| *** klamath has joined #openstack-swift | 13:38 | |
| *** vinsh_ has quit IRC | 13:38 | |
| *** vinsh has joined #openstack-swift | 13:39 | |
| *** mikecmpbll has joined #openstack-swift | 13:44 | |
| *** bkopilov has joined #openstack-swift | 13:47 | |
| *** mvk has quit IRC | 13:52 | |
| openstackgerrit | Tovin Seven proposed openstack/swift master: OSprofiler in OpenStack Swift https://review.openstack.org/468316 | 13:57 |
| *** cbartz has quit IRC | 13:58 | |
| *** mikecmpb_ has joined #openstack-swift | 14:01 | |
| *** mikecmpbll has quit IRC | 14:02 | |
| tdasilva | yuxin_: great to hear your interest on swift3, but as kota_ mentioned, I'd make sure to send any patches to s3api and yeah, object versioning is difentely welcome | 14:09 |
| *** gyankum has quit IRC | 14:09 | |
| tdasilva | remix_tj: WORM functionality is one of those requests that always come up, the real challenge with WORM in swift is eventual consistency, preventing overwrites is very difficult | 14:10 |
| *** zaitcev has joined #openstack-swift | 14:11 | |
| *** ChanServ sets mode: +v zaitcev | 14:11 | |
| tdasilva | remix_tj: and thanks for the PRs on ansible-saio, will look at those today. | 14:12 |
| remix_tj | tdasilva: i'm trying to think like centera does. In centera once you wrote an object and set the retention, you're denied to do any edit/delete operations. I understand the issue of eventual consistency and i'll try to take in account, but i'll first start with a filter | 14:13 |
| remix_tj | we've several applications requiring to store data for an indefinite time ensuring that no changes will happen to those data | 14:13 |
| remix_tj | up to now only some propretary solutions does this and in several cases it's done only with a flag on the data container saying that nothing can be deleted or changed before X time | 14:15 |
| remix_tj | anyway it's an excercise :-) | 14:16 |
| tdasilva | remix_tj: one way to achieve is with object versioning, that way even if an object got overwritten you were able to maintain an old copy | 14:17 |
| tdasilva | but i realize that not really WORM functionality like centera does | 14:18 |
| remix_tj | no, because versioning means that at time t0 i'll have the good content, after the update with a request i'll get the updated content that is not what i require | 14:19 |
| remix_tj | *requested | 14:19 |
| remix_tj | i require to have the original content served at every time. | 14:20 |
| remix_tj | i have to check on enterprise storages what happens on retained objects if i try to update their content. | 14:20 |
| DHE | so you need eventually consistent with the OLDEST copy served? | 14:21 |
| remix_tj | DHE: indeed i require to have only one version of data | 14:22 |
| remix_tj | in centera if i write an object it gets written with new identifier, so is not possible to overwrite the existing content. That's due to the nature of that kind of storage. But i've seen that EMC implemented the same retention on S3 buckets, so i'll try to explore the behaviour and then test | 14:25 |
| *** germs has joined #openstack-swift | 14:27 | |
| *** germs has quit IRC | 14:27 | |
| *** germs has joined #openstack-swift | 14:27 | |
| *** mvk has joined #openstack-swift | 14:29 | |
| tdasilva | remix_tj: ack...please do keep us in the loop | 14:30 |
| remix_tj | sure | 14:30 |
| remix_tj | tdasilva: i'll start with the simplest thing which is filtering delete requests, when i'll move on more complex things :-) | 14:32 |
| *** germs has quit IRC | 14:32 | |
| remix_tj | is there any blueprint about it on launchpad? | 14:33 |
| notmyname | preventing deletes isnt' too hard. preventing overwrites is very very difficult | 14:33 |
| tdasilva | remix_tj: i believe the centera approach could be replicated. i.e., give a UID to an object (probably the object url itself), but then keep that list somewhere that is strong consistent | 14:34 |
| tdasilva | was just having this conversation with notmyname yesterday... | 14:34 |
| remix_tj | tdasilva: on centera if you write an object you obtain an ID. If you loose the ID there's only one way to recover data: obtain the list of all the stored data and look into the metadata to see if matches with what you're looking at | 14:35 |
| remix_tj | metadata of each stored object | 14:36 |
| notmyname | fortunately, swift doesn't work that way ;-) | 14:37 |
| remix_tj | yeah, fortunately :-) | 14:38 |
| notmyname | reminder: the september PTG tickets early-bird pricing ends tomorrow. register now, if you haven't already | 14:43 |
| notmyname | https://www.eventbrite.com/e/project-teams-gathering-denver-2018-tickets-45296703660 | 14:43 |
| *** armaan has quit IRC | 14:47 | |
| remix_tj | i'm "very happy" i can't be at berlin summit this year. I think i'll get ATC discount code, but my wife should give birth to my first son/daughter during the summit, so i'll skip and wait for another opportunity in my zone :-) | 14:47 |
| *** cbartz has joined #openstack-swift | 14:54 | |
| *** links has quit IRC | 14:58 | |
| *** ukaynar has joined #openstack-swift | 15:00 | |
| tdasilva | remix_tj: congrats! that is a good excuse indeed :) | 15:08 |
| remix_tj | :-) | 15:09 |
| *** cshastri has joined #openstack-swift | 15:30 | |
| *** gyankum has joined #openstack-swift | 15:39 | |
| clayg | notmyname: I always assumed a storage policy with a different cleanup hashdir could just prefer first timestamp - done. | 16:00 |
| *** itlinux has joined #openstack-swift | 16:05 | |
| timburke | good morning | 16:11 |
| notmyname | clayg: interesting | 16:24 |
| *** germs has joined #openstack-swift | 16:28 | |
| *** germs has quit IRC | 16:28 | |
| *** germs has joined #openstack-swift | 16:28 | |
| *** gyee has joined #openstack-swift | 16:29 | |
| *** germs has quit IRC | 16:32 | |
| *** itlinux has quit IRC | 16:36 | |
| *** ccamacho has quit IRC | 16:37 | |
| *** gkadam has quit IRC | 16:46 | |
| *** links has joined #openstack-swift | 16:53 | |
| *** germs has joined #openstack-swift | 17:04 | |
| *** germs has quit IRC | 17:04 | |
| *** germs has joined #openstack-swift | 17:05 | |
| *** germs has quit IRC | 17:05 | |
| *** germs has joined #openstack-swift | 17:05 | |
| *** germs has quit IRC | 17:05 | |
| *** germs has joined #openstack-swift | 17:06 | |
| *** germs has quit IRC | 17:06 | |
| *** SkyRocknRoll has joined #openstack-swift | 17:07 | |
| *** germs has joined #openstack-swift | 17:07 | |
| *** itlinux has joined #openstack-swift | 17:12 | |
| *** psachin has quit IRC | 17:14 | |
| *** links has quit IRC | 17:16 | |
| *** SkyRocknRoll has quit IRC | 17:20 | |
| *** mikecmpb_ has quit IRC | 17:23 | |
| *** tesseract has quit IRC | 17:27 | |
| *** mvk has quit IRC | 17:27 | |
| *** SkyRocknRoll has joined #openstack-swift | 17:33 | |
| *** SkyRocknRoll has quit IRC | 17:35 | |
| *** cshastri has quit IRC | 17:38 | |
| *** NM has joined #openstack-swift | 17:40 | |
| *** mikecmpbll has joined #openstack-swift | 17:46 | |
| *** cbartz has quit IRC | 17:53 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient master: Back out some version bumps https://review.openstack.org/568914 | 17:54 |
| *** NM has quit IRC | 18:00 | |
| *** NM has joined #openstack-swift | 18:01 | |
| *** NM has quit IRC | 18:01 | |
| *** NM has joined #openstack-swift | 18:03 | |
| *** NM has quit IRC | 18:05 | |
| *** geaaru has quit IRC | 18:13 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient master: Back out some version bumps https://review.openstack.org/568914 | 18:15 |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient master: Back out some version bumps https://review.openstack.org/568914 | 18:16 |
| *** NM has joined #openstack-swift | 18:16 | |
| *** eandersson has joined #openstack-swift | 18:17 | |
| eandersson | Is this the recommended way to provide region_name for Swift? https://review.openstack.org/#/c/568921/1/zaqar/storage/swift/driver.py | 18:18 |
| patchbot | patch 568921 - zaqar - [WIP] Adding Swift backend Region support | 18:18 |
| notmyname | eandersson: that doesn't look right | 18:19 |
| eandersson | Do you have an example? I was looking at this https://github.com/openstack/python-swiftclient/blob/master/swiftclient/client.py#L1561 | 18:19 |
| notmyname | hang on. maybe it is. checking | 18:19 |
| notmyname | yes. that's the way to pass a region name to keystone for auth | 18:21 |
| eandersson | cool thanks notmyname | 18:21 |
| notmyname | I see some references to os_region_name, too. TBH, I'm not entirely sure ont he difference | 18:21 |
| eandersson | I might follow up with an update to the Swift examples after this :P | 18:22 |
| notmyname | cool, thanks! | 18:22 |
| *** gyankum has quit IRC | 18:24 | |
| DHE | swift regions and openstack regions don't align. openstack regions will provide different URLs to different services depending on regions. hence openstack directs you to a local proxy server, but that's different from storage regions | 18:27 |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient master: Back out some version bumps https://review.openstack.org/568914 | 18:29 |
| eandersson | Yea - in this case the issue was that it was using the first keystone catalog endpoint. | 18:30 |
| eandersson | Which lead it to connect to a Swift backend in a different region. | 18:30 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Add support for sharding in ContainerBroker https://review.openstack.org/565744 | 18:47 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Add shard range support to container server https://review.openstack.org/565745 | 18:47 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Enable proxy to build listings from shards https://review.openstack.org/565746 | 18:47 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Redirect object updates to shard containers https://review.openstack.org/565747 | 18:47 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Add sharder daemon, manage_shard_ranges tool and probe tests https://review.openstack.org/565748 | 18:47 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: Add container sharding documentation https://review.openstack.org/565749 | 18:47 |
| openstackgerrit | Alistair Coles proposed openstack/swift feature/deep-review: SQUASH: handle delimiter listings in proxy https://review.openstack.org/568930 | 18:47 |
| acoles | timburke: sorry I got bogged down with some tests today and didn't get through all your review comments today so that's just what I have...feel free to push new patchsets if you want, or I'll crack on tomorrow. | 18:49 |
| *** mvk has joined #openstack-swift | 19:22 | |
| *** NM has quit IRC | 19:51 | |
| *** NM has joined #openstack-swift | 19:54 | |
| *** silor has quit IRC | 19:58 | |
| *** NM1 has joined #openstack-swift | 20:03 | |
| *** NM has quit IRC | 20:03 | |
| timburke | dependencies are the worst. when i did https://github.com/openstack/python-swiftclient/commit/ff505d7 i made a liar of our impressively-low requests>=1.1 requirement, 'cause the exception came in with https://github.com/requests/requests/commit/811ee4e (so, requests 2.4.0) | 20:19 |
| timburke | also, http://logs.openstack.org/14/568914/4/check/openstack-tox-lower-constraints/bca39dc/job-output.txt.gz#_2018-05-16_18_34_43_901554 is some pretty *horrible* test-failure output | 20:21 |
| *** ukaynar has quit IRC | 20:28 | |
| notmyname | swift team meeting in 10 minutes | 20:50 |
| clayg | wooooooo meetings! | 20:50 |
| notmyname | I didn't get re-banned in -meeting, so we'll have it in #openstack-meeting ;-) | 20:50 |
| notmyname | this week we'll talk about feature/deep-review and specifically when to merge it | 20:51 |
| *** m_kazuhiro has joined #openstack-swift | 20:52 | |
| *** NM1 has quit IRC | 20:53 | |
| *** rledisez1 has joined #openstack-swift | 20:55 | |
| *** rledisez1 has quit IRC | 20:57 | |
| kota_ | good morning | 20:58 |
| *** alecuyer has quit IRC | 20:58 | |
| *** rledisez has quit IRC | 20:58 | |
| *** rledisez has joined #openstack-swift | 20:58 | |
| notmyname | meeting time | 21:00 |
| clayg | wooooo! | 21:02 |
| *** itlinux has quit IRC | 21:13 | |
| *** NM has joined #openstack-swift | 21:23 | |
| zaitcev | acoles: is there a comment or a function with evident code that explains how params['marker'] += '\x00' works in feature/deep? | 21:25 |
| zaitcev | oh, wait. Meeting | 21:25 |
| timburke | is this about making end_marker inclusive? tacking on a null byte gets you the very next possible string | 21:29 |
| acoles | zaitcev: timburke I'll need to reload context some more but IIRC it's to do with shard range upper being inclusive but end marker not being inclusive | 21:33 |
| *** d0ugal_ has joined #openstack-swift | 21:39 | |
| *** d0ugal has quit IRC | 21:41 | |
| *** itlinux has joined #openstack-swift | 21:51 | |
| *** NM has quit IRC | 22:11 | |
| *** portante has quit IRC | 22:15 | |
| *** ndk- has quit IRC | 22:16 | |
| *** rcernin has joined #openstack-swift | 22:25 | |
| *** rledisez has quit IRC | 22:26 | |
| *** alecuyer has joined #openstack-swift | 22:26 | |
| *** rledisez has joined #openstack-swift | 22:27 | |
| *** m_kazuhiro has quit IRC | 22:27 | |
| *** itlinux has quit IRC | 22:33 | |
| *** portante has joined #openstack-swift | 22:35 | |
| *** ndk_ has joined #openstack-swift | 22:36 | |
| *** d0ugal__ has joined #openstack-swift | 22:40 | |
| *** d0ugal_ has quit IRC | 22:41 | |
| clayg | notmyname: epic meeting bruh | 22:44 |
| eandersson | btw the region code is not working as intended | 22:49 |
| eandersson | https://github.com/openstack/python-swiftclient/blob/master/swiftclient/client.py#L647 | 22:49 |
| eandersson | That line should be passing on region_name from os_options | 22:50 |
| eandersson | Had to do it like this https://review.openstack.org/#/c/568921/3/zaqar/storage/swift/driver.py | 22:56 |
| patchbot | patch 568921 - zaqar - Adding Swift backend Region support | 22:56 |
| tdasilva | sorry i missed missed, just caught up reading scroll back! go container sharding! go acoles ! | 22:58 |
| openstackgerrit | Erik Olof Gunnar Andersson proposed openstack/python-swiftclient master: Make swiftclient respect region_name when using sessions https://review.openstack.org/568979 | 23:04 |
| eandersson | ^ should fix the region issue | 23:07 |
| notmyname | thanks | 23:09 |
| eandersson | No more using the endpoint for China when trying to access Swift from the US :D | 23:10 |
| *** ndk_ has quit IRC | 23:17 | |
| *** portante has quit IRC | 23:18 | |
| *** ndk_ has joined #openstack-swift | 23:20 | |
| *** portante has joined #openstack-swift | 23:21 | |
| *** kei_yama has joined #openstack-swift | 23:21 | |
| *** germs has quit IRC | 23:33 | |
| *** germs has joined #openstack-swift | 23:35 | |
| *** germs has quit IRC | 23:35 | |
| *** germs has joined #openstack-swift | 23:35 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!