| *** rcernin has joined #openstack-swift | 00:00 | |
| *** mingyu has quit IRC | 00:01 | |
| *** NM has joined #openstack-swift | 00:18 | |
| *** NM has quit IRC | 00:26 | |
| *** oshritf__ has quit IRC | 00:27 | |
| *** _JZ_ has quit IRC | 00:34 | |
| *** rcernin has quit IRC | 00:51 | |
| *** rcernin has joined #openstack-swift | 00:54 | |
| *** rcernin has quit IRC | 00:55 | |
| *** rcernin has joined #openstack-swift | 00:55 | |
| *** sams-gleb has joined #openstack-swift | 01:54 | |
| *** sams-gleb has quit IRC | 01:59 | |
| *** dims_ has joined #openstack-swift | 02:14 | |
| *** dims has quit IRC | 02:14 | |
| *** clu_ has quit IRC | 02:14 | |
| *** dims_ has quit IRC | 02:47 | |
| *** dims has joined #openstack-swift | 02:50 | |
| *** yarkot has joined #openstack-swift | 03:31 | |
| openstackgerrit | Joshua White proposed openstack/swift: WIP Swift multinode; Do not merge https://review.openstack.org/426043 | 03:47 |
|---|---|---|
| openstackgerrit | Joshua White proposed openstack/swift: WIP Swift multinode; Do not merge https://review.openstack.org/426043 | 03:49 |
| *** tqtran has quit IRC | 03:55 | |
| *** sams-gleb has joined #openstack-swift | 03:57 | |
| *** sams-gleb has quit IRC | 04:01 | |
| *** mmmucky has quit IRC | 04:04 | |
| *** mmmucky has joined #openstack-swift | 04:04 | |
| *** chosafine has joined #openstack-swift | 04:08 | |
| mahatic | good morning | 04:15 |
| mattoliverau | mahatic: morning | 04:15 |
| mahatic | mattoliverau: o/ it was a holiday in India too yesterday, republic day for us | 04:15 |
| mattoliverau | \o/ twins! | 04:16 |
| mahatic | :D | 04:16 |
| *** klrmn has quit IRC | 04:18 | |
| *** chosafin_ has joined #openstack-swift | 04:20 | |
| *** mtreinish has quit IRC | 04:22 | |
| *** mtreinish has joined #openstack-swift | 04:23 | |
| *** chosafine has quit IRC | 04:23 | |
| *** psachin has joined #openstack-swift | 04:24 | |
| *** SkyRocknRoll has joined #openstack-swift | 04:26 | |
| *** ppai has joined #openstack-swift | 04:28 | |
| mahatic | oh progress/new bug in patch 419787 | 04:32 |
| patchbot | https://review.openstack.org/#/c/419787/ - swift - Better optimistic lock in get_hashes | 04:32 |
| *** ppai has quit IRC | 04:48 | |
| *** ppai has joined #openstack-swift | 05:01 | |
| *** ppai has quit IRC | 05:04 | |
| *** ppai has joined #openstack-swift | 05:05 | |
| *** ppai has quit IRC | 05:17 | |
| *** david-lyle has quit IRC | 05:46 | |
| *** winggundamth has joined #openstack-swift | 05:55 | |
| *** sams-gleb has joined #openstack-swift | 05:59 | |
| *** sams-gleb has quit IRC | 06:03 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Better optimistic lock in get_hashes https://review.openstack.org/419787 | 06:17 |
| *** silor has joined #openstack-swift | 06:20 | |
| *** silor1 has joined #openstack-swift | 06:25 | |
| *** silor has quit IRC | 06:26 | |
| *** silor1 is now known as silor | 06:26 | |
| *** ukaynar has quit IRC | 06:33 | |
| *** JimCheung has joined #openstack-swift | 06:43 | |
| *** JimCheung has quit IRC | 06:48 | |
| *** philipw has quit IRC | 07:02 | |
| *** philipw has joined #openstack-swift | 07:02 | |
| *** sams-gleb has joined #openstack-swift | 07:22 | |
| *** rcernin has quit IRC | 07:24 | |
| *** clu_ has joined #openstack-swift | 07:30 | |
| *** tanee is now known as tanee_away | 08:13 | |
| *** geaaru has joined #openstack-swift | 08:18 | |
| *** clu_ has quit IRC | 08:18 | |
| *** rledisez has joined #openstack-swift | 08:25 | |
| *** hseipp has joined #openstack-swift | 08:26 | |
| openstackgerrit | Merged openstack/python-swiftclient: 3.3.0 authors/changelog update https://review.openstack.org/425454 | 08:46 |
| *** kei_yama has quit IRC | 08:50 | |
| *** tanee_away is now known as tanee | 09:05 | |
| *** JimCheung has joined #openstack-swift | 09:15 | |
| *** JimCheung has quit IRC | 09:20 | |
| *** sams-gleb has quit IRC | 09:42 | |
| *** sams-gleb has joined #openstack-swift | 09:43 | |
| *** sams-gleb has quit IRC | 09:47 | |
| *** gabor_antal has joined #openstack-swift | 09:52 | |
| *** tqtran has joined #openstack-swift | 09:58 | |
| *** tqtran has quit IRC | 10:02 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 10:14 |
| *** jordanP has joined #openstack-swift | 10:21 | |
| *** sams-gleb has joined #openstack-swift | 10:28 | |
| *** mvk has quit IRC | 11:06 | |
| *** vint_bra has joined #openstack-swift | 11:08 | |
| *** mkrai has quit IRC | 11:10 | |
| *** vint_bra has quit IRC | 11:12 | |
| *** mkrai has joined #openstack-swift | 11:18 | |
| *** chlong has joined #openstack-swift | 11:24 | |
| *** sanchitmalhotra has quit IRC | 11:29 | |
| *** sanchitmalhotra has joined #openstack-swift | 11:29 | |
| *** NM has joined #openstack-swift | 11:37 | |
| *** oshritf__ has joined #openstack-swift | 11:52 | |
| *** catintheroof has joined #openstack-swift | 12:04 | |
| *** mvk has joined #openstack-swift | 12:04 | |
| *** oshritf__ has quit IRC | 12:29 | |
| *** vint_bra has joined #openstack-swift | 12:30 | |
| *** McMurlock1 has joined #openstack-swift | 12:33 | |
| *** geaaru has quit IRC | 12:35 | |
| *** acoles_ is now known as acoles | 12:40 | |
| *** glb1 has joined #openstack-swift | 12:50 | |
| *** tqtran has joined #openstack-swift | 13:00 | |
| *** psachin has quit IRC | 13:02 | |
| *** tqtran has quit IRC | 13:04 | |
| *** catintheroof has quit IRC | 13:05 | |
| *** catintheroof has joined #openstack-swift | 13:16 | |
| *** winggundamth has quit IRC | 13:53 | |
| *** tqtran has joined #openstack-swift | 14:01 | |
| *** tqtran has quit IRC | 14:05 | |
| *** sams-gleb has quit IRC | 14:19 | |
| *** sams-gleb has joined #openstack-swift | 14:20 | |
| *** geaaru has joined #openstack-swift | 14:22 | |
| *** sams-gleb has quit IRC | 14:25 | |
| *** sams-gleb has joined #openstack-swift | 14:25 | |
| *** sams-gleb has quit IRC | 14:29 | |
| *** Renich has quit IRC | 15:08 | |
| *** sams-gleb has joined #openstack-swift | 15:12 | |
| *** _JZ_ has joined #openstack-swift | 15:39 | |
| *** chsc has joined #openstack-swift | 15:45 | |
| *** mvk has quit IRC | 15:58 | |
| *** chsc has quit IRC | 16:03 | |
| openstackgerrit | Janie Richling proposed openstack/swift: WIP: Symlink support in container-sync https://review.openstack.org/421669 | 16:06 |
| clayg | acoles: how we looking bro? we on fleek? | 16:07 |
| *** oshritf has joined #openstack-swift | 16:07 | |
| acoles | clayg: I think so. I've not had time to go over the tests, just posted to gerrit though | 16:09 |
| clayg | i'm just reading cschwede's comment trying to understand why your test passed on master | 16:10 |
| acoles | clayg: I replied to that comment | 16:10 |
| clayg | he seems to be saying that the writeback in consolidate_hashes from the concurrent call is surprising - he expected modified to be False | 16:10 |
| acoles | clayg: oh, maybe that's sa different comment? | 16:10 |
| * cschwede is looking at the patch again | 16:11 | |
| acoles | clayg: I thought the test in my gist passed on master because mtime changed on the hashes.pkl even though content remained same | 16:11 |
| *** JimCheung has joined #openstack-swift | 16:14 | |
| acoles | I think that now if a suffix is invalidated then that is either (a) left in hashes.invalid journal or (b) committed to hashes.pkl in a way that any other process than the committer will notice (via 'updated') if it has read an older hashes.pkl | 16:15 |
| * acoles afk | 16:16 | |
| clayg | it passes 7/10 times on master anyway | 16:17 |
| cschwede | what?! | 16:18 |
| cschwede | why is that? | 16:18 |
| *** JimCheung has quit IRC | 16:18 | |
| clayg | cschwede: I'm printfing right now - i'm with you - how is this not failing? modified should be false! ;) | 16:23 |
| clayg | I mean... i can see it's not but I don't understand | 16:23 |
| clayg | i think it's just the performance regression | 16:30 |
| clayg | modified *is* false - the "other process" in non_local['called'] *is* eating the invalidation - but the outer caller of get_hashes does a second rehash anyway (we always do on master) so on the secon go round it picks up what it wrote last time in conslidate and rehashes to the correct state | 16:31 |
| clayg | pretty wild | 16:32 |
| *** donagh has joined #openstack-swift | 16:35 | |
| *** donagh has left #openstack-swift | 16:44 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Better optimistic lock in get_hashes https://review.openstack.org/419787 | 16:47 |
| notmyname | I woke up with a fever this morning, so I'm going to try to take it easy today. https://review.openstack.org/426301 has been proposed for the swiftclient release. | 16:48 |
| patchbot | patch 426301 - releases - swiftclient 3.3.0 release | 16:48 |
| notmyname | ping me if something else comes up, and I'll get back to you when I can | 16:48 |
| *** SkyRocknRoll has quit IRC | 16:48 | |
| clayg | oh... setting base to 7 in gerrit seems to show what you'd expect despite the rebase <shrug> | 16:49 |
| *** jordanP has quit IRC | 16:54 | |
| *** tqtran has joined #openstack-swift | 17:03 | |
| *** tqtran has quit IRC | 17:07 | |
| jrichli | tdasilva: looks like the SwiftStack cluster that runs gate jobs still runs with post-as-copy, so symlinks patch fails to run there | 17:07 |
| tdasilva | jrichli: can't send a patch to swiftstack cluster ;) | 17:08 |
| tdasilva | charz_: hello! we have recently changed the default for post-as-copy, but looks like swiftstack clusters have that value hardcoded, when you have a chance, can you change it? | 17:09 |
| notmyname | I'll follow up with charz_ | 17:10 |
| tdasilva | notmyname: thanks! and hope you get better soon | 17:11 |
| notmyname | thanks | 17:11 |
| *** chsc has joined #openstack-swift | 17:16 | |
| *** chsc has joined #openstack-swift | 17:16 | |
| *** rledisez has quit IRC | 17:17 | |
| notmyname | swiftclient 3.3.0 was tagged | 17:17 |
| *** tqtran has joined #openstack-swift | 17:21 | |
| *** Jeffrey4l_ has quit IRC | 17:24 | |
| *** Jeffrey4l_ has joined #openstack-swift | 17:24 | |
| *** JimCheung has joined #openstack-swift | 17:29 | |
| jrichli | notmyname: fever - that's aweful. I have had a cold the last couple days. And of course you know about poor kota_ with the flu | 17:30 |
| *** bikmak has quit IRC | 17:31 | |
| *** hseipp has quit IRC | 17:35 | |
| *** chlong has quit IRC | 18:03 | |
| *** klrmn has joined #openstack-swift | 18:08 | |
| *** mvk has joined #openstack-swift | 18:16 | |
| *** psachin has joined #openstack-swift | 18:19 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Fix race when consolidating new partition (WIP) https://review.openstack.org/426336 | 18:28 |
| acoles | clayg: don't give up! we will conquer hashing ;) | 18:44 |
| acoles | goodnight | 18:44 |
| *** geaaru has quit IRC | 18:47 | |
| *** ChubYann has joined #openstack-swift | 18:51 | |
| *** oshritf has quit IRC | 18:58 | |
| *** acoles is now known as acoles_ | 19:02 | |
| *** arch-nemesis has joined #openstack-swift | 19:08 | |
| *** McMurlock1 has quit IRC | 19:13 | |
| *** tqtran has quit IRC | 19:20 | |
| clayg | whooooo | 19:20 |
| *** silor has quit IRC | 19:23 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Fix race when consolidating new partition (WIP) https://review.openstack.org/426336 | 19:33 |
| *** psachin has quit IRC | 20:00 | |
| *** david-lyle has joined #openstack-swift | 20:01 | |
| *** oshritf has joined #openstack-swift | 20:02 | |
| *** varunsomani has joined #openstack-swift | 20:15 | |
| *** varunsomani has quit IRC | 20:16 | |
| *** ukaynar has joined #openstack-swift | 20:22 | |
| *** NM has quit IRC | 20:27 | |
| *** JimCheung has quit IRC | 20:44 | |
| *** caiobrentano has quit IRC | 20:45 | |
| clayg | tdasilva: so are you warming up on patch 425496 at all? | 20:53 |
| patchbot | https://review.openstack.org/#/c/425496/ - swift - Warn about using EC with isa_l_rs_vand and nparity... | 20:53 |
| clayg | I was thinking of kicking the tires - but to my understanding and notmyname's comments - it *seems* to faithfully reproduce what was agreeded to in the meeting | 20:54 |
| clayg | ... so unless we agreed to the wrong thing - I'm probably happy as long as it does what's on the tin | 20:54 |
| *** arch-nemesis has quit IRC | 21:01 | |
| *** Jeffrey4l_ has quit IRC | 21:35 | |
| *** Jeffrey4l_ has joined #openstack-swift | 21:41 | |
| MooingLemur | is liberasurecode_rs_vand a compatible drop in for jerasure_rs_vand? | 21:43 |
| MooingLemur | for an existing policy? | 21:43 |
| *** catintheroof has quit IRC | 21:44 | |
| *** catintheroof has joined #openstack-swift | 21:45 | |
| clayg | MooingLemur: no definately not | 21:48 |
| MooingLemur | roger | 21:48 |
| clayg | MooingLemur: maybe I should say "definately not intended to be; but to my knowledge no one has been crazy enough to try - because if it worked it'd be like breeding a dog with a cat and getting chicken" | 21:49 |
| clayg | i.e. YOU SHOULD TOTALLY TRY IT AND LET US KNOW HOW IT GOES | 21:49 |
| MooingLemur | delicious | 21:49 |
| *** catintheroof has quit IRC | 21:49 | |
| MooingLemur | yeah, I think it's quite unhappy with the attempt :) | 21:50 |
| clayg | ;) | 21:50 |
| MooingLemur | since this is a personal cluster that I'm converting from gentoo to centos, I'll just migrate the data manually away from jerasure. | 21:52 |
| MooingLemur | easy to get in gentoo, not supplied with centos without extra work so *shrug* | 21:52 |
| *** vint_bra has quit IRC | 21:54 | |
| *** _JZ_ has quit IRC | 21:55 | |
| zaitcev | Guys, is there a chance for some to review https://review.openstack.org/337297 in like a couple of hours? | 22:04 |
| patchbot | patch 337297 - swift - Add support to increase object ring partition power | 22:04 |
| zaitcev | I think it's fine, but acoles_ and mattoliverau caught things in it previously that I missed | 22:04 |
| zaitcev | Er | 22:05 |
| zaitcev | actually it was Clay | 22:05 |
| zaitcev | But anyway, I'm irrationally uneasy about landing it. | 22:05 |
| *** oshritf has quit IRC | 22:12 | |
| *** JimCheung has joined #openstack-swift | 22:12 | |
| clayg | i don't think i've ever even looked at it :'( | 22:23 |
| MooingLemur | clayg.clone() | 22:24 |
| zaitcev | "Patch Set 4: // (5 comments) // Still reading this over, I think it makes some tradeoffs over the epoch approach that are have some very promising merit - looking forward to talking with you more about it." | 22:24 |
| clayg | zaitcev: it looks like you and cschwede are feeling pretty good out it? | 22:24 |
| zaitcev | clayg: yes | 22:24 |
| clayg | sorry zaitcev, cschwede, mattoliverau - that's huge - that's 3 cores | 22:25 |
| zaitcev | well, I was sceptical originally. It seems complex and I hate additing code | 22:25 |
| clayg | If you guys are feeling good about it the only advice I probably have to offer is checklist stuff | 22:25 |
| zaitcev | ok | 22:25 |
| clayg | double check probetests - double check docs | 22:25 |
| clayg | yes code is the worst +2 | 22:26 |
| clayg | but... you can't make the impossible possible w/o code it seems :'( | 22:26 |
| zaitcev | I tested it exactly once. It was fine and I was able to add-remove devices and rebalance the ring afterwards without imposing too much of data movement, so at least that worked. | 22:27 |
| clayg | zaitcev: mattoliverau: cschwede: I don't think anyone cares if it goes in this week or next? If you want to stew on it go for it! tdasilva timburke or myself I'm sure would all *love* to look into it more! | 22:27 |
| clayg | but we can look at it on master too ;) | 22:27 |
| zaitcev | Today is Ocata cutoff | 22:27 |
| clayg | what!? | 22:27 |
| clayg | no it's not | 22:27 |
| clayg | zaitcev: I think notmyname is doing a python swiftclient thingy | 22:28 |
| zaitcev | Well, tdasilva told me so. | 22:28 |
| clayg | but we have some time for Octocat | 22:28 |
| clayg | tdasilva: !!! you sure!? | 22:28 |
| zaitcev | Hrm. Well, I needed to review it again anyway, so I don't I wasted any time today. | 22:28 |
| zaitcev | Sorry for the noise if so... | 22:29 |
| zaitcev | You know what's funny, I distinctly remember a talk about this in 2012. It was 5 years ago when some guy, I think his name was Luis, tried to adjust the partition power. | 22:30 |
| zaitcev | It was back when gholt was active | 22:30 |
| zaitcev | He was not smart enough to invent cschwede's relinker and realize that +1 is doable. | 22:31 |
| clayg | zaitcev: I think lots have people have realized you can do the 2^+1 thing - but figuring out what to do on disk and the process to get all the bits in place w/o loosing any was cschwede's invention AFAIK | 22:32 |
| clayg | cschwede is sort of a genius like that | 22:32 |
| *** ukaynar has quit IRC | 22:33 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Better optimistic lock in get_hashes https://review.openstack.org/419787 | 22:51 |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Fix race when consolidating new partition https://review.openstack.org/426336 | 22:51 |
| *** tqtran has joined #openstack-swift | 23:28 | |
| *** chsc has quit IRC | 23:31 | |
| *** sams-gleb has quit IRC | 23:39 | |
| notmyname | today (actually yesterday, it turns out) was the ocata cutoff for library releases | 23:59 |
| notmyname | the week of feb 13 is the cutoff for swift | 23:59 |
| notmyname | https://releases.openstack.org/ocata/schedule.html | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!