| *** haomaiwang has quit IRC | 00:01 | |
| *** haomaiwang has joined #openstack-swift | 00:01 | |
| ho_ | OMG! the patch was merged. acoles and cschwede, thanks for reviews and notmyname, thanks for your help :-) | 00:03 |
|---|---|---|
| openstackgerrit | Tim Burke proposed openstack/swift: Simplify policy-name validation slightly https://review.openstack.org/279333 | 00:09 |
| *** jamielennox is now known as jamielennox|away | 00:10 | |
| openstackgerrit | Merged openstack/swift-bench: authors/changelog update for 1.1. release https://review.openstack.org/279302 | 00:12 |
| *** clyps____ has joined #openstack-swift | 00:12 | |
| *** zul__ has joined #openstack-swift | 00:13 | |
| *** rickyrem1 has joined #openstack-swift | 00:14 | |
| *** daemontool_ has joined #openstack-swift | 00:14 | |
| *** clyps__ has quit IRC | 00:15 | |
| *** zul_ has quit IRC | 00:15 | |
| *** rickyrem has quit IRC | 00:15 | |
| *** daemontool has quit IRC | 00:15 | |
| *** jmccarthy has quit IRC | 00:15 | |
| *** jmccarthy has joined #openstack-swift | 00:15 | |
| *** zhill has joined #openstack-swift | 00:15 | |
| *** Jeffrey4l has joined #openstack-swift | 00:20 | |
| *** trifon has joined #openstack-swift | 00:26 | |
| *** cdelatte has quit IRC | 00:28 | |
| *** diazjf1 has quit IRC | 00:34 | |
| openstackgerrit | Merged openstack/swift-bench: version bump for 1.1. release https://review.openstack.org/279310 | 00:36 |
| openstackgerrit | Merged openstack/swift-bench: version bump to open 1.2 dev work https://review.openstack.org/279311 | 00:36 |
| openstackgerrit | Merged openstack/swift: Fixed params on autocreate_account docstring https://review.openstack.org/279264 | 00:38 |
| openstackgerrit | Merged openstack/swift: Removed unused parameter in server.py https://review.openstack.org/278441 | 00:39 |
| notmyname | woo hoo. I actually still can push tags for swift-bench. so I think the 1.1 release there is done. I'll check later to see if it triggered in -infra stuff for pushing to pypi | 00:49 |
| kota_ | good morning | 00:51 |
| kota_ | ho_: congrats, the RBAC patch merged :) | 00:51 |
| *** arthus has joined #openstack-swift | 00:52 | |
| *** zhill has quit IRC | 00:55 | |
| *** haomaiwang has quit IRC | 01:01 | |
| *** haomaiwang has joined #openstack-swift | 01:01 | |
| *** garthb has quit IRC | 01:03 | |
| openstackgerrit | Reedip proposed openstack/swift-bench: Remove support for py33/py26 https://review.openstack.org/264486 | 01:10 |
| *** gyee has joined #openstack-swift | 01:21 | |
| ho_ | kota_: thanks! actually one of the RBAC patches merged now :-) | 01:23 |
| kota_ | ho_: oh, you said "*one* of"!? | 01:26 |
| kota_ | ho_: how many patches are remaining? | 01:26 |
| ho_ | kota_: 4 patches i think. i will check it. | 01:27 |
| ho_ | kota_: patch 213608, patch 212810, patch 212825 and patch 253371. these patches are almost same as first one. | 01:29 |
| patchbot | ho_: https://review.openstack.org/#/c/213608/ - swift - Add functional test for access control (container ... | 01:29 |
| patchbot | ho_: https://review.openstack.org/#/c/212810/ - swift - Add functional test for access control (RBAC INFO)... | 01:29 |
| patchbot | ho_: https://review.openstack.org/#/c/212825/ - swift - Add functional test for access control (COPY) with... | 01:29 |
| patchbot | ho_: https://review.openstack.org/#/c/253371/ - swift - Improves RBAC related functional test | 01:29 |
| kota_ | hew | 01:30 |
| ho_ | kota_: and cschwede's improvement patch | 01:30 |
| kota_ | thanks for explanaiton | 01:31 |
| ho_ | kota_: thanks for asking :-) | 01:32 |
| *** jamielennox|away is now known as jamielennox | 01:34 | |
| *** arthus has quit IRC | 01:41 | |
| *** takashi has joined #openstack-swift | 01:44 | |
| *** lyrrad has quit IRC | 01:54 | |
| *** shakamunyi has joined #openstack-swift | 01:58 | |
| *** haomaiwang has quit IRC | 02:01 | |
| *** trifon has quit IRC | 02:02 | |
| *** haomaiwa_ has joined #openstack-swift | 02:02 | |
| *** shakamunyi has quit IRC | 02:12 | |
| *** asettle has quit IRC | 02:16 | |
| *** baojg has joined #openstack-swift | 02:24 | |
| *** arthus has joined #openstack-swift | 02:26 | |
| *** shakamunyi has joined #openstack-swift | 02:31 | |
| *** baojg has quit IRC | 02:36 | |
| *** baojg_ has joined #openstack-swift | 02:36 | |
| openstackgerrit | Michael Barton proposed openstack/swift: go: object engine abstraction layer https://review.openstack.org/278932 | 02:43 |
| redbo | oops | 02:47 |
| openstackgerrit | Michael Barton proposed openstack/swift: go: object engine abstraction layer https://review.openstack.org/278932 | 02:47 |
| *** diazjf has joined #openstack-swift | 02:48 | |
| *** natarej_ has quit IRC | 02:50 | |
| *** natarej has joined #openstack-swift | 02:50 | |
| *** sanchitmalhotra has joined #openstack-swift | 02:55 | |
| *** dmorita has quit IRC | 02:55 | |
| *** asettle has joined #openstack-swift | 02:57 | |
| *** dmorita has joined #openstack-swift | 02:59 | |
| *** haomaiwa_ has quit IRC | 03:01 | |
| *** diazjf has quit IRC | 03:01 | |
| *** haomaiwa_ has joined #openstack-swift | 03:02 | |
| *** asettle has quit IRC | 03:02 | |
| *** dmorita has quit IRC | 03:04 | |
| *** gyee has quit IRC | 03:09 | |
| *** dmorita has joined #openstack-swift | 03:10 | |
| *** chirag has quit IRC | 03:10 | |
| openstackgerrit | Merged openstack/swift: Make handoffs_first a more useful "mode" https://review.openstack.org/215867 | 03:26 |
| *** links has joined #openstack-swift | 03:32 | |
| *** nadeem has joined #openstack-swift | 03:35 | |
| *** arthus has quit IRC | 03:40 | |
| mahatic | good morning | 03:53 |
| ho_ | mahatic: morning! | 03:55 |
| mahatic | hello ho_ | 03:56 |
| ho_ | all: do you know why container-sync uses urllib2 for sending data? | 03:59 |
| *** haomaiwa_ has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-swift | 04:01 | |
| *** baojg_ has quit IRC | 04:03 | |
| *** baojg has joined #openstack-swift | 04:13 | |
| notmyname | good evening, world | 04:14 |
| *** vinsh_ has quit IRC | 04:16 | |
| mattoliverau | notmyname: evening | 04:16 |
| takashi | notmyname: good evening. | 04:16 |
| *** vinsh has joined #openstack-swift | 04:16 | |
| takashi | ho_: That is because it should send requests to another swift cluster (It can't use internal/direct client). | 04:17 |
| takashi | ho_: I suppose container-sync used python-swiftclient in very old version, but we changed it when we removed swift's dependency on python-swiftclient | 04:18 |
| pdardeau | mattoliverau: good evening | 04:19 |
| *** asettle has joined #openstack-swift | 04:20 | |
| *** arthus has joined #openstack-swift | 04:26 | |
| jrichli | hello all. patch 218490 only needs one more +2. It was a feature talked about at Vancouver summit. | 04:27 |
| patchbot | jrichli: https://review.openstack.org/#/c/218490/ - swift - Automatic refresh of memcache config settings | 04:27 |
| notmyname | ah yes. that one :-) | 04:27 |
| ho_ | takashi: thanks. what do you think is there any possibility to change the library? | 04:28 |
| mattoliverau | I believe I've already +2'ed it :) | 04:28 |
| ho_ | notmyname: good evening! | 04:28 |
| notmyname | ho_: what do you want to change it to? | 04:28 |
| ho_ | notmyname: i would like to change the library to ie pycurl. | 04:28 |
| jrichli | yes, mattoliverau. Thank you! | 04:28 |
| notmyname | ho_: why? | 04:28 |
| ho_ | notmyname: now i have been investigating container-sync related problem and know the library can not handle 100-continue | 04:29 |
| ho_ | notmyname: s/know/knew/ | 04:30 |
| takashi | ho_: I know that may cause a heavy load on network. | 04:31 |
| notmyname | ho_: ah. thanks, I was jsut curious | 04:31 |
| takashi | ho_: The proxy server has some logics to remove duplicated PUT from container-sync, but container-sync always send object data to the proxy-server, because it can't handle 100-continue, right. | 04:32 |
| takashi | ho_: I talked about this problem with eranrom at Tokyo summit, and he told me that he also faced that problem. | 04:33 |
| takashi | ho_: He told me that he was planning to add independent HEAD request before PUT insted of replacing library, but I think you can exchange your ideas about that. | 04:34 |
| openstackgerrit | janonymous proposed openstack/swift: test for txn-id capture https://review.openstack.org/278176 | 04:34 |
| ho_ | takashi: yeah, really? thanks for good information. i will ask eranrom. hehe. i don't like additional head req... :-) | 04:35 |
| ho_ | takashi: i will submit my image as code to share the problem. thanks again! | 04:36 |
| takashi | ho_: I agree with you. That may be helpful with python-swiftclient. I suppose now it also does not support 100-continue now. | 04:37 |
| jrichli | btw, I really learned a lot from the LCA talks from mattoliverau and notmyname! They were great. we need more swift deep dives! | 04:37 |
| zigo | notmyname: Thanks in advance if you work on the backport of the CVE-2016-0738 for Jessie. That would really be helpful for us. | 04:37 |
| notmyname | jrichli: thanks. | 04:37 |
| *** rickyrem1 has quit IRC | 04:38 | |
| notmyname | jrichli: yeah, I've had some other people ask for more of those (eg support and sales engineers at swiftstack). I'd like to do more, but it's one of those things that always seems to be somewhere lower on the TODO list | 04:38 |
| notmyname | zigo: speaking of stuff lower on the TODO list... ;-) | 04:38 |
| notmyname | zigo: but seriously, I did add that to my list :-) | 04:39 |
| zigo | :) | 04:39 |
| mattoliverau | jrichli: thanks :) | 04:39 |
| notmyname | takashi: yeah, we lost 100-continue support when we ported over to use requests :-( | 04:39 |
| notmyname | takashi: it was a big loss, I think. (but we did gain some other things that were very important) | 04:39 |
| notmyname | jrichli: did it make sense? that was the first time I had given a description of the on-disk EC layout in a presentation format | 04:40 |
| notmyname | jrichli: previously I've done it 1:1 on whiteboards, and I think my presentation version needs a little work | 04:41 |
| mattoliverau | it did, just make everything polynomials! | 04:41 |
| notmyname | mattoliverau: simple! | 04:41 |
| *** nadeem has quit IRC | 04:42 | |
| jrichli | notmyname: yes. I do have to admit, I am still conceptualizing the fragment archives enough to *really* understand how we can do ranged reads. but i get it at a high level. | 04:45 |
| takashi | notmyname: I understand why we did it. :-) Nothing can be done only without taking its cons, but I sometimes hope if requests supported 100-continue... :-( | 04:45 |
| takashi | s/only// | 04:45 |
| notmyname | jrichli: modulo math. we know the size of the segment in the FA and how long the total FA is. so therefore we can scale the request range to get the right segments out of the FA. and then we apply the real range to that (with the right initial offset) and you get the right range to respond with | 04:47 |
| notmyname | takashi: YES!!! | 04:47 |
| notmyname | jrichli: so we can do that on reads. but I want to apply similar math to the read path so we can get the total FA size up front to do the right fallocate. right now EC objects dont' fallocate | 04:49 |
| jrichli | right, i actually thought about digging into that. is on the list :-) I see somebody is assigned to that. | 04:50 |
| notmyname | https://bugs.launchpad.net/swift/+bug/1532008 | 04:51 |
| openstack | Launchpad bug 1532008 in OpenStack Object Storage (swift) "EC policy doesn't fallocate objects" [Undecided,New] - Assigned to Azhagu Selvan SP (tamizhgeek) | 04:51 |
| notmyname | tamizh_geek: are you still working on this? | 04:51 |
| jrichli | it was just last week when he took it :-) but I will try to learn enough to be able to review if nothing else! | 04:53 |
| *** baojg has quit IRC | 04:53 | |
| notmyname | jrichli: range magic for EC https://github.com/openstack/swift/blob/master/swift/proxy/controllers/obj.py#L1477-L1585 | 04:54 |
| jrichli | cool, thx! | 04:58 |
| * jrichli is calling it a night | 05:00 | |
| *** haomaiwang has quit IRC | 05:01 | |
| *** haomaiwa_ has joined #openstack-swift | 05:01 | |
| *** klrmn1 has quit IRC | 05:06 | |
| *** proteusguy__ has joined #openstack-swift | 05:09 | |
| *** charz_ has quit IRC | 05:09 | |
| *** hrou_ has quit IRC | 05:10 | |
| *** proteusguy_ has quit IRC | 05:10 | |
| *** arthus has quit IRC | 05:10 | |
| *** hrou has joined #openstack-swift | 05:10 | |
| *** changbl has quit IRC | 05:11 | |
| *** wbhuber has quit IRC | 05:11 | |
| *** stevemar has quit IRC | 05:11 | |
| *** wbhuber has joined #openstack-swift | 05:12 | |
| *** charz has joined #openstack-swift | 05:12 | |
| *** stevemar has joined #openstack-swift | 05:13 | |
| *** changbl has joined #openstack-swift | 05:13 | |
| *** arthus has joined #openstack-swift | 05:14 | |
| *** Jeffrey4l has quit IRC | 05:19 | |
| *** chlong has quit IRC | 05:19 | |
| *** garthb has joined #openstack-swift | 05:20 | |
| *** chlong has joined #openstack-swift | 05:33 | |
| openstackgerrit | Alexandra Settle proposed openstack/swift: WIP: New operational procedures guide https://review.openstack.org/278245 | 05:33 |
| openstackgerrit | Alexandra Settle proposed openstack/swift: WIP: New operational procedures guide https://review.openstack.org/278245 | 05:40 |
| asettle | mattoliverau: if you're around ^^ a note for that, for reviewing purposes, I've found a way to start splitting it out :) | 05:41 |
| asettle | notmyname for you too ^ | 05:41 |
| notmyname | ack | 05:41 |
| *** Jeffrey4l has joined #openstack-swift | 05:41 | |
| *** chlong has quit IRC | 05:42 | |
| notmyname | for those of you who like to think on concurrency stuff ( clayg redbo acoles_ ), do you think the lack of a green version of os.path causes much/any harm? so if we do os.path.isfile or .getmtime, the stat of the file ends up being non-green, right? won't that block the whole process/ | 05:44 |
| notmyname | mattoliverau: had to put a -1 on your existing +2 for patch 218490 | 05:46 |
| patchbot | notmyname: https://review.openstack.org/#/c/218490/ - swift - Automatic refresh of memcache config settings | 05:46 |
| notmyname | ahale: there's some back and forth "is this good for ops" discussion on that patch^. I'd appreciate any feedback you have | 05:47 |
| *** ppai has joined #openstack-swift | 05:47 | |
| redbo | notmyname: Yeah, all of those call stat, which can block. | 05:54 |
| mattoliverau | notmyname: how dare you! :P | 05:54 |
| *** chlong has joined #openstack-swift | 05:54 | |
| notmyname | redbo: so eg an auto ring reload could theoretically block that whole proxy server process | 05:55 |
| *** mattoliverau has quit IRC | 05:55 | |
| notmyname | (and the proposed memcache.conf reload) | 05:56 |
| redbo | Yeah. I mean it will block the whole process, just usually not for very long. But if the disk is bad or slow, the whole thing will wait for it. | 05:56 |
| notmyname | yeah :-( | 05:57 |
| notmyname | in a quick test, it doesn't seem like a green version exists | 05:57 |
| notmyname | I wonder how possible it is to make one | 05:58 |
| redbo | You'd have to send it off to a threadpool. But then I think probably just sending the whole ring reload into a thread would be easier and not too terrible. | 06:00 |
| *** haomaiwa_ has quit IRC | 06:01 | |
| notmyname | no calls to isfile or isdir in the object server. just the replicator and updater processes | 06:01 |
| *** haomaiwang has joined #openstack-swift | 06:01 | |
| notmyname | so while it could make the background stuff hang (which admittedly is a known problem in general), at least it's not causing the data path to slow down | 06:02 |
| notmyname | /redbo thinks, "who cares about your silly python object servers..." | 06:02 |
| *** asettle has quit IRC | 06:03 | |
| redbo | haha... well, I'm tired of fighting with those kinds of problems. | 06:05 |
| notmyname | oh? are we doing erlang now? ;-) | 06:05 |
| notmyname | rust? | 06:05 |
| redbo | No, I mean process locking up problems. Go is working great. But if everyone wanted to try rust instead, I'd be down. | 06:06 |
| notmyname | :-) | 06:06 |
| *** garthb has quit IRC | 06:10 | |
| *** chlong has quit IRC | 06:11 | |
| *** zhill has joined #openstack-swift | 06:12 | |
| *** asettle has joined #openstack-swift | 06:13 | |
| *** asettle has quit IRC | 06:15 | |
| *** ChubYann has quit IRC | 06:19 | |
| * notmyname goes to bed | 06:19 | |
| *** chlong has joined #openstack-swift | 06:34 | |
| jith_ | hi all.. i have three disk each one is a replica, i reduced the replica count from 3 to 2 and made one of the disk weight as 0.. and i removed the third disk(mpathf) from the cluster.. but when i try to unmount it.. it says device is busy... when i do lsof --> out put is swift-obj 28114 root 3r REG 254,14 174850048 1291624466 /srv/node/mpathf/objects/390/c07/6183d7bb47899405 | 06:35 |
| jith_ | 63976fd3a519ec07/1447398087.61070.data | 06:35 |
| jith_ | what it means??? | 06:35 |
| jith_ | is it safe to remove the disk | 06:37 |
| *** arthus has quit IRC | 06:40 | |
| *** arthus has joined #openstack-swift | 06:41 | |
| *** silor has joined #openstack-swift | 06:41 | |
| *** silor1 has joined #openstack-swift | 06:44 | |
| *** silor has quit IRC | 06:46 | |
| *** silor1 is now known as silor | 06:46 | |
| *** haomaiwang has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-swift | 07:01 | |
| *** matt6434 has joined #openstack-swift | 07:02 | |
| *** cbartz has joined #openstack-swift | 07:02 | |
| *** chlong has quit IRC | 07:03 | |
| *** matt6434 is now known as mattoliverau | 07:04 | |
| *** ChanServ sets mode: +v mattoliverau | 07:04 | |
| mattoliverau | I'm back | 07:04 |
| openstackgerrit | janonymous proposed openstack/swift: Fix missing txn_id logs in GreenAsyncPile's spawned functions https://review.openstack.org/278176 | 07:12 |
| janonymous | I am sorry for the delay in this patch, please review the same :) | 07:16 |
| janonymous | i am not sure though i merged it correctly :D | 07:18 |
| jith_ | mattoliverau: any suggestion?? | 07:25 |
| jith_ | i have three disk each one is a replica, i reduced the replica count from 3 to 2 and made one of the disk weight as 0.. and i removed the third disk(mpathf) from the cluster.. but when i try to unmount it.. it says device is busy... when i do lsof --> out put is swift-obj 28114 root 3r REG 254,14 174850048 1291624466 /srv/node/mpathf/objects/390/c07/6183d7bb47899405 | 07:25 |
| jith_ | what it means??? | 07:25 |
| jith_ | is it safe to remove the disk | 07:25 |
| *** zaitcev has quit IRC | 07:40 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** 64MAA2KZJ has joined #openstack-swift | 08:01 | |
| openstackgerrit | janonymous proposed openstack/swift: Fix missing txn_id logs in GreenAsyncPile's spawned functions https://review.openstack.org/278176 | 08:08 |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: WIP: Fix container-sync behavior https://review.openstack.org/279417 | 08:13 |
| *** trifon has joined #openstack-swift | 08:14 | |
| *** esker has joined #openstack-swift | 08:24 | |
| *** esker has quit IRC | 08:29 | |
| *** arthus has quit IRC | 08:39 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: WIP: Fix container-sync behavior https://review.openstack.org/279417 | 08:42 |
| *** kei_yama has quit IRC | 08:44 | |
| *** ig0r_ has quit IRC | 08:45 | |
| *** esker has joined #openstack-swift | 08:45 | |
| *** arthus has joined #openstack-swift | 08:45 | |
| *** jmccarthy1 has quit IRC | 08:48 | |
| *** esker has quit IRC | 08:49 | |
| *** acoles_ is now known as acoles | 08:55 | |
| *** jmccarthy1 has joined #openstack-swift | 08:55 | |
| acoles | ho_: congrats on RBAC patch! | 08:57 |
| ho_ | acoles: thanks! | 09:00 |
| *** 64MAA2KZJ has quit IRC | 09:01 | |
| *** haomaiwang has joined #openstack-swift | 09:01 | |
| ho_ | acoles: i don't know how to express my feeling even if i can use japanese :-) | 09:01 |
| acoles | ho_: go ahead and use Japanese :) | 09:01 |
| ho_ | acoles: :-) | 09:02 |
| openstackgerrit | Mustafa ÇELİK proposed openstack/swift: Extra X-Timestamp are removed from error responses https://review.openstack.org/268163 | 09:05 |
| *** esker has joined #openstack-swift | 09:06 | |
| *** takashi has quit IRC | 09:09 | |
| *** esker has quit IRC | 09:11 | |
| *** mvk has joined #openstack-swift | 09:25 | |
| *** esker has joined #openstack-swift | 09:27 | |
| *** Jeffrey4l has quit IRC | 09:28 | |
| *** esker has quit IRC | 09:32 | |
| openstackgerrit | Peter Lisák proposed openstack/swift: Change schedule priority of daemon/server in config https://review.openstack.org/238799 | 09:39 |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift: Remove execute permissions from doc files and swift.conf-sample https://review.openstack.org/265145 | 09:39 |
| openstackgerrit | Christian Schwede proposed openstack/swift: Start object-auditing with last partition after restart https://review.openstack.org/279440 | 09:44 |
| *** esker has joined #openstack-swift | 09:48 | |
| *** joeljwright has joined #openstack-swift | 09:50 | |
| *** ChanServ sets mode: +v joeljwright | 09:50 | |
| acoles | ho_: I think you should perhaps treat yourself to a some fine Japanese whisky this evening :-) | 09:53 |
| *** esker has quit IRC | 09:53 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift: Remove execute permissions from codes and tests without "#! /usr/bin/env python" https://review.openstack.org/279441 | 09:53 |
| *** mvk has quit IRC | 09:59 | |
| *** joeljwright has quit IRC | 09:59 | |
| *** haomaiwang has quit IRC | 10:01 | |
| *** haomaiwang has joined #openstack-swift | 10:01 | |
| *** stantonnet has quit IRC | 10:02 | |
| *** stantonnet has joined #openstack-swift | 10:06 | |
| acoles | janonymous: so is patch 278176 now your squashed version for review? | 10:06 |
| patchbot | acoles: https://review.openstack.org/#/c/278176/ - swift - Fix missing txn_id logs in GreenAsyncPile's spawne... | 10:06 |
| acoles | looks that way | 10:07 |
| *** ahale has quit IRC | 10:08 | |
| *** ahale has joined #openstack-swift | 10:08 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift: Remove execute permission from swift/cli/ringbuilder.py https://review.openstack.org/279449 | 10:08 |
| *** esker has joined #openstack-swift | 10:09 | |
| *** dmorita has quit IRC | 10:10 | |
| *** ig0r_ has joined #openstack-swift | 10:13 | |
| *** esker has quit IRC | 10:14 | |
| *** dmorita has joined #openstack-swift | 10:15 | |
| ho_ | acoles: that's a perfect idea :-) | 10:16 |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: WIP: Fix container-sync behavior https://review.openstack.org/279417 | 10:18 |
| ho_ | eranrom: oshritf: could you please check my approach to solve bug 1277223 ? | 10:22 |
| openstack | bug 1277223 in OpenStack Object Storage (swift) "container-sync: multiple sync PUTs per user PUT, then remote cluster syncs back to sending cluster" [Undecided,Confirmed] https://launchpad.net/bugs/1277223 - Assigned to Eran Rom (eranr) | 10:22 |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift: Remove execute permission from swift/cli/ringbuilder.py https://review.openstack.org/279449 | 10:24 |
| openstackgerrit | Ondřej Nový proposed openstack/swift: Additional info log message for drive-audit https://review.openstack.org/263691 | 10:26 |
| *** daemontool has joined #openstack-swift | 10:28 | |
| *** daemontool_ has quit IRC | 10:31 | |
| *** openstackgerrit has quit IRC | 10:32 | |
| *** mvk has joined #openstack-swift | 10:32 | |
| *** openstackgerrit has joined #openstack-swift | 10:32 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift: Remove execute permission from swift/cli/ringbuilder.py https://review.openstack.org/279449 | 10:38 |
| *** ho_ has quit IRC | 10:40 | |
| *** joeljwright has joined #openstack-swift | 10:41 | |
| *** ChanServ sets mode: +v joeljwright | 10:41 | |
| *** wer has quit IRC | 10:42 | |
| *** wer has joined #openstack-swift | 10:43 | |
| *** haomaiwang has quit IRC | 11:01 | |
| *** haomaiwang has joined #openstack-swift | 11:01 | |
| *** arthus has quit IRC | 11:07 | |
| *** ppai has quit IRC | 11:15 | |
| *** zhill has quit IRC | 11:17 | |
| *** sileht has quit IRC | 11:20 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Start object-auditing with last partition after restart https://review.openstack.org/279440 | 11:22 |
| *** sanchitmalhotra has quit IRC | 11:29 | |
| *** marcusvrn_ has joined #openstack-swift | 11:37 | |
| *** sileht has joined #openstack-swift | 11:38 | |
| *** slo_ has joined #openstack-swift | 11:40 | |
| openstackgerrit | Merged openstack/swift: Fix typos in comments https://review.openstack.org/279233 | 11:41 |
| *** jith_ has quit IRC | 11:43 | |
| *** baffle___ is now known as baffle | 11:45 | |
| *** ig0r_ has quit IRC | 11:48 | |
| *** ig0r_ has joined #openstack-swift | 11:49 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift: Remove '#! /usr/bin/env python' from unexecutable files https://review.openstack.org/279493 | 11:50 |
| *** ankurgupta1 has joined #openstack-swift | 11:56 | |
| *** haomaiwang has quit IRC | 12:01 | |
| *** haomaiwang has joined #openstack-swift | 12:01 | |
| *** links has quit IRC | 12:10 | |
| *** cdelatte has joined #openstack-swift | 12:11 | |
| *** dmorita has quit IRC | 12:13 | |
| *** dmorita has joined #openstack-swift | 12:16 | |
| *** dmorita has quit IRC | 12:20 | |
| *** cdelatte has quit IRC | 12:20 | |
| *** Jeffrey4l has joined #openstack-swift | 12:22 | |
| *** ankurgupta1 has quit IRC | 12:23 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Start object-auditing with last partition after restart https://review.openstack.org/279440 | 12:32 |
| *** daemontool has quit IRC | 12:36 | |
| *** cdelatte has joined #openstack-swift | 12:46 | |
| *** daemontool has joined #openstack-swift | 12:50 | |
| *** haomaiwang has quit IRC | 13:01 | |
| *** haomaiwa_ has joined #openstack-swift | 13:01 | |
| *** jordanP has joined #openstack-swift | 13:03 | |
| *** jordanP has quit IRC | 13:04 | |
| *** jordanP has joined #openstack-swift | 13:04 | |
| *** cschwede has left #openstack-swift | 13:10 | |
| *** cschwede has joined #openstack-swift | 13:11 | |
| *** ChanServ sets mode: +v cschwede | 13:11 | |
| *** thumpba has joined #openstack-swift | 13:13 | |
| *** thumpba has quit IRC | 13:14 | |
| *** thumpba has joined #openstack-swift | 13:14 | |
| *** cbartz has left #openstack-swift | 13:17 | |
| *** thumpba has quit IRC | 13:19 | |
| tdasilva | good morning | 13:25 |
| *** asettle has joined #openstack-swift | 13:31 | |
| asettle | notmyname: you around? :) | 13:32 |
| *** asettle has quit IRC | 13:43 | |
| *** km has quit IRC | 13:49 | |
| *** petertr7_away is now known as petertr7 | 13:49 | |
| openstackgerrit | Peter Lisák proposed openstack/swift: Change schedule priority of daemon/server in config https://review.openstack.org/238799 | 13:54 |
| *** haomaiwa_ has quit IRC | 14:01 | |
| *** haomaiwa_ has joined #openstack-swift | 14:02 | |
| *** MVenesio has joined #openstack-swift | 14:05 | |
| *** haypo has joined #openstack-swift | 14:14 | |
| *** daemontool has quit IRC | 14:16 | |
| *** petertr7 is now known as petertr7_away | 14:23 | |
| openstackgerrit | Victor Stinner proposed openstack/swift: Port get_hmac() and hash_path() to Python 3 https://review.openstack.org/236998 | 14:28 |
| haypo | acoles: hi. i read again my change for hmac, and i think that i found a solution to respect your requirements ;) | 14:28 |
| haypo | acoles: sorry, i'm grumpy, it was a bad week for me :-( | 14:29 |
| haypo | (sorry for being grumpy) | 14:29 |
| acoles | haypo: np we all get grumpy sometimes ;) | 14:29 |
| haypo | acoles: with my new change, hash prefix/suffix are raw bytes as written in the configuration file | 14:29 |
| haypo | acoles: i used the latin1 encoding for the temporary file on disk => unicode conversion, and then unicode => bytes hash prefix/suffix | 14:30 |
| haypo | acoles: latin1 is kind of magic. decoding from latin1 never fails, and later you retrieve your "raw" bytes unchanged | 14:30 |
| acoles | haypo: just so you know, i am unlikely to cycle back to it this week and OOO early next week | 14:30 |
| acoles | haypo: sounds magic | 14:31 |
| haypo | torgomatic, cschwede : would you mind reviewing my fixed https://review.openstack.org/236998 change for hmac? | 14:32 |
| haypo | acoles: don't worry, you already helped on my other py3 patches. thanks, i appreciate your effort ;) | 14:33 |
| haypo | crap. the test fails :-p | 14:34 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Port get_hmac() and hash_path() to Python 3 https://review.openstack.org/236998 | 14:38 |
| *** daemontool has joined #openstack-swift | 14:40 | |
| *** haomaiwa_ has quit IRC | 15:01 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Bump PyECLib requirement to >= 1.2.0 https://review.openstack.org/278944 | 15:01 |
| *** haomaiwang has joined #openstack-swift | 15:01 | |
| *** changbl has quit IRC | 15:03 | |
| *** mrmoje has joined #openstack-swift | 15:21 | |
| *** _JZ_ has joined #openstack-swift | 15:32 | |
| *** changbl has joined #openstack-swift | 15:32 | |
| *** mragupat has joined #openstack-swift | 15:32 | |
| openstackgerrit | Peter Lisák proposed openstack/swift: Change schedule priority of daemon/server in config https://review.openstack.org/238799 | 15:33 |
| *** changbl has quit IRC | 15:37 | |
| *** changbl has joined #openstack-swift | 15:38 | |
| *** bjkeller has joined #openstack-swift | 15:43 | |
| *** dmsimard has joined #openstack-swift | 15:43 | |
| *** petertr7_away is now known as petertr7 | 15:45 | |
| *** klrmn has joined #openstack-swift | 15:47 | |
| *** MVenesio has quit IRC | 15:53 | |
| notmyname | good morning | 15:58 |
| notmyname | acoles: nice. patch 276823 landed | 15:59 |
| patchbot | notmyname: https://review.openstack.org/#/c/276823/ - openstack-infra/project-config - Add job gate-swift-tox-func-in-process-fast-post (MERGED) | 15:59 |
| acoles | notmyname: yup :) do you know if there is an easy way to list all job votes for a specific job over a period of time (i.e. analyse the failure rate of the job before making it voting)? I'm not ex[ecting it to fail of course! | 16:00 |
| *** haomaiwang has quit IRC | 16:01 | |
| *** haomaiwang has joined #openstack-swift | 16:01 | |
| *** barker has joined #openstack-swift | 16:01 | |
| *** klrmn has quit IRC | 16:02 | |
| *** trifon has quit IRC | 16:16 | |
| *** ig0r_ has quit IRC | 16:20 | |
| *** jordanP has quit IRC | 16:20 | |
| *** arch-nemesis has joined #openstack-swift | 16:20 | |
| openstackgerrit | Victor Stinner proposed openstack/swift: Parse HTTP as bytes on Python 3 https://review.openstack.org/237042 | 16:21 |
| *** links has joined #openstack-swift | 16:23 | |
| *** jmccarthy1 has quit IRC | 16:29 | |
| *** MVenesio has joined #openstack-swift | 16:31 | |
| openstackgerrit | Victor Stinner proposed openstack/swift: Port get_hmac() and hash_path() to Python 3 https://review.openstack.org/236998 | 16:33 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Parse HTTP as bytes on Python 3 https://review.openstack.org/237042 | 16:33 |
| openstackgerrit | Jonathan Hinson proposed openstack/swift: WIP Conditional GETs Fix https://review.openstack.org/257603 | 16:35 |
| acoles | pchng: sorry, I am not going to get to your patch this week but I have it on my todo list | 16:38 |
| openstackgerrit | Ben Keller proposed openstack/swift: Headers now use a case incensitive dictionary https://review.openstack.org/274906 | 16:44 |
| *** mrmoje has quit IRC | 16:44 | |
| *** nadeem has joined #openstack-swift | 16:52 | |
| *** zaitcev has joined #openstack-swift | 16:53 | |
| *** ChanServ sets mode: +v zaitcev | 16:53 | |
| *** haomaiwang has quit IRC | 17:01 | |
| *** haomaiwang has joined #openstack-swift | 17:01 | |
| *** lyrrad has joined #openstack-swift | 17:05 | |
| tdasilva | notmyname: hi, wasn't there a matrix of swift released versions and openstack releases? | 17:05 |
| notmyname | tdasilva: yes. it's in the changelog and also at https://wiki.openstack.org/wiki/Swift/version_map | 17:08 |
| *** MVenesio has quit IRC | 17:09 | |
| tdasilva | notmyname: thanks! | 17:11 |
| *** chsc has joined #openstack-swift | 17:12 | |
| *** mvk has quit IRC | 17:12 | |
| *** gyee has joined #openstack-swift | 17:14 | |
| *** pgbridge has quit IRC | 17:15 | |
| *** trifon has joined #openstack-swift | 17:15 | |
| *** Jeffrey4l has quit IRC | 17:15 | |
| *** openstackgerrit has quit IRC | 17:17 | |
| *** openstackgerrit has joined #openstack-swift | 17:17 | |
| *** joeljwright has quit IRC | 17:22 | |
| *** mrmoje has joined #openstack-swift | 17:34 | |
| *** pgbridge has joined #openstack-swift | 17:38 | |
| *** MVenesio has joined #openstack-swift | 17:39 | |
| *** MVenesio has quit IRC | 17:39 | |
| openstackgerrit | Victor Stinner proposed openstack/swift: Port swob to Python 3 https://review.openstack.org/279704 | 17:56 |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: Drop testtools from test-requirements.txt https://review.openstack.org/253678 | 17:58 |
| *** links has quit IRC | 17:58 | |
| *** nadeem has quit IRC | 17:58 | |
| *** nadeem has joined #openstack-swift | 17:59 | |
| *** haomaiwang has quit IRC | 18:01 | |
| *** haomaiwa_ has joined #openstack-swift | 18:01 | |
| *** MVenesio has joined #openstack-swift | 18:02 | |
| haypo | cschwede: you shoud not fuzz ConfigParser with random bytes :-D (see my reply on https://review.openstack.org/#/c/236998/) -- i understand that my patch works, just don't use "random bytes" ;) | 18:06 |
| patchbot | haypo: https://review.openstack.org/#/c/236998/ - swift - Port get_hmac() and hash_path() to Python 3 | 18:06 |
| haypo | have a nice week-end ;) | 18:07 |
| *** aunnam has joined #openstack-swift | 18:07 | |
| *** zhill has joined #openstack-swift | 18:08 | |
| *** ChubYann has joined #openstack-swift | 18:09 | |
| *** haomaiwa_ has quit IRC | 18:10 | |
| *** esker has joined #openstack-swift | 18:10 | |
| clayg | I'm *always* grumpy | 18:10 |
| *** haomaiwa_ has joined #openstack-swift | 18:11 | |
| *** petertr7 is now known as petertr7_away | 18:11 | |
| *** haomaiwa_ has quit IRC | 18:17 | |
| *** haomaiwa_ has joined #openstack-swift | 18:18 | |
| *** dmorita has joined #openstack-swift | 18:20 | |
| clayg | I've never even *heard* of "Depends-On" | 18:20 |
| clayg | it's not documented here -> https://wiki.openstack.org/wiki/GitCommitMessages | 18:21 |
| *** dmorita has quit IRC | 18:21 | |
| clayg | acoles: where'd you hear about "Depends-On" | 18:22 |
| acoles | clayg: http://docs.openstack.org/infra/manual/developers.html#cross-repository-dependencies | 18:23 |
| *** openstack has joined #openstack-swift | 18:26 | |
| -asimov.freenode.net- [freenode-info] channel flooding and no channel staff around to help? Please check with freenode support: http://freenode.net/faq.shtml#gettinghelp | 18:26 | |
| acoles | clayg: some (but apparently not all) of those wiki pages got made obsolete e.g. https://wiki.openstack.org/wiki/Obsolete:GerritJenkinsGit | 18:27 |
| clayg | lol @ "do not create a cycle" | 18:29 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Enable object server to return non-durable data https://review.openstack.org/215276 | 18:29 |
| clayg | *don't cross the beams!" | 18:29 |
| acoles | have a good weekend! | 18:32 |
| *** bjkeller has quit IRC | 18:35 | |
| *** acoles is now known as acoles_ | 18:37 | |
| *** forcebanana has quit IRC | 18:38 | |
| *** bjkeller has joined #openstack-swift | 18:42 | |
| *** ntt has quit IRC | 18:46 | |
| *** openstackgerrit has quit IRC | 18:47 | |
| *** openstackgerrit has joined #openstack-swift | 18:47 | |
| *** esker has quit IRC | 18:50 | |
| openstackgerrit | David Goetz proposed openstack/swift: go: fix requests with Range header for 0-bytes files (DLO or special objects like links for example) https://review.openstack.org/274331 | 18:53 |
| notmyname | summary of the "what's going on in swift" questions that I asked everyone about recently | 18:54 |
| notmyname | http://lists.openstack.org/pipermail/openstack-dev/2016-February/086446.html | 18:54 |
| *** dmorita has quit IRC | 18:57 | |
| *** dmorita has joined #openstack-swift | 18:58 | |
| *** dmorita has quit IRC | 18:59 | |
| *** dmorita has joined #openstack-swift | 18:59 | |
| *** haomaiwa_ has quit IRC | 19:01 | |
| *** haomaiwang has joined #openstack-swift | 19:01 | |
| timburke | clayg: i *love* Depends-On! fairly certain you can use it within a single repo, too, which i maybe should've done for patch 234391 | 19:05 |
| patchbot | timburke: https://review.openstack.org/#/c/234391/ - swift - Have versioned_writes use reverse listings (MERGED) | 19:05 |
| notmyname | here's a list of community-ranked patches in gerrit for swift. https://gist.github.com/notmyname/8a8d2403d24e39f59173 | 19:08 |
| notmyname | current review timings for swift are: Patch owners at 4 days 7 hours. Patch reviewers at 4 days 23 hours. patch owners are winning | 19:09 |
| notmyname | there are 21 open patches in swift with zero reviewer comments | 19:09 |
| portante | okay, so POSIX on swift? | 19:10 |
| *** dmsimard has quit IRC | 19:10 | |
| *** nadeem has quit IRC | 19:12 | |
| notmyname | portante: ? | 19:12 |
| portante | FILE this: Swift API, then S3 API, and now POSIX access to OpenStack Swift | 19:13 |
| clayg | notmyname: the graph is cool | 19:15 |
| clayg | timburke: well w/i a repo we have patch chains | 19:15 |
| clayg | timburke: either way I'd never heard about it before today - but am glad it exists | 19:15 |
| clayg | portante: stranger things have happened | 19:17 |
| clayg | notmyname: I like the community starred patches too! | 19:17 |
| *** petertr7_away is now known as petertr7 | 19:17 | |
| portante | didn't somebody do gmail as a filesystem? | 19:17 |
| clayg | lol | 19:17 |
| notmyname | that sounds awesome | 19:17 |
| timburke | clayg: yeah, but the changes in versioned_writes didn't overlap with the changes in container/* at all. if i'd used Depends-On, i never would have needed to submit a new patchset just because there was a new reverse-listings patch | 19:17 |
| portante | https://sr71.net/projects/gmailfs/ | 19:18 |
| onovy | https://en.wikipedia.org/wiki/GmailFS // this? | 19:18 |
| portante | sorry, I should be banned from IRC chat sessions for I only contribute digressions | 19:18 |
| onovy | btw: this is much better idea: http://lcamtuf.coredump.cx/juggling_with_packets.txt | 19:19 |
| notmyname | I like the idea of using 4k metadata on zero byte swift objects to implement a block storage system. also, you'll likely end up with zero changeback costs because most swift providers don't count metadata! | 19:21 |
| *** bgmccollum has left #openstack-swift | 19:21 | |
| *** dmorita has quit IRC | 19:23 | |
| *** dmsimard has joined #openstack-swift | 19:26 | |
| *** dmorita has joined #openstack-swift | 19:26 | |
| *** dmorita_ has joined #openstack-swift | 19:26 | |
| jrichli | notmyname: Thanks for all the stats! | 19:26 |
| notmyname | here are the completely unreviewed open patches https://gist.github.com/notmyname/1b93fa4c955fd82e7994 | 19:27 |
| openstackgerrit | Tim Burke proposed openstack/swift: Add Expires header for successful GETs using tempurls https://review.openstack.org/279737 | 19:28 |
| *** dmorita_ has quit IRC | 19:28 | |
| notmyname | bah. looks like I inadvertantly removed some | 19:28 |
| notmyname | will update | 19:28 |
| *** dmorita_ has joined #openstack-swift | 19:28 | |
| *** dmorita_ has quit IRC | 19:28 | |
| notmyname | ok, updated https://gist.github.com/notmyname/1b93fa4c955fd82e7994 | 19:29 |
| *** dmorita_ has joined #openstack-swift | 19:30 | |
| onovy | cschwede: ad 279440. you can save timestamp of last successfull audit inside same json files and audit in ts ascending order. and without json files at first place? | 19:30 |
| *** dmorita has quit IRC | 19:30 | |
| onovy | and devices without... | 19:30 |
| gmmaha | notmyname: would having that list as part of the dashboard help? | 19:31 |
| notmyname | gmmaha: yes. but I can't just paste an arbitrary list to the gerrit dashboard. I'll probably end up created and hosting my own html dashboard somewhere | 19:32 |
| *** trifon has quit IRC | 19:32 | |
| gmmaha | notmyname: was thinking a little less complicated.. :) gerrit search and adding it to the swift gerrit dash creator | 19:33 |
| gmmaha | i havent looked into all the new gerrit search params.. guessing its do-able. | 19:33 |
| * gmmaha goes to see if there are some search params that do that | 19:33 | |
| notmyname | gmmaha: right, but that's my point. eg I'd have to figure out the right search syntax for those exact results. and gerrit dashboards do not let your sort at all | 19:33 |
| notmyname | gmmaha: finding the unreviewed patches might be doable. at least "patches with no +/-1 comments is pretty easy | 19:34 |
| notmyname | but I couldn't do the community stars list at all | 19:34 |
| gmmaha | aah didnt know the community stars is that hard.. never had to search anything over that | 19:35 |
| notmyname | you can get a list of patches that anyone has starred | 19:36 |
| notmyname | but my list weights by how active that person is in the community | 19:36 |
| *** dmsimard has quit IRC | 19:36 | |
| gmmaha | aaah. | 19:36 |
| *** dmorita_ has quit IRC | 19:36 | |
| notmyname | basically, each reviewer gets a number of points between 0 and 100. each patch they star adds that many points to that patch. then I sort the patches by points and take the top 20 to get my list | 19:37 |
| gmmaha | thats pretty neat!! :) | 19:38 |
| notmyname | teh number of points is calculated by percent of time you've been active since you started contributing times the percent of time you've been active in the overall life of the project | 19:38 |
| cschwede | onovy: thx for your feedback, responded in the review | 19:38 |
| cschwede | notmyname: thx; I re-sent my tweet. note to myself: check for typos before hitting send ;) | 19:39 |
| notmyname | cschwede: heh | 19:39 |
| *** dmorita has joined #openstack-swift | 19:39 | |
| onovy | cschwede: hmm, good idea, better than mine :] | 19:39 |
| notmyname | so a person who left a comment 3 years ago and today isn't counted very highly, nor is a person who only started contributing yesterday. but someone who's been actively contributing for a long time is counted highly | 19:39 |
| notmyname | gmmaha: it's very close to how "red" your name is on http://d.not.mn/contrib_activity.png | 19:40 |
| gmmaha | notmyname: right.. get the active ones the high say | 19:40 |
| onovy | ha! my name is red :) | 19:41 |
| gmmaha | notmyname: :D wow | 19:41 |
| gmmaha | i have to wrap up the other project and get my street cred up again! | 19:42 |
| *** CaioBrentano has joined #openstack-swift | 19:47 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Remove execute permission from swift/cli/ringbuilder.py https://review.openstack.org/279449 | 19:48 |
| *** haypo has left #openstack-swift | 19:51 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Simplify policy-name validation slightly https://review.openstack.org/279333 | 19:52 |
| notmyname | http://d.not.mn/contrib_activity.png updated to include the weight value for each person. | 19:54 |
| notmyname | (cdn hosted, so a cached version may show) | 19:55 |
| onovy | notmyname: please :) https://review.openstack.org/#/c/263691/ | 19:55 |
| patchbot | onovy: https://review.openstack.org/#/c/263691/ - swift - Additional info log message for drive-audit | 19:55 |
| onovy | and btw: where is legend for that image? | 19:56 |
| notmyname | onovy: just ask acoles_. he rolls his eyes every time I bring it out ;-) | 19:56 |
| onovy | blue=patch, green=review? | 19:56 |
| notmyname | onovy: so here's how to read it | 19:56 |
| notmyname | actually, wait. I just recently wrote this down for someone... let me find it | 19:57 |
| notmyname | onovy: https://gist.github.com/notmyname/a60673d006419fff78cd | 19:58 |
| clayg | notmyname: you need a version that sorts by redness | 19:58 |
| onovy | clayg: +1 | 19:59 |
| notmyname | I had an original version that did that. but it was harder to read, actually. didnt' show the flow of activity over time | 19:59 |
| clayg | notmyname: well idk that would be interesting or not - but it might be | 19:59 |
| clayg | notmyname: yeah i might could see that | 19:59 |
| onovy | notmyname: thanks, nice | 20:00 |
| *** haomaiwang has quit IRC | 20:01 | |
| onovy | green=review, blue=patch // please put this simple legend into image :) | 20:01 |
| *** haomaiwang has joined #openstack-swift | 20:01 | |
| clayg | notmyname: the words are great! | 20:02 |
| jrichli | I always love seeing the progression of this graph! just one suggestion - some way to represent work on feature branches. :-) | 20:04 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 20:06 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 20:07 |
| timburke | notmyname: also, might be nice to include swiftclient. i don't think i see joel on there at all | 20:07 |
| notmyname | yeah, this is just for master branch on swift. I'd like to combine it with swiftclient too | 20:08 |
| *** joeljwright has joined #openstack-swift | 20:08 | |
| *** ChanServ sets mode: +v joeljwright | 20:08 | |
| *** changbl has quit IRC | 20:10 | |
| *** silor has quit IRC | 20:13 | |
| *** ankurgupta has left #openstack-swift | 20:14 | |
| openstackgerrit | Shashirekha Gundur proposed openstack/swift: change default ports for servers https://review.openstack.org/274840 | 20:17 |
| openstackgerrit | John Dickinson proposed openstack/swift-bench: added metadata section to setup.cfg https://review.openstack.org/279771 | 20:23 |
| openstackgerrit | John Dickinson proposed openstack/swift-bench: version bump for 1.2 release https://review.openstack.org/279772 | 20:23 |
| openstackgerrit | John Dickinson proposed openstack/swift-bench: version bump to open 1.2.1 dev work https://review.openstack.org/279773 | 20:23 |
| notmyname | take 2 on a swift-bench release | 20:24 |
| *** joeljwright has quit IRC | 20:24 | |
| openstackgerrit | Victor Stinner proposed openstack/swift: Port swob to Python 3 https://review.openstack.org/279704 | 20:24 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Port get_hmac() and hash_path() to Python 3 https://review.openstack.org/236998 | 20:24 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Parse HTTP as bytes on Python 3 https://review.openstack.org/237042 | 20:24 |
| jrichli | I am trying to change the internal-client.conf pipeline used during probetests. I modified both /etc/swift/internal-client.conf and the sample config in the repo, but it *seems* neither of them are taking effect. | 20:26 |
| *** KriSstaL has joined #openstack-swift | 20:28 | |
| clayg | jrichli: some of the probetests define their own configuration for their internal client instance (I think) | 20:28 |
| jrichli | hmm, I see an "allow_modify_pipeline" in the InternalClient ... | 20:28 |
| jrichli | clayg: oh, ok. i will look for that | 20:29 |
| KriSstaL | hola | 20:29 |
| jrichli | clayg: if the test doesn't define its own, what would the default be? the sample config? | 20:30 |
| *** KriSstaL has left #openstack-swift | 20:30 | |
| clayg | jrichli: I think /etc/swift/internal-client.conf - maybe /etc/swift/expiring-objects.conf in the expirier test | 20:31 |
| clayg | jrichli: which one is failing for you? | 20:31 |
| jrichli | this is for encryption: there are new container-sync probetests that reveal that you do indeed need to add encryption to the internal client pipeline (or something) to make container-sync work. | 20:33 |
| jrichli | we knew that this was needed, but there were no tests to reveal it until now | 20:33 |
| *** daemontool has quit IRC | 20:33 | |
| jrichli | so its a good time to address it :-) | 20:33 |
| *** daemontool has joined #openstack-swift | 20:33 | |
| jrichli | test_sync (test.probe.test_container_sync.TestContainerSync) is one of them | 20:34 |
| jrichli | but only fails with crypto - again, as expected | 20:34 |
| *** daemontool has quit IRC | 20:34 | |
| jrichli | but I am hoping that all I need to do is add crypto mw to its pipeline | 20:35 |
| *** daemontool has joined #openstack-swift | 20:35 | |
| clayg | timburke: how do you find this stuff? RE: patch 279333 | 20:37 |
| patchbot | clayg: https://review.openstack.org/#/c/279333/ - swift - Simplify policy-name validation slightly | 20:37 |
| clayg | timburke is always like "saw this stupid thing - fixed it" | 20:37 |
| clayg | I'm constly like "saw this stupid thing - not worth it" | 20:38 |
| *** MVenesio has quit IRC | 20:41 | |
| clayg | siva_krishnan: np, re patch #274840 on the requirements.txt - it probably just needs a rebase | 20:41 |
| patchbot | clayg: https://review.openstack.org/#/c/274840/ - swift - change default ports for servers | 20:41 |
| clayg | I'm really not sure how the probe test thing is going to effect people (developers) | 20:41 |
| clayg | *hopefully* everyone can just do a sed on their remakerings scirpts or pull from vagrant-swift-all-in-one branch (when someone/me writes it :\) and things will mostly just work? | 20:42 |
| clayg | siva_krishnan: acctually - where *you* able to get probetests passing with this branch? what's your development setup like? | 20:42 |
| openstackgerrit | Merged openstack/swift-bench: added metadata section to setup.cfg https://review.openstack.org/279771 | 20:43 |
| openstackgerrit | Merged openstack/swift-bench: version bump for 1.2 release https://review.openstack.org/279772 | 20:43 |
| siva_krishnan | clayg: sgundur is working on this patch | 20:44 |
| clayg | siva_krishnan: a thousand apologies! | 20:45 |
| clayg | gah - so sorry :'( | 20:45 |
| siva_krishnan | clayg no problem. wil ask sgundur to respond | 20:45 |
| sgundur | clayg: I checked the unittests were passing | 20:49 |
| openstackgerrit | Merged openstack/swift-bench: version bump to open 1.2.1 dev work https://review.openstack.org/279773 | 20:50 |
| clayg | sgundur: ok, might double check functests (gate will too, I think the in-process would be more likely to be effected than functional - devstacks setup should still work) | 20:52 |
| clayg | sgundur: but the probetests I think have been failing on the community cluster - not sure why | 20:52 |
| clayg | charz: ^ ??? | 20:52 |
| sgundur | clayg: just now check the probetests - they r failing , will check the functests ? | 20:53 |
| sgundur | clayg: thanks | 20:53 |
| *** ankurgupta1 has joined #openstack-swift | 20:54 | |
| *** CaioBrentano has quit IRC | 20:56 | |
| notmyname | fresh link to avoid caching issues that has the weights on it (the weights used to calculate the stared patches) http://d.not.mn/swift_contrib_activity.png | 20:59 |
| notmyname | https://gist.github.com/notmyname/8a8d2403d24e39f59173 is generated from those scores | 21:00 |
| sgundur | clayg: functests failing too , here is the link http://paste.openstack.org/show/486890/ | 21:00 |
| *** haomaiwang has quit IRC | 21:01 | |
| notmyname | (the scores are based on contrib activity over the last 2 years instead of activity for all time) | 21:01 |
| *** haomaiwang has joined #openstack-swift | 21:01 | |
| *** dmorita has quit IRC | 21:04 | |
| *** mrmoje has quit IRC | 21:04 | |
| *** changbl has joined #openstack-swift | 21:05 | |
| *** mrmoje has joined #openstack-swift | 21:06 | |
| notmyname | oh cool. the swift-bench release worked this time. https://pypi.python.org/pypi/swift-bench | 21:10 |
| notmyname | ta da! | 21:10 |
| *** petertr7 is now known as petertr7_away | 21:15 | |
| clayg | ~*~magic~*~ | 21:15 |
| *** petertr7_away is now known as petertr7 | 21:16 | |
| onovy | swift-bench release?! | 21:16 |
| notmyname | onovy: done. | 21:16 |
| onovy | two releases in 24 hours? | 21:16 |
| onovy | wow! | 21:16 |
| notmyname | onovy: no, just one | 21:17 |
| notmyname | the first one didn't take | 21:17 |
| onovy | https://github.com/openstack/swift-bench/releases | 21:17 |
| onovy | tag is here :] | 21:17 |
| notmyname | yeah, but it never got to pypi | 21:17 |
| notmyname | so IMO that one doesn't count | 21:17 |
| onovy | just write infra, they will fix it | 21:17 |
| onovy | not needed to do another release | 21:17 |
| onovy | (had same problem with swauth) | 21:17 |
| onovy | (maybe) | 21:18 |
| notmyname | no, I'm good. I did talk to -infra. we're good as-is, and that's the way it's staying | 21:18 |
| onovy | ah! setup.cfg, sry | 21:18 |
| *** nadeem has joined #openstack-swift | 21:28 | |
| onovy | and btw: thanks! packaging... | 21:30 |
| *** nadeem has quit IRC | 21:32 | |
| *** nadeem has joined #openstack-swift | 21:33 | |
| *** changbl has quit IRC | 21:35 | |
| *** aunnam has quit IRC | 21:37 | |
| *** marcusvrn_ has quit IRC | 21:37 | |
| *** dmorita has joined #openstack-swift | 21:37 | |
| openstackgerrit | Ankur proposed openstack/swift: Docstring omission in class BaseDiskFileManager. https://review.openstack.org/278615 | 21:39 |
| -openstackstatus- NOTICE: The infrastructure team is taking gerrit offline for maintenance this afternoon, beginning at 22:00 utc. We should have it back online around 23:00 utc. http://lists.openstack.org/pipermail/openstack-dev/2016-February/086195.html | 21:47 | |
| openstackgerrit | Thiago da Silva proposed openstack/swift: decouple versioned writes from COPY https://review.openstack.org/260179 | 21:49 |
| openstackgerrit | Merged openstack/swift: Remove execute permission from swift/cli/ringbuilder.py https://review.openstack.org/279449 | 21:54 |
| *** mrmoje has quit IRC | 21:55 | |
| *** daemontool has quit IRC | 21:58 | |
| *** haomaiwang has quit IRC | 22:01 | |
| *** haomaiwa_ has joined #openstack-swift | 22:01 | |
| *** loth has joined #openstack-swift | 22:01 | |
| -openstackstatus- NOTICE: Gerrit is offline for maintenacne until 23:00 utc | 22:06 | |
| *** ChanServ changes topic to "Gerrit is offline for maintenacne until 23:00 utc" | 22:06 | |
| *** petertr7 is now known as petertr7_away | 22:10 | |
| notmyname | if you didn't see the notice or the topic changes, seems gerrit is undergoing some maintenance and is offline for another 45 minutes or so | 22:13 |
| notmyname | still seems available to me, but that's what the message says | 22:14 |
| loth | Hey all, I have a bunch of e3-1230's with 1x1tb installed sitting unused, would it be terrible to use these for proxy and/or storage nodes? All the docs I've seen call for huge storage nodes but I've never seen much on tons of smaller servers. | 22:15 |
| notmyname | loth: the cool thing is that you can run swift on just about anything (I've installed it on a raspberry pi) | 22:16 |
| notmyname | loth: it's really a question of being able to give the performance you need for your workload | 22:16 |
| loth | Well id imagine the performance would be better with 4x servers with 1tb each than 1 server with 4x 1tb | 22:17 |
| notmyname | yeah | 22:17 |
| notmyname | in general, that's true | 22:17 |
| notmyname | but also, in general, drives dominate the cost of a storage cluster and people use swift for really large storage | 22:17 |
| notmyname | so to answer your question, yes, swift will work with that sort of setup that you have | 22:18 |
| loth | True, if i were to buy hardware I'd probably look for whatever is most efficent gb/ru | 22:18 |
| notmyname | but there are other configs that would likely be more efficient :-) | 22:18 |
| notmyname | right | 22:18 |
| loth | Are there any issues colocating the OS and swift storage drive? | 22:19 |
| notmyname | it could work. but normally that would be a bad idea. swift expects drives to fail and treats that differently than a server failing. so if your combined OS/data drive fails, swift will treat it like a server failure | 22:20 |
| notmyname | normally a server failure is an availability issue and not a durability issue (eg power loss). so in your case you'd have to make sure you more proactively respond to the OS/data drive failing | 22:20 |
| notmyname | so it will work. but it would take more operational effort to handle failures | 22:21 |
| loth | Wouldnt it replicate that data to other avalible servers and continue on? | 22:21 |
| notmyname | swift does that on a drive failure, but not a server failure. the normal case is to have 12, 24, or more drives in one server. you don't want that to be dumped onto the network every time there's a small blip with a storage node | 22:22 |
| loth | so standard practice on a drive failure is to move that data locally onto another disk instead of another node on the network? | 22:22 |
| notmyname | no, it moves in to another drive in the cluster. but data on one drive is spread throughout the rest of the cluster. there are no two mirrors of a drive in a swift cluster | 22:23 |
| notmyname | loth: here's a talk I gave a few years ago on swift's failure handling https://www.youtube.com/watch?v=_sUvfGKhaMo | 22:24 |
| loth | Cool, i'll take a watch | 22:26 |
| notmyname | loth: I need to run, but definitely other people in here will be able to answer questions too | 22:27 |
| loth | Thanks! | 22:27 |
| *** ankurgupta1 has left #openstack-swift | 22:34 | |
| mattoliverau | But just not as eloquently as notmyname :) | 22:38 |
| *** DericHorn-HP has joined #openstack-swift | 22:38 | |
| loth | I see, so if a server were to go down, it wouldnt replicate that data automatically since it thinks its a short-term outage? | 22:39 |
| *** changbl has joined #openstack-swift | 22:44 | |
| timburke | clayg: you already nailed it. there's a reason i've got 30+ open patches in gerrit | 22:45 |
| timburke | that one in particular i meant to leave as a review comment, but didn't get around to actually submitting it before the patch merged | 22:45 |
| clayg | nice | 22:47 |
| timburke | clayg: and like 1/3 of those aren't even swift/swiftclient. some are at least somewhat related (patch 233245, patch 276517), but a fair few are just out of left-field (patch 272727, patch 276500) | 22:50 |
| patchbot | timburke: https://review.openstack.org/#/c/233245/ | 22:50 |
| patchbot | timburke: https://review.openstack.org/#/c/276517/ | 22:50 |
| patchbot | timburke: https://review.openstack.org/#/c/272727/ | 22:50 |
| patchbot | timburke: https://review.openstack.org/#/c/276500/ | 22:50 |
| timburke | ah, right, gerrit's down. those are from like os-loganalyze, hacking, meetbot... | 22:50 |
| loth | Is it possible to change how swift handles server-down and have it replicate data? | 22:52 |
| clayg | timburke: you're weird | 22:55 |
| timburke | yup | 22:55 |
| clayg | did gerrit *know* it was going down - or was this like a surpise friday afternoon thing? | 22:56 |
| *** chsc has quit IRC | 22:56 | |
| redbo | http://lists.openstack.org/pipermail/openstack-dev/2016-February/086195.html | 22:56 |
| redbo | It was scheduled | 22:57 |
| clayg | redbo: whoa! look at you keeping up with the ML like a pro! | 22:57 |
| *** nadeem has quit IRC | 22:58 | |
| redbo | Nah, I only looked when "git review" failed :) | 22:58 |
| redbo | If you put a 0-byte object in swift and send "Range: bytes=-1", it returns 206 with "Content-Range: bytes 0--1/0". But it doesn't send me the -1th byte of the file. | 23:00 |
| *** haomaiwa_ has quit IRC | 23:01 | |
| *** nadeem has joined #openstack-swift | 23:01 | |
| *** haomaiwang has joined #openstack-swift | 23:01 | |
| *** bjkeller has quit IRC | 23:03 | |
| redbo | Or is it supposed to take a byte from me? RFC unclear. | 23:05 |
| -openstackstatus- NOTICE: Gerrit is offline for maintenance, ETA updated to 23:30 utc | 23:07 | |
| *** ChanServ changes topic to "Gerrit is offline for maintenance, ETA updated to 23:30 utc" | 23:07 | |
| *** DericHorn-HP has quit IRC | 23:23 | |
| -openstackstatus- NOTICE: Gerrit is offline for maintenance, ETA updated to 23:59 utc | 23:42 | |
| *** ChanServ changes topic to "Gerrit is offline for maintenance, ETA updated to 23:59 utc" | 23:42 | |
| *** mragupat has quit IRC | 23:52 | |
| *** mragupat has joined #openstack-swift | 23:55 | |
| *** mragupat has quit IRC | 23:57 | |
| *** _JZ_ has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
