| *** gyee has quit IRC | 00:20 | |
| timburke | clayg, do we have a set of recommended/supported REGIONS/ZONES/NODES/DISKS/EC_DISKS/REPLICAS for https://github.com/swiftstack/vagrant-swift-all-in-one/ ? i'm trying some kinda off-the-wall configs and it keeps failing bring-up because the dispersion's crap... | 00:53 |
|---|---|---|
| timburke | which, fair enough -- it is -- but idk that we should abort the rest of building the rings because of it... | 00:53 |
| openstackgerrit | Arzhna Lee proposed openstack/swift master: Allow lowercase region name for AWS .NET SDK compatibility https://review.opendev.org/742322 | 00:55 |
| openstackgerrit | Arzhna Lee proposed openstack/swift master: Allow lowercase region name for AWS .NET SDK compatibility https://review.opendev.org/742322 | 01:01 |
| openstackgerrit | Arzhna Lee proposed openstack/swift master: Allow lowercase region name for AWS .NET SDK compatibility https://review.opendev.org/742322 | 01:05 |
| openstackgerrit | Arzhna Lee proposed openstack/swift master: Allow lowercase region name for AWS .NET SDK compatibility https://review.opendev.org/742322 | 01:07 |
| openstackgerrit | Arzhna Lee proposed openstack/swift master: Allow lowercase region name for AWS .NET SDK compatibility https://review.opendev.org/742322 | 01:08 |
| seongsoocho | morning! ^ above issue occured in my cluster. The AWS only has a lowercase region name . So some of sdk change the region name to lowercase ... | 01:31 |
| timburke | man, that makes our old default before https://github.com/openstack/swift/commit/692a034 even more of a pain... | 01:38 |
| timburke | will review tomorrow | 01:38 |
| seongsoocho | timburke: oh I see. As my customers use s3 api more and more, there are many issue report. :-) . | 01:47 |
| *** baojg has quit IRC | 03:10 | |
| openstackgerrit | Merged openstack/swift master: Decouple proxy_server tests https://review.opendev.org/742281 | 03:23 |
| *** psachin has joined #openstack-swift | 03:31 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: py3: Work with proper native string paths in crypto meta https://review.opendev.org/742033 | 03:38 |
| openstackgerrit | Tim Burke proposed openstack/swift stable/ussuri: py3: Stop munging RAW_PATH_INFO https://review.opendev.org/742338 | 03:44 |
| openstackgerrit | Tim Burke proposed openstack/swift stable/train: py3: Stop munging RAW_PATH_INFO https://review.opendev.org/742339 | 03:44 |
| *** mahatic has joined #openstack-swift | 03:45 | |
| *** ChanServ sets mode: +v mahatic | 03:45 | |
| ormandj | timburke: burning the midnight oil | 03:58 |
| timburke | i shouldn't -- but my gate's finally (mostly) fixed, so i wanted to get some stuff moving | 03:59 |
| ormandj | i'll look around tomorrow and see how that other patch is behaving to give some more feedback | 04:00 |
| openstackgerrit | Merged openstack/swift master: probe tests: Get rid of `server` arg for device_dir() and storage_dir() https://review.opendev.org/741544 | 04:12 |
| openstackgerrit | Merged openstack/swift master: Stop syncing empty suffixes list https://review.opendev.org/741538 | 04:22 |
| *** camelCaser has quit IRC | 04:23 | |
| *** camelCaser has joined #openstack-swift | 04:26 | |
| openstackgerrit | Merged openstack/swift stable/ussuri: Use latest pip in gate jobs https://review.opendev.org/742304 | 04:48 |
| openstackgerrit | Merged openstack/swift stable/train: Use latest pip in gate jobs https://review.opendev.org/742305 | 04:48 |
| *** m75abrams has joined #openstack-swift | 04:53 | |
| openstackgerrit | zhouyunyang proposed openstack/swift master: skip same suffixes sync when update_delete https://review.opendev.org/742342 | 05:06 |
| openstackgerrit | Merged openstack/swift master: py3: Stop munging RAW_PATH_INFO https://review.opendev.org/737856 | 05:07 |
| *** rcernin has quit IRC | 07:32 | |
| *** gtema has joined #openstack-swift | 07:33 | |
| openstackgerrit | zhouyunyang proposed openstack/swift master: skip same suffixes sync when update_delete https://review.opendev.org/742360 | 07:43 |
| *** gtema has quit IRC | 08:20 | |
| *** gtema has joined #openstack-swift | 08:24 | |
| *** baojg has joined #openstack-swift | 08:26 | |
| *** cschwede has joined #openstack-swift | 08:32 | |
| *** ChanServ sets mode: +v cschwede | 08:32 | |
| *** gtema has quit IRC | 08:39 | |
| *** gtema has joined #openstack-swift | 08:41 | |
| *** rcernin has joined #openstack-swift | 09:11 | |
| *** rcernin has quit IRC | 09:35 | |
| openstackgerrit | David Sariel proposed openstack/swift master: Let developers/operators add watchers to object audit (simplified) https://review.opendev.org/706653 | 10:35 |
| *** tkajinam has quit IRC | 10:59 | |
| *** gtema has quit IRC | 11:34 | |
| *** baojg has quit IRC | 11:49 | |
| *** rcernin has joined #openstack-swift | 11:49 | |
| *** baojg has joined #openstack-swift | 11:50 | |
| *** gtema has joined #openstack-swift | 11:58 | |
| *** gtema has quit IRC | 12:03 | |
| *** gtema has joined #openstack-swift | 12:09 | |
| *** baojg has quit IRC | 12:11 | |
| *** baojg has joined #openstack-swift | 12:12 | |
| *** nottrobin_ has joined #openstack-swift | 12:31 | |
| *** nottrobin has quit IRC | 12:38 | |
| *** nottrobin_ is now known as nottrobin | 12:38 | |
| *** mahatic has quit IRC | 12:41 | |
| *** baojg has quit IRC | 13:23 | |
| *** baojg has joined #openstack-swift | 13:24 | |
| *** gtema has quit IRC | 13:31 | |
| *** rcernin has quit IRC | 13:32 | |
| *** baojg has quit IRC | 13:41 | |
| *** baojg has joined #openstack-swift | 13:43 | |
| openstackgerrit | David Sariel proposed openstack/swift master: Let developers/operators add watchers to object audit (simplified) https://review.opendev.org/706653 | 13:57 |
| *** ormandj has quit IRC | 14:06 | |
| *** ormandj has joined #openstack-swift | 14:09 | |
| openstackgerrit | David Sariel proposed openstack/swift master: Let developers/operators add watchers to object audit (simplified) https://review.opendev.org/706653 | 14:18 |
| *** baojg has quit IRC | 14:22 | |
| *** baojg has joined #openstack-swift | 14:24 | |
| *** psachin has quit IRC | 14:24 | |
| *** paladox has quit IRC | 14:32 | |
| *** paladox has joined #openstack-swift | 14:53 | |
| *** gyee has joined #openstack-swift | 15:48 | |
| *** baojg has quit IRC | 15:54 | |
| *** baojg has joined #openstack-swift | 15:56 | |
| *** m75abrams has quit IRC | 16:21 | |
| *** jadonn has joined #openstack-swift | 16:33 | |
| *** baojg has quit IRC | 16:34 | |
| *** cschwede has quit IRC | 16:41 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Start to decouple Object GET path https://review.opendev.org/733911 | 17:20 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Add concurrent_gets to EC GET requests https://review.opendev.org/711342 | 17:20 |
| *** Nisha_Agarwal has joined #openstack-swift | 17:22 | |
| Nisha_Agarwal | Hello all | 17:23 |
| Nisha_Agarwal | I wanted to ask a query....Does swift container configuration support NFS URL scheme configuration as well? | 17:24 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: swift-container-info: Show shard ranges summary https://review.opendev.org/737056 | 17:29 |
| timburke | Nisha_Agarwal, i'm not quite sure what that would mean -- would you want to expose a swift container as a filesystem over nfs? i think there are some other projects that might be able to do that, but we don't support it natively | 17:34 |
| Nisha_Agarwal | timburke, Thanks. I wanted to ask from tempurl perspective. In ironic we create the media and place it in swift containers and generally swift temp url scheme is http(s) | 17:35 |
| Nisha_Agarwal | timburke, in this aspect is it possible to configure swift containers in such a way that the tempurls generated at run time has nfs url scheme | 17:36 |
| Nisha_Agarwal | the hardware has a limitation to use nfs or cifs | 17:37 |
| Nisha_Agarwal | so just wanted to know on that aspect | 17:37 |
| timburke | all of swift's apis are built on http, so it would make sense to me that tempurls would be http (or https) -- if there's a hard requirement to use nfs, you'll likely need to have some sort of a gateway that speaks nfs on one side and translates it to swift (http) requests on the other | 17:39 |
| timburke | i think that'll be entirely separate from thinking about tempurls -- they only really make sense in the context of http (where you can have query parameters as part of the protocol); i don't think nfs supports anything quite like that | 17:40 |
| timburke | the auth mechanism will likely be very different | 17:40 |
| Nisha_Agarwal | timburke, ok ...the temporary url generated has to be used for inserting to virtual cdrom of the server | 17:40 |
| Nisha_Agarwal | timburke, agreed | 17:40 |
| Nisha_Agarwal | timburke, i couldnt find any documentation related to nfs hence i pinged here to validate if my findings are correct. | 17:41 |
| timburke | fwiw, there's https://github.com/NVIDIA/vdisc/ which says it's "a tool for creating and mounting virtual CD-ROM images backed by object storage" | 17:42 |
| timburke | (full disclosure: that's put out by my employer, and i haven't actually played around with it yet) | 17:42 |
| *** baojg has joined #openstack-swift | 19:07 | |
| *** Nisha_Agarwal has quit IRC | 19:27 | |
| *** baojg has quit IRC | 19:31 | |
| timburke | hrm. i think bug 1446873 and p 740867 complicate the upgrade story for p 735751 a decent bit... is it worth adding a use_legacy_network option, or should we just include a release note like "When upgrading from earlier versions of Swift, expect replication errors until all nodes are upgraded"? | 19:32 |
| openstack | bug 1446873 in OpenStack Object Storage (swift) "ssync doesn't work with replication_server = true" [Medium,In progress] https://launchpad.net/bugs/1446873 | 19:32 |
| patchbot | https://review.opendev.org/#/c/740867/ - swift - Allow replication servers to handle all request me... - 2 patch sets | 19:32 |
| patchbot | https://review.opendev.org/#/c/735751/ - swift - Allow direct and internal clients to use the repli... - 4 patch sets | 19:32 |
| *** baojg has joined #openstack-swift | 19:32 | |
| clayg | ouch, upgrades 😞 | 19:48 |
| timburke | if we go with an option for it, it won't be as narrow as i'd originally done, either -- we'd want it for *everything* :-( | 19:50 |
| clayg | LAME! | 19:52 |
| clayg | so but we already deploy with replication_server = True commnted out? I mean... what is actually blocking requests until post upgrade? | 19:53 |
| clayg | like we might have replication_server = True turned on the database servers; so maybe UpgradeImpact just reminds us to turn that off - then upgrade - then you can turn it back on (because post upgrade `replication_server = True` is the same as `# replication_server`) | 19:55 |
| clayg | I guess that's not ideal (you have to change your configs or get errors during upgrade) ... but I can live with it (if it works) | 19:55 |
| timburke | *for us*? hopefully nothing? i guess the release note could be like "if you have replication_servers=true in any configs, remove that line to ensure a smooth rolling upgrade"... not so bad... | 19:56 |
| timburke | did you just check object, or a/c as well? | 19:56 |
| clayg | yeah we HAD to turn off object - but until now there was no obvious reason to disable it for a/c | 19:58 |
| clayg | so yeah; I think this is all workable 💪 | 19:58 |
| clayg | unless we forgot about something 🤣 | 19:58 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: container-sharding: Stable overlap order https://review.opendev.org/742535 | 20:21 |
| zaitcev | timburke: are having a meeting in 40 minutes? | 20:21 |
| timburke | that's my plan -- sorry, haven't updated the agenda yet | 20:22 |
| openstackgerrit | Merged openstack/swift stable/train: Breakup reclaim into batches https://review.opendev.org/741755 | 20:40 |
| openstackgerrit | Merged openstack/swift stable/ussuri: Breakup reclaim into batches https://review.opendev.org/741753 | 20:59 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Allow replication servers to handle all request methods https://review.opendev.org/740867 | 20:59 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Allow direct and internal clients to use the replication network https://review.opendev.org/735751 | 20:59 |
| openstackgerrit | Tim Burke proposed openstack/swift master: wip: Allow probe tests to run with separate replication network https://review.opendev.org/741723 | 20:59 |
| *** baojg has quit IRC | 21:22 | |
| *** baojg has joined #openstack-swift | 21:23 | |
| clayg | lots to talk about today! | 22:01 |
| timburke | :-( so direct_client (at least) *will* spew tracebacks on 405... | 22:23 |
| timburke | seems like internal_client must not | 22:28 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Actually interpolate error message https://review.opendev.org/742549 | 22:43 |
| *** rcernin has joined #openstack-swift | 22:52 | |
| *** rcernin has quit IRC | 22:58 | |
| *** rcernin has joined #openstack-swift | 23:01 | |
| *** rcernin has quit IRC | 23:04 | |
| *** rcernin has joined #openstack-swift | 23:05 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!