| *** tonanhngo has joined #openstack-swift | 00:03 | |
| *** tonanhngo has quit IRC | 00:07 | |
| *** ^_andrea_^ has quit IRC | 00:13 | |
| openstackgerrit | Merged openstack/swift master: Sync metadata in 'rsync_then_merge' in db_replicator https://review.openstack.org/302494 | 00:43 |
|---|---|---|
| *** klrmn has quit IRC | 00:56 | |
| openstackgerrit | Merged openstack/swift master: Use config_number method instead of node_id + 1 https://review.openstack.org/471613 | 01:13 |
| openstackgerrit | Merged openstack/swift master: Using assertIsNone() instead of assertEqual(None) https://review.openstack.org/471570 | 01:13 |
| tdasilva | timburke, kota_ how do you guys feel about the libec/pyeclib release patches? | 01:14 |
| *** zhengyin has joined #openstack-swift | 01:15 | |
| timburke | tdasilva: i don't really like that compiles are broken on a supported platform (https://review.openstack.org/#/c/459028/), but i'm not willing to veto a release over it | 01:15 |
| patchbot | patch 459028 - liberasurecode - Un-inline get/set_metatdata_chksum | 01:15 |
| tdasilva | timburke: mmm...who's got a mac to review that? | 01:22 |
| timburke | i do! i do! :P | 01:23 |
| tdasilva | timburke: "liberasurecode is known to work" != 'supported' :P | 01:23 |
| timburke | notmyname: clayg: feel like compiling liberasurecode on your host? :-) | 01:23 |
| timburke | tdasilva: at the very least, we ought to update the README to say "known to *not* work" then :P | 01:24 |
| tdasilva | timburke: lol | 01:24 |
| tdasilva | i don't mind removing the reference to OSX | 01:25 |
| tdasilva | timburke: but honestly, i'm not in a super hurry mode to get this in, but would like to see it in pike | 01:25 |
| tdasilva | so I think M3 is the cut off for me | 01:26 |
| timburke | i definitely agree it's worth having another release soon, though. between all the memory-leak fixes, zaitcev's patches, and the libphazr support, it's gonna be great | 01:30 |
| zaitcev | I can review 459028 without a Mac. | 01:36 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Send Connection:close headers when we haven't read the request body https://review.openstack.org/471993 | 01:46 |
| *** zhurong has joined #openstack-swift | 01:48 | |
| *** jamielennox is now known as jamielennox|away | 02:04 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Limit number of revert tombstone SSYNC requests https://review.openstack.org/439572 | 02:09 |
| zaitcev | Honestly... When I see liberasurecode, I think that the author had no idea how to write in C in general, didn't know basic techniques and common pitfalls. For example, packed structures are stupidly popular, but their useful application range is not as wide as some assume. Throwing the natural alignment out has costs, and not just performance. | 02:11 |
| zaitcev | Tushar and Kevin have some to answer for | 02:11 |
| zaitcev | But then I, too, ignored that library and was too happy to trust them without looking. Too late now! | 02:11 |
| *** bkopilov has quit IRC | 02:12 | |
| zaitcev | Still... If you ever write something to disk, like a header, not making its fields endian-independent is a crime. | 02:13 |
| *** jamielennox|away is now known as jamielennox | 02:15 | |
| *** thurloat[m] has left #openstack-swift | 02:23 | |
| kota_ | oops | 02:26 |
| kota_ | clayg: sorry, I overwrote your fix, I didn't notice on that | 02:26 |
| *** gyee has quit IRC | 02:27 | |
| *** klrmn has joined #openstack-swift | 02:36 | |
| *** Sukhdev has joined #openstack-swift | 02:36 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Limit number of revert tombstone SSYNC requests https://review.openstack.org/439572 | 02:36 |
| *** klrmn has quit IRC | 03:23 | |
| *** bkopilov has joined #openstack-swift | 03:32 | |
| *** links has joined #openstack-swift | 03:34 | |
| *** gkadam has joined #openstack-swift | 03:42 | |
| *** zhurong has quit IRC | 03:45 | |
| *** kei_yama has quit IRC | 03:45 | |
| timburke | [ahead 1, behind 1458] yowza! | 03:50 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Improve content negotiation https://review.openstack.org/207276 | 03:50 |
| *** kei_yama has joined #openstack-swift | 04:00 | |
| *** psachin has joined #openstack-swift | 04:02 | |
| *** itlinux has joined #openstack-swift | 04:03 | |
| *** pcaruana has joined #openstack-swift | 04:31 | |
| *** aselius has quit IRC | 04:32 | |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Use more backends in existing tests https://review.openstack.org/438215 | 04:32 |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Make test format strings consistent with test function names https://review.openstack.org/438214 | 04:32 |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Use preprocessor to build test suites https://review.openstack.org/431794 | 04:32 |
| *** zhengyin has quit IRC | 04:34 | |
| *** zhurong has joined #openstack-swift | 04:34 | |
| *** zhengyin has joined #openstack-swift | 04:36 | |
| *** pcaruana has quit IRC | 04:57 | |
| *** jaosorior_away is now known as jaosorior | 05:05 | |
| *** jamielennox is now known as jamielennox|away | 05:12 | |
| *** jamielennox|away is now known as jamielennox | 05:19 | |
| *** itlinux has quit IRC | 05:21 | |
| *** zhengyin has quit IRC | 05:34 | |
| *** Sukhdev has quit IRC | 05:35 | |
| *** zhengyin has joined #openstack-swift | 05:35 | |
| *** cshastri has joined #openstack-swift | 05:44 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 05:55 |
| *** jamielennox is now known as jamielennox|away | 05:56 | |
| *** jamielennox|away is now known as jamielennox | 06:03 | |
| *** ChubYann has quit IRC | 06:05 | |
| *** rcernin has joined #openstack-swift | 06:15 | |
| *** skudlik has joined #openstack-swift | 06:24 | |
| *** zhurong has quit IRC | 06:32 | |
| *** hseipp has joined #openstack-swift | 06:44 | |
| *** pcaruana has joined #openstack-swift | 06:57 | |
| openstackgerrit | Mahati Chamarthy proposed openstack/swift master: Limit number of revert tombstone SSYNC requests https://review.openstack.org/439572 | 07:08 |
| mahatic | kota_: clayg : ^ it's a typo in commit msg, (s/RIVERT/REVERT). No further changes. Thank you for the discussion and modifications! | 07:08 |
| kota_ | mahatic: thx! +2'ed | 07:10 |
| mahatic | kota_: thanks, not sure if I could +A it though | 07:10 |
| *** tesseract has joined #openstack-swift | 07:12 | |
| *** mingyu has joined #openstack-swift | 07:13 | |
| *** mingyu has quit IRC | 07:22 | |
| *** gkadam is now known as gkadam-afk | 07:30 | |
| *** mingyu has joined #openstack-swift | 07:30 | |
| *** mingyu has quit IRC | 07:40 | |
| *** jaosorior has quit IRC | 07:42 | |
| *** zhurong has joined #openstack-swift | 07:46 | |
| *** bkopilov has quit IRC | 07:47 | |
| *** jaosorior has joined #openstack-swift | 07:54 | |
| *** cbartz has joined #openstack-swift | 07:58 | |
| *** hseipp has quit IRC | 08:01 | |
| *** oshritf has joined #openstack-swift | 08:04 | |
| mathiasb | good morning | 08:06 |
| mathiasb | mahatic acoles jrichli timburke: I updated the script for setting up a VM with devstack with swift+barbican patch https://review.openstack.org/364878/ | 08:06 |
| patchbot | patch 364878 - swift - Retrieve encryption root secret from Barbican | 08:06 |
| mathiasb | and I also added a Vagrantfile for creating the VM using vagrant and virtualbox | 08:07 |
| mathiasb | links to both files can be found at https://etherpad.openstack.org/p/swift-kms-keymaster-setup | 08:07 |
| mathiasb | in case anyone has some spare time on their hands and feels like reviewing/testing crypto stuff :) | 08:07 |
| *** zhurong has quit IRC | 08:13 | |
| *** zhurong has joined #openstack-swift | 08:15 | |
| *** gkadam-afk is now known as gkadam | 08:16 | |
| *** geaaru has joined #openstack-swift | 08:16 | |
| *** qwertyco has joined #openstack-swift | 08:17 | |
| mahatic | mathiasb: thank you! I've marked those links, I hope to leave a review soon | 08:22 |
| *** links has quit IRC | 08:25 | |
| *** links has joined #openstack-swift | 08:38 | |
| mathiasb | great, thanks mahatic! | 08:45 |
| *** hseipp has joined #openstack-swift | 08:53 | |
| *** winggundamth has joined #openstack-swift | 08:53 | |
| *** winggundamth has quit IRC | 08:58 | |
| *** winggundamth has joined #openstack-swift | 09:00 | |
| *** mvk has joined #openstack-swift | 09:02 | |
| *** Dw_Sn has joined #openstack-swift | 09:05 | |
| *** zhengyin has quit IRC | 09:21 | |
| *** zhengyin has joined #openstack-swift | 09:21 | |
| *** qwertyco has quit IRC | 09:23 | |
| acoles | rledisez: I updated and added some comment to bug 1652323 | 09:26 |
| openstack | bug 1652323 in OpenStack Object Storage (swift) "ssync syncs an expired object as a tombstone" [Medium,Confirmed] https://launchpad.net/bugs/1652323 | 09:26 |
| acoles | rledisez: and based on your and clayg's irc comments the consensus seems to be that the 'right' fix is to sync the expired data, so in EC case make changes to allow GETs of expired fragments for rebuild | 09:28 |
| acoles | rledisez: now we just need dev resource to write the patch :) | 09:28 |
| acoles | cc notmyname ^^ | 09:30 |
| *** zhengyin has quit IRC | 09:33 | |
| *** cshastri has quit IRC | 09:33 | |
| *** zhengyin has joined #openstack-swift | 09:33 | |
| *** cshastri has joined #openstack-swift | 09:43 | |
| *** mat128 has joined #openstack-swift | 10:06 | |
| *** zhengyin has quit IRC | 11:05 | |
| Dw_Sn | any idea where I can find a doc. that show all URL translation between AWS S3 and swif3 ? I know I have been asking this question for few days | 11:15 |
| Dw_Sn | whenever I have temp_url_sig and temp_url_experis everything works... once I change those to be like AWS S3 " Expreis and Signutre " the authentication doesn't work .. any idea where I can find any kind of guide ? | 11:16 |
| *** mvk has quit IRC | 11:16 | |
| Dw_Sn | timburke: You mentioned to me before about Swift3 pre-signed URL that is compatible with S3 and it is not swift tempurl ... any idea where I can read / use that ? | 11:18 |
| *** hseipp has quit IRC | 11:29 | |
| *** mingyu has joined #openstack-swift | 11:35 | |
| *** hseipp has joined #openstack-swift | 11:35 | |
| *** mingyu has quit IRC | 11:45 | |
| *** mingyu has joined #openstack-swift | 11:46 | |
| *** kei_yama has quit IRC | 11:47 | |
| *** mvk has joined #openstack-swift | 11:47 | |
| *** mingyu has quit IRC | 11:53 | |
| *** mingyu has joined #openstack-swift | 11:58 | |
| openstackgerrit | Christopher Bartz proposed openstack/swift master: Exclude containers for account quota https://review.openstack.org/414232 | 12:02 |
| *** mingyu has quit IRC | 12:11 | |
| *** mingyu has joined #openstack-swift | 12:11 | |
| *** mingyu has quit IRC | 12:16 | |
| *** winggundamth has quit IRC | 12:22 | |
| *** ianychoi has quit IRC | 12:27 | |
| *** ianychoi has joined #openstack-swift | 12:28 | |
| rledisez | acoles: i'll try to have a look at the code tomorrow to evaluate the amount of work | 12:28 |
| *** catintheroof has joined #openstack-swift | 12:32 | |
| *** mat128 has quit IRC | 12:34 | |
| *** zhurong has quit IRC | 12:36 | |
| *** jroll has left #openstack-swift | 12:46 | |
| *** gkadam has quit IRC | 12:50 | |
| *** klamath has joined #openstack-swift | 13:01 | |
| *** klamath has quit IRC | 13:02 | |
| *** klamath has joined #openstack-swift | 13:03 | |
| *** oshritf has quit IRC | 13:03 | |
| *** hseipp has quit IRC | 13:04 | |
| *** oshritf has joined #openstack-swift | 13:08 | |
| *** mat128 has joined #openstack-swift | 13:09 | |
| *** mat128 has quit IRC | 13:10 | |
| *** vint_bra has joined #openstack-swift | 13:12 | |
| *** klamath_ has joined #openstack-swift | 13:12 | |
| *** klamath has quit IRC | 13:13 | |
| *** mat128 has joined #openstack-swift | 13:15 | |
| *** links has quit IRC | 13:15 | |
| *** lucasxu has joined #openstack-swift | 13:16 | |
| *** joeljwright has joined #openstack-swift | 13:20 | |
| *** ChanServ sets mode: +v joeljwright | 13:20 | |
| *** SkyRocknRoll has joined #openstack-swift | 13:28 | |
| *** oshritf has quit IRC | 13:36 | |
| *** lucasxu has quit IRC | 13:38 | |
| *** lucasxu has joined #openstack-swift | 13:40 | |
| *** lucasxu has quit IRC | 13:51 | |
| *** oshritf has joined #openstack-swift | 13:52 | |
| *** ianychoi_ has joined #openstack-swift | 13:54 | |
| *** ianychoi has quit IRC | 13:57 | |
| *** ianychoi_ is now known as ianychoi | 13:57 | |
| acoles | rledisez: ack, thanks | 13:57 |
| *** oshritf has quit IRC | 13:58 | |
| acoles | rledisez: I think it will need a probe test, I can at least help with that if needed | 13:59 |
| *** oshritf has joined #openstack-swift | 14:01 | |
| *** lucasxu has joined #openstack-swift | 14:02 | |
| *** chlong has joined #openstack-swift | 14:02 | |
| *** oshritf has quit IRC | 14:08 | |
| jrichli | mathiasb: That is great! Thanks for doing that. I hope to take a look this weekend. | 14:11 |
| *** oshritf has joined #openstack-swift | 14:16 | |
| *** ganders has joined #openstack-swift | 14:17 | |
| *** oshritf has quit IRC | 14:24 | |
| tdasilva | mathiasb: very cool, thanks for working on that. | 14:31 |
| *** oshritf has joined #openstack-swift | 14:36 | |
| *** skudlik has quit IRC | 14:40 | |
| *** mat128 has quit IRC | 14:43 | |
| *** aselius has joined #openstack-swift | 14:46 | |
| *** klrmn has joined #openstack-swift | 14:47 | |
| *** cshastri has quit IRC | 14:49 | |
| *** logan- has joined #openstack-swift | 14:52 | |
| *** psachin has quit IRC | 14:57 | |
| *** SkyRocknRoll has quit IRC | 14:58 | |
| *** hseipp has joined #openstack-swift | 15:07 | |
| *** mat128 has joined #openstack-swift | 15:14 | |
| *** rcernin has quit IRC | 15:16 | |
| asettle | notmyname: you up? :) | 15:17 |
| *** SkyRocknRoll has joined #openstack-swift | 15:32 | |
| *** chsc has joined #openstack-swift | 15:42 | |
| *** chsc has joined #openstack-swift | 15:42 | |
| *** gyee has joined #openstack-swift | 15:42 | |
| *** chsc has quit IRC | 15:48 | |
| *** gyee has quit IRC | 15:58 | |
| *** jaosorior has quit IRC | 16:00 | |
| *** mat128 has quit IRC | 16:08 | |
| *** gkadam has joined #openstack-swift | 16:12 | |
| *** ganders has quit IRC | 16:13 | |
| *** geaaru has quit IRC | 16:14 | |
| *** gyee has joined #openstack-swift | 16:16 | |
| *** cbartz has quit IRC | 16:17 | |
| *** skudlik has joined #openstack-swift | 16:19 | |
| *** hseipp has quit IRC | 16:26 | |
| *** chsc has joined #openstack-swift | 16:29 | |
| *** chsc has joined #openstack-swift | 16:29 | |
| *** mingyu has joined #openstack-swift | 16:29 | |
| notmyname | good morning | 16:29 |
| notmyname | asettle: was commuting. what's up? | 16:29 |
| notmyname | tdasilva: cschwede: do either of you know venkata edara at redhat? http://lists.openstack.org/pipermail/openstack-dev/2017-June/118027.html | 16:31 |
| *** tesseract has quit IRC | 16:32 | |
| tdasilva | notmyname: I don't know Venkata Edara, but that's probably from the guys working on gluster-swift project | 16:34 |
| *** chsc has quit IRC | 16:39 | |
| *** ganders has joined #openstack-swift | 16:43 | |
| *** chsc has joined #openstack-swift | 16:47 | |
| *** chsc has joined #openstack-swift | 16:47 | |
| *** itlinux has joined #openstack-swift | 16:48 | |
| *** mat128 has joined #openstack-swift | 16:50 | |
| *** pcaruana has quit IRC | 17:00 | |
| *** mat128 has quit IRC | 17:03 | |
| *** gkadam has quit IRC | 17:05 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Minor cleanup in test.unit.obj.test_reconstructor https://review.openstack.org/472347 | 17:07 |
| *** mat128 has joined #openstack-swift | 17:09 | |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Use preprocessor to build test suites https://review.openstack.org/431794 | 17:10 |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Remove testcase.skip support https://review.openstack.org/472351 | 17:10 |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Print test that's going to run *before* running test https://review.openstack.org/472352 | 17:10 |
| timburke | 1 file changed, 54 insertions(+), 633 deletions(-) | 17:11 |
| timburke | i *love* macros! | 17:11 |
| *** hseipp has joined #openstack-swift | 17:11 | |
| clayg | timburke: I forget the pattern you think it easiest to read/write for "demonstrate/assert that the one and only log message I'm expecting includes the pharse xyz" | 17:12 |
| *** Dw_Sn has quit IRC | 17:13 | |
| clayg | timburke: I feel like you cleaned some up recently - and called out some existing pattern i was riding on in patch 428408 | 17:13 |
| patchbot | https://review.openstack.org/#/c/428408/ - swift - Don't rehash primaries in reconstructor handoffs_o... | 17:13 |
| clayg | maybe I could do better? | 17:13 |
| *** lucasxu has quit IRC | 17:14 | |
| timburke | clayg: i liked where i was going in https://review.openstack.org/#/c/467710/ -- but i think for the log lines i just started adding messages like 'Expected only one error, got %r' % error_lines | 17:15 |
| patchbot | patch 467710 - swift - Clean up some assertions in test_reconstructor | 17:15 |
| *** hseipp has quit IRC | 17:16 | |
| timburke | i suppose you could do something like assertFalse(error_lines[N:]) then make assertions on those first N? but if something new snuck in at the start instead of the end, you're still gonna need to get more info on the failure | 17:17 |
| *** joeljwright has quit IRC | 17:19 | |
| clayg | it seemed low-value/burdensome to me to add the help text... maybe if it was a macro ;) | 17:20 |
| clayg | not so much in typing it out... more like when you add a log line to methods that were getting hit - s/1/2 (and maybe s/[0]/[-1]) and then also s/one/two - it's not great - maybe assertions on log lines are more trouble than they're worth? | 17:21 |
| timburke | may well be the case | 17:22 |
| *** ChubYann has joined #openstack-swift | 17:27 | |
| clayg | timburke: I think we just need an assertLength(3, container) that keeps the value and the error message consistent - but displays the contents of container when the length is unexpected | 17:32 |
| clayg | next time I find myself writing assertEqual(3, len(container)) I will definitely spin out a helper - thanks! | 17:32 |
| timburke | clayg: yeah, i was just thinking something similar... | 17:32 |
| clayg | well maybe I'll let you do it then - you're better at that frameworky level stuff anyway ;) | 17:33 |
| timburke | trouble is that we don't really have a common base test class. idk that i want it badly enough to touch *all* the tests | 17:35 |
| clayg | *I* sure don't! | 17:39 |
| clayg | add a MixIn to utils and we can apply it as we go? | 17:40 |
| *** itlinux has quit IRC | 17:45 | |
| *** tonanhngo has joined #openstack-swift | 17:50 | |
| *** tonanhngo has quit IRC | 17:52 | |
| *** tonanhngo has joined #openstack-swift | 17:53 | |
| *** oshritf has joined #openstack-swift | 17:54 | |
| *** lucasxu has joined #openstack-swift | 18:00 | |
| *** oshritf has quit IRC | 18:01 | |
| notmyname | FYI https://review.openstack.org/#/c/472275/ is the proposed plan for docs changes, given the changes in the docs team itself | 18:05 |
| patchbot | patch 472275 - docs-specs - Adds spec for migrating the documentation suite to... | 18:05 |
| notmyname | in general, I think this is a very good idea, but be aware that there is a big todo list that will fall to our team | 18:05 |
| notmyname | well, maybe not a huge list, but at least ongoing maintenance work | 18:05 |
| notmyname | the most important near-term thing we'll see as the swift team is a patch (or patches?) that will be to import the docs team docs into the swift tree | 18:06 |
| notmyname | but this work will fall to *us*; the docs team will not be pushing these patches | 18:07 |
| notmyname | so be aware of this proposed idea, comment as you see fit, and know that we'll need to decide as a group how we want to facilitate this docs migration | 18:07 |
| tdasilva | notmyname: today we have how many sets of docs? api, dev, install-guide, admin-guide? | 18:07 |
| notmyname | one of the good things is that this work should end up with us having one one docs tree and one docs job (potentially). but it might take a bit to get there | 18:08 |
| notmyname | a la https://review.openstack.org/#/c/386834/ | 18:09 |
| patchbot | patch 386834 - swift - Unify docs trees | 18:09 |
| openstackgerrit | Merged openstack/swift master: Limit number of revert tombstone SSYNC requests https://review.openstack.org/439572 | 18:10 |
| notmyname | kota_: mahatic: clayg: nice! | 18:10 |
| *** Sukhdev has joined #openstack-swift | 18:16 | |
| *** ganders has quit IRC | 18:19 | |
| *** skudlik has quit IRC | 18:22 | |
| *** mvk has quit IRC | 18:24 | |
| *** ganders has joined #openstack-swift | 18:29 | |
| *** itlinux_ has joined #openstack-swift | 18:38 | |
| asettle | notmyname: tagged you in relevant post :) as per our discussion of being le guinea pig for our transition, we would definitely appreciate your review. It's slightly different to your plan, but it is as the community voted :) | 18:38 |
| notmyname | asettle: see my comments here in channel on it | 18:38 |
| notmyname | asettle: the only part I'm thinking about right now is the dirtree layout in the repo | 18:39 |
| asettle | notmyname: yep :) all valid comments made above. Going to be pretty hefty. | 18:39 |
| notmyname | asettle: while the proposed one isn't bad or wrong, it's certainly more exhaustive that might be needed. I assume the proposed list is a superset? | 18:40 |
| notmyname | asettle: and I would very much like there to be no external dependencies on an in-tree structure. ie the only external usage would be how to do the docs build, not the layout of the docs in tree | 18:40 |
| *** skudlik has joined #openstack-swift | 18:42 | |
| asettle | notmyname: it's as exhausted as we have felt fit for now. We are admittedly running against a clock, so we may be over thinking some courses of action, but it is all to solve not just the install-guide issue. But the *docs team* issue as well :) | 18:42 |
| notmyname | right. I get that. the only thing that makes me hesitate on the proposal is the amount of work I'd be signing up the whole swift team for. but (1) it's necessary and (2) it's a good idea regardless of the state of the docs team and (3) it's necessary given the actual state of the docs team | 18:43 |
| notmyname | asettle: I do not envy your position | 18:44 |
| asettle | notmyname: yo tell me about it. Literally been tearing my hair out. I geniunely understand that putting this back on the project teams make sense, but it is just a horrible amount of work for everyone :( so, yeah, sympathies right back at you | 18:44 |
| asettle | Fortunately, I'm leaving the country tomorrow. I've thrown a spec in the ring and I'm hoping all the horribleness will happen while I'm getting fat on pasta :p | 18:45 |
| notmyname | so the amount of work (hey! maybe it's "just" copying some docs trees in) and slight trepidation about a leaky API (what I was saying about external knowledge of the internal structure) are the only things I'm thinking over | 18:45 |
| notmyname | but I think the proposal is sound | 18:45 |
| asettle | notmyname: totally makes sense. Completely understandable :) we wanna work with you guys first because we figured there's some good critical thinkers in here ;) and, well, to be honest, other than nova you're one of the bigger groups at the moment. | 18:45 |
| notmyname | ...and I hope you enjoy italy :-) | 18:45 |
| asettle | Thanks :D | 18:46 |
| asettle | So, you let us know everything you're thinking, and we'll work together :) | 18:46 |
| notmyname | I'm not sure how to phrase what I said in gerrit | 18:46 |
| notmyname | asettle: do you expect that the initial migration work is simply a copy of a directory tree from one project to another? or is it more complicated than that? | 18:47 |
| notmyname | eg will someone have to go through existing docs and pull apart per-project pieces paragraph by paragraph? | 18:48 |
| asettle | notmyname: hmmm, yes, to be honest. I mean, it'll be a fiddly copy. But that's literally how I'm thinking it'll be right now. | 18:48 |
| asettle | I can't imagine so, so, Doug linked to: https://etherpad.openstack.org/p/doc-migration-tracking | 18:48 |
| asettle | In the spec | 18:48 |
| asettle | He has gone through each guide and figured out what goes where | 18:49 |
| asettle | So, in theory, it'll be a case of "go find that file, get the thing, move out" | 18:49 |
| asettle | We very rarely mixed project subjects in files | 18:49 |
| notmyname | ah, but if it's not mixed per file, it *does* seem like it's mixed in that one existing docs project has different chapters that will split to different projects | 18:51 |
| notmyname | eg admin guide has an object storage chapter that goes to swift, networking to neutron,e tc | 18:51 |
| notmyname | asettle: do you know to what extent those are cross-linked? | 18:52 |
| notmyname | hmm... maybe that doesn't matter too much for this conversation | 18:52 |
| asettle | Yis, that is true. Hmmm, I can't tell you off the top of my head, but I'm sure we could check that out. | 18:52 |
| notmyname | that comes down to the redirect links that the docs team is managing | 18:52 |
| asettle | But yes, they're good thoughts, something we need to be mindful of when we start | 18:53 |
| asettle | Yeah christ. | 18:53 |
| asettle | o.o | 18:53 |
| notmyname | and now I understand why that's been a big deal for you | 18:53 |
| asettle | Yep | 18:53 |
| notmyname | ah, I see now... | 18:54 |
| notmyname | so the proposed docs dir structure is important in that we'll end up needing a docs job for each | 18:54 |
| notmyname | even if it's just generating one page | 18:54 |
| notmyname | so that the redirects can provide continuity | 18:54 |
| notmyname | eg https://docs.openstack.org/cli-reference/swift.html | 18:55 |
| notmyname | TBH I'm not sure I knew that page even existed. that's one reason I think this is a good thing to do anyway, regardless of the unexpected time crunch on it | 18:57 |
| notmyname | asettle: where is https://etherpad.openstack.org/p/doc-migration-tracking linked? | 18:59 |
| notmyname | it's a good overview that helps understand what's going on | 19:00 |
| asettle | Second link in 'references' in the spec :) | 19:00 |
| asettle | Sorry, was doing the dishes | 19:00 |
| notmyname | I guess I missed it in the spec? | 19:00 |
| notmyname | ah yes :-) | 19:00 |
| asettle | notmyname: well, the idea is that we'll end up with one docs job - one to rule them all. | 19:00 |
| notmyname | yes. of course I'm a fan of that :-) | 19:01 |
| asettle | dhellmann might be able to explain more about the intricacies there | 19:01 |
| * dhellmann perks up his ears | 19:01 | |
| notmyname | well, i don't care as much about one docs *job* as much as one docs *tree* and one conf.py file | 19:01 |
| notmyname | no, the etherpad is clear | 19:01 |
| notmyname | asettle: ok, I gotta head to lunch now. thanks for taking the time to chat about this | 19:01 |
| dhellmann | yeah, I agree having these separate trees & jobs has gotten a bit silly | 19:01 |
| notmyname | dhellmann: I'll leave this in gerrit, too: my only concern/request is that the structure of the in-tree docs are not critical (or even known) to tools outisde of the repo. external tools should *only* be calling the build job | 19:03 |
| dhellmann | yeah | 19:03 |
| dhellmann | the reason for specifying the structure is that with one build job, we'll be publishing everything to one directory on the web server | 19:03 |
| dhellmann | but we still want landing pages like the list of install guides to be able to easily deep-link into those results | 19:04 |
| dhellmann | and specifying at least some top-level structure helps there | 19:04 |
| dhellmann | I have no particular affinity to these names. They seemed quite like what you proposed for swift, just without "guide" in the names of some of the dirs. | 19:04 |
| notmyname | sure, it's fine. but if/when something is changed or consolidated or whatver, that shouldn't break anything external as long as calling the job stays the same | 19:05 |
| dhellmann | oh, yeah, definitely | 19:05 |
| *** ganders has quit IRC | 19:05 | |
| dhellmann | the job is exactly the same as the one that publishes the /developer docs now, so it just calls "python setup.py build_sphinx" | 19:05 |
| dhellmann | and then rsyncs the output into place | 19:05 |
| notmyname | but for the time being, as I said earlier, the dir names seem find, especially given the need to have the existing links redirect to per-project artifacts | 19:06 |
| dhellmann | the *job* doesn't care what the structure is, it's those landing pages and redirects | 19:06 |
| dhellmann | right | 19:06 |
| clayg | notmyname: charz: I keep seeing this probetest fail on the community cluster - passed on my machine last night tho https://8b86aea46fb38e6450f2-0e5f4c086da474abc1df58826577db2f.ssl.cf1.rackcdn.com/472347/6794/probetests/console.txt | 19:06 |
| notmyname | and at some point in the future, `tox -e docs` is *much* better than assuming sphinx/python is used to build docs | 19:06 |
| dhellmann | if you have proposals for alternative names, that's all fine, we can go through them, too. | 19:06 |
| dhellmann | yeah, that's a legacy issue. I hesitated to convert to that right now because some projects do slightly different things for the tox docs target right now | 19:07 |
| dhellmann | one step at a time | 19:07 |
| notmyname | it's orthogonal | 19:07 |
| dhellmann | but I would also like to get to that point | 19:07 |
| notmyname | clayg: ack. i'll start a probe tests as I walk out for lunch | 19:08 |
| dhellmann | I think there was a target like that proposed for the golang interfaces, so we could do a separate migration to use the same or a similar name after the content is in place | 19:08 |
| dhellmann | that might even have been a makefile target, I don't remember off the top of my head | 19:08 |
| notmyname | yeah, IIRC | 19:09 |
| notmyname | ok, /me goes afk | 19:09 |
| dhellmann | ttyl | 19:09 |
| * asettle is out from here on in | 19:10 | |
| asettle | o/ | 19:10 |
| *** itlinux_ has quit IRC | 19:28 | |
| *** mvk has joined #openstack-swift | 19:41 | |
| *** itlinux_ has joined #openstack-swift | 19:44 | |
| clayg | rledisez: if you make the config option "devices_per_worker" and you set it to 10, and you have 24 devices... I think it's reasonable you get 3 workers. But each one should have *8* devices right? | 19:45 |
| clayg | you asked for 10 "devices_per_worker" - but you get 3 workers with 8 devices instead of two with 10 and one with 4 our something else useless and stupid? | 19:45 |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Negative data or parity args are invalid https://review.openstack.org/472394 | 19:51 |
| *** mingyu has quit IRC | 19:52 | |
| clayg | rledisez: i also don't really like how disk_per_worker means bigger == slower | 20:34 |
| clayg | faster is better, bigger number is better, it is known | 20:35 |
| notmyname | clayg: oh, probetests passed in my SAIO | 20:59 |
| *** lucasxu has quit IRC | 21:11 | |
| *** Sukhdev has quit IRC | 21:14 | |
| *** Sukhdev has joined #openstack-swift | 21:17 | |
| *** Sukhdev has quit IRC | 21:21 | |
| *** mat128 has quit IRC | 21:41 | |
| *** skudlik has quit IRC | 21:45 | |
| *** catintheroof has quit IRC | 21:50 | |
| *** joeljwright has joined #openstack-swift | 21:53 | |
| *** ChanServ sets mode: +v joeljwright | 21:53 | |
| *** chlong has quit IRC | 21:53 | |
| *** joeljwright has quit IRC | 21:55 | |
| *** timss has quit IRC | 22:04 | |
| *** timss has joined #openstack-swift | 22:04 | |
| *** spiette has joined #openstack-swift | 22:11 | |
| *** vint_bra has quit IRC | 22:30 | |
| *** vint_bra has joined #openstack-swift | 22:32 | |
| *** klamath_ has quit IRC | 22:41 | |
| *** mingyu has joined #openstack-swift | 22:53 | |
| *** mingyu has quit IRC | 22:58 | |
| *** vint_bra has quit IRC | 22:58 | |
| *** tonanhngo has quit IRC | 23:01 | |
| *** itlinux_ has quit IRC | 23:03 | |
| *** tonanhngo has joined #openstack-swift | 23:05 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 23:08 |
| *** tonanhngo has quit IRC | 23:10 | |
| *** kei_yama has joined #openstack-swift | 23:24 | |
| *** chsc has quit IRC | 23:36 | |
| *** jamielennox is now known as jamielennox|away | 23:57 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!