| *** rickyrem has joined #openstack-swift | 00:06 | |
| *** diogogmt has quit IRC | 00:21 | |
| *** jamielennox|away is now known as jamielennox | 00:33 | |
| *** zhill has joined #openstack-swift | 00:36 | |
| timburke | clayg: thanks for rebasing patch 264961! so is that essentially a +2? | 00:38 |
|---|---|---|
| patchbot | timburke: https://review.openstack.org/#/c/264961/ - swift - Make write-only container access consistent | 00:38 |
| openstackgerrit | Michael Barton proposed openstack/swift: go: ObjectEngine abstraction layer https://review.openstack.org/278932 | 00:43 |
| openstackgerrit | Michael Barton proposed openstack/swift: go: ObjectEngine abstraction layer https://review.openstack.org/278932 | 00:44 |
| mattoliverau | ho_: morning | 00:54 |
| ho_ | mattoliverau: morning! | 01:02 |
| clayg | timburke: oh did i not +2 it :\ | 01:15 |
| *** diogogmt has joined #openstack-swift | 01:18 | |
| *** jroll has quit IRC | 01:22 | |
| *** jroll has joined #openstack-swift | 01:24 | |
| *** zhill has quit IRC | 01:30 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Fix object-server to handle newer ts file https://review.openstack.org/281636 | 01:32 |
| *** lyrrad has quit IRC | 01:35 | |
| *** trifon has quit IRC | 01:36 | |
| *** baojg has joined #openstack-swift | 01:48 | |
| *** chlong_ has joined #openstack-swift | 01:55 | |
| *** asettle has joined #openstack-swift | 01:59 | |
| *** forcebanana has quit IRC | 02:01 | |
| *** mingdang1 has joined #openstack-swift | 02:01 | |
| *** jamielennox is now known as jamielennox|away | 02:02 | |
| *** david-lyle_ has joined #openstack-swift | 02:04 | |
| *** david-lyle_ is now known as david-lyle | 02:05 | |
| *** asettle has quit IRC | 02:07 | |
| *** jamielennox|away is now known as jamielennox | 02:12 | |
| clayg | notmyname: tsg and kota aren't in channel; I'm getting stupid log lines everywhere I install liberasurecode -> https://gist.github.com/clayg/af012b4793954461f862 | 02:16 |
| *** haomaiwang has joined #openstack-swift | 02:16 | |
| *** haomaiwang has quit IRC | 02:21 | |
| ho_ | clayg: any change i can get re-review for patch 271142 and patch 271160 ? | 02:21 |
| patchbot | ho_: https://review.openstack.org/#/c/271142/ - swift - Add unit tests for checking responses in AccountCo... | 02:21 |
| patchbot | ho_: https://review.openstack.org/#/c/271160/ - swift - Add unit tests for checking responses in Container... | 02:21 |
| *** jamielennox is now known as jamielennox|away | 02:24 | |
| portante | notmyname, clayg, torgomatic, have you folks looked at http://pmem.io/nvml/libpmemobj/ | 02:24 |
| portante | would be fun, huh? | 02:24 |
| *** jamielennox|away is now known as jamielennox | 02:25 | |
| *** jamielennox is now known as jamielennox|away | 02:26 | |
| *** forcebanana has joined #openstack-swift | 02:28 | |
| *** forcebanana has quit IRC | 02:30 | |
| *** forcebanana has joined #openstack-swift | 02:32 | |
| *** haomaiwang has joined #openstack-swift | 02:32 | |
| *** jamielennox|away is now known as jamielennox | 02:33 | |
| clayg | portante: only in relation to xpoint - general idea seems fine - not directly relevant to me today | 02:33 |
| clayg | portante: but if we ever get away from those pesky filesystems - it's ON | 02:34 |
| *** jamielennox is now known as jamielennox|away | 02:34 | |
| *** jamielennox|away is now known as jamielennox | 02:35 | |
| *** jamielennox is now known as jamielennox|away | 02:36 | |
| *** haomaiwang has quit IRC | 02:37 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 02:37 |
| portante | I hear tell that Intel was to be shipping NVM hardware is systems by 2017 | 02:37 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 02:38 |
| *** haomaiwa_ has joined #openstack-swift | 02:48 | |
| *** forcebanana has quit IRC | 02:50 | |
| *** jamielennox|away is now known as jamielennox | 02:51 | |
| ho_ | clayg: s/change/chance/ | 02:56 |
| *** StraubTW_ has quit IRC | 02:58 | |
| *** Zyric_ has quit IRC | 02:58 | |
| *** sanchitmalhotra has joined #openstack-swift | 03:00 | |
| *** haomaiwa_ has quit IRC | 03:01 | |
| *** haomaiwa_ has joined #openstack-swift | 03:01 | |
| *** asettle has joined #openstack-swift | 03:06 | |
| *** rickyrem has quit IRC | 03:10 | |
| *** Jeffrey4l__ has joined #openstack-swift | 03:11 | |
| *** asettle has quit IRC | 03:11 | |
| *** Zyric_ has joined #openstack-swift | 03:14 | |
| *** esker has joined #openstack-swift | 03:27 | |
| *** links has joined #openstack-swift | 03:30 | |
| *** asettle has joined #openstack-swift | 03:30 | |
| *** asettle has quit IRC | 03:34 | |
| *** david-lyle_ has joined #openstack-swift | 03:42 | |
| *** david-lyle has quit IRC | 03:45 | |
| *** baojg has quit IRC | 03:53 | |
| *** haomaiwa_ has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-swift | 04:01 | |
| mahatic | good morning | 04:02 |
| *** david-lyle_ has quit IRC | 04:03 | |
| *** david_lyle__ has joined #openstack-swift | 04:03 | |
| ho_ | mahatic: morning! | 04:05 |
| *** mingyu has quit IRC | 04:06 | |
| *** mingyu has joined #openstack-swift | 04:06 | |
| *** gyee has quit IRC | 04:07 | |
| *** david_lyle__ has quit IRC | 04:08 | |
| clayg | ho_: oh no - what did I do? | 04:22 |
| clayg | oh - patches - right on | 04:22 |
| notmyname | good evening | 04:24 |
| notmyname | clayg: are those liberasurecode things maybe related to ldconfig? | 04:26 |
| clayg | i don't know?! | 04:26 |
| clayg | i type ldconfig sometimes | 04:26 |
| clayg | I don't know what it does | 04:26 |
| clayg | it doesn't seem like ti does anything | 04:26 |
| clayg | it prints stuff I can't understand on my terminal | 04:27 |
| notmyname | google tells me that it's the shared library cache. and tsg always tells me to do that when i have problems ;-) | 04:27 |
| clayg | lol - k we're totally on the same page then | 04:27 |
| clayg | but in this case anyway it doesn't make the errors go away | 04:28 |
| notmyname | it seems your errors have to do with all of the actual ec libraries that you might have installed, therefore it's obviously a caching problem, right? that's one of those "hard things" | 04:28 |
| notmyname | ah. | 04:28 |
| notmyname | yeah, we're totally on the same page :-) | 04:28 |
| clayg | notmyname: check your syslog on your saio - do you have them messages too? | 04:28 |
| clayg | notmyname: I mean like my stuff is working - it's just being stupid noisy about stuff I dont't care about | 04:28 |
| notmyname | oh, interesting | 04:29 |
| notmyname | do you have storage policies configured that reference all those libraries? | 04:29 |
| notmyname | in swift.conf | 04:29 |
| clayg | notmyname: NO WAY | 04:29 |
| clayg | i just have the liberasure_vand thing | 04:29 |
| clayg | rs_ in there somehwere | 04:29 |
| clayg | idk | 04:29 |
| clayg | the default | 04:29 |
| notmyname | oh look. it's in my syslog too. we're TOTALLY on the same page | 04:30 |
| clayg | notmyname: then i have some other nodes with isa-l configured/installed - no warning about missing isa-l.so on those nodes | 04:30 |
| notmyname | why would it have references to those? | 04:30 |
| clayg | i filed it upstream - tsg says its going to get better when we have bigger numbers | 04:30 |
| clayg | bigger numbers are key | 04:30 |
| notmyname | yeah, I've got ISA-L on my saio, so no errors about that one | 04:31 |
| clayg | "I think what you are seeing is a result of VALID_EC_TYPES evaluation in pyeclib" | 04:31 |
| clayg | ^ from tsg on https://bitbucket.org/tsg-/liberasurecode/issues/23/spurious-messages-in-syslog | 04:31 |
| clayg | but he says bigger numbers will make it better (upcoming release/version w/e - point is bigger) | 04:32 |
| notmyname | my initial guess is that's a wrong guess | 04:33 |
| notmyname | because VALIE_EC_TYPES doesn't have anything about libshss in it, and that's one of the messages | 04:34 |
| notmyname | argh! | 04:34 |
| notmyname | or I can't read at all | 04:34 |
| notmyname | either I need more of or less of this margarita I'm halfway through ;-) | 04:34 |
| clayg | that's the ticket! | 04:35 |
| * notmyname finds it very weird that pyeclib has special casing for shss (or, really, any 3rd party library. but especially shss) | 04:35 | |
| *** nakagawamsa has joined #openstack-swift | 04:37 | |
| *** nakagawamsa has quit IRC | 04:38 | |
| *** ppai has joined #openstack-swift | 04:39 | |
| openstackgerrit | Mingyu Li proposed openstack/swift: Order the devices by (region, zone, ip, device) in the output of swift-ring-builder https://review.openstack.org/277956 | 04:42 |
| notmyname | clayg: based on commit message subject, what's your gut reaction to that patch? ^ | 04:44 |
| notmyname | for reference, https://github.com/MicrowiseOnGitHub/tempfiles/blob/master/reorder_ring_output | 04:45 |
| clayg | notmyname: big of buckets of win! | 04:52 |
| notmyname | I guess I can see that. my first reaction is that it's ordering on something that's not the primary key (the dev_id) so things get "mixed up". but the gist example seems reasonable... | 04:53 |
| clayg | notmyname: idk - there's a linked bug right? | 04:53 |
| notmyname | oh so we have a bug filed and it's something that has to be done? | 04:54 |
| notmyname | ah ha! of course you like it. you filed that bug! ;-) | 04:54 |
| clayg | no the bug might be invalid - but the justifcation had something to do with devices no longer show up in "the time your added them" order | 04:55 |
| clayg | which was... neat? but now it doesn't make any sense... so... some order is better than... none? | 04:55 |
| openstackgerrit | Tim Burke proposed openstack/swift: Add Expires header for successful GETs using tempurls https://review.openstack.org/279737 | 04:56 |
| notmyname | there's an order! and it's inscrutable. how else are we to keep our aura of mystery?! | 04:56 |
| notmyname | I haven't played much with updating rings in a post resuse dev_ids world | 04:56 |
| clayg | lol - dunno - we've got bugs filed on swift-ring-builder help message grammer - so you know... brave new world | 04:56 |
| notmyname | yeah, I think my initial reaction was "you're changing what?! the order of devices in the ring? can you even do that?". but yeah. seems totally reasonable | 04:58 |
| notmyname | good job, mingyu. thanks for working on it :-) | 04:58 |
| clayg | ho_: do you want to take a stab to DRY out that common test case setup in patch 271160 like acoles_ nit'd on? | 04:59 |
| patchbot | clayg: https://review.openstack.org/#/c/271160/ - swift - Add unit tests for checking responses in Container... | 04:59 |
| clayg | ho_: I have to agree, it seems gross to have that all duplicated? | 04:59 |
| *** haomaiwang has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-swift | 05:01 | |
| openstackgerrit | Bing Hu proposed openstack/python-swiftclient: ecnode auth token in swiftclient log message https://review.openstack.org/259273 | 05:02 |
| timburke | hrm. it makes me sad that you can't just do `swift-ring-builder /etc/swift/object.ring.gz | sort -nk 1` to get the old order... | 05:02 |
| clayg | ho_: I'd probably just do it like this -> https://gist.github.com/clayg/0658a29a29090c14d54b | 05:04 |
| notmyname | timburke: but why would you want the old order? couldn't you also use `sort -R` to get something effectively similar? ;-) | 05:04 |
| clayg | timburke: --json | 05:05 |
| ho_ | clayg: thanks! i will do it | 05:06 |
| clayg | ho_: too late! | 05:06 |
| notmyname | timburke: clayg: and after, --json, just use http://jsontoyaml.com | 05:07 |
| openstackgerrit | Clay Gerrard proposed openstack/swift: dry out container unittests https://review.openstack.org/282173 | 05:07 |
| clayg | ^ ho_ could use some help with the commit message | 05:07 |
| clayg | notmyname: --json-to-yaml-to-humman-to-xml | 05:07 |
| ho_ | clayg: oh, i understand "too late" means. i will update the commit message :-) | 05:08 |
| timburke | notmyname: *shrug* you were the one that seemed to freak out over a change in order... | 05:10 |
| timburke | clayg: don't tempt me :P | 05:10 |
| notmyname | timburke: but now I've seen the light | 05:10 |
| timburke | hehe | 05:11 |
| clayg | notmyname: IME timburke makes even *MOAR* sense when I've been drinking | 05:11 |
| notmyname | timburke: new plan. when you come over to my desk in the office, bring a bottle ;-) | 05:11 |
| openstackgerrit | Bing Hu proposed openstack/python-swiftclient: ecnode auth token in swiftclient log message https://review.openstack.org/259273 | 05:12 |
| timburke | clayg: when you've been drinking, or me? maybe both? either way, seems like i've got a plan for Bristol! | 05:13 |
| notmyname | lol | 05:13 |
| timburke | notmyname: why bring a bottle when the good tequila's practically already there? | 05:14 |
| *** SkyRocknRoll has joined #openstack-swift | 05:14 | |
| clayg | timburke: like I said, I'm taking my week in Bristol to merge fast-POST - everything else is lagniappe | 05:16 |
| openstackgerrit | Bing Hu proposed openstack/python-swiftclient: encode auth token in swiftclient log message https://review.openstack.org/259273 | 05:16 |
| timburke | clayg: that's a good word. i'll have to try to remember that one. TIL | 05:17 |
| mingyu | notmyname: clayg: thank you for reviewing that! I was having my launch :-) | 05:19 |
| notmyname | clayg: wow. bringing out the cajun words ;-) | 05:19 |
| mingyu | I'm really happy that I can do something to contribute back to Swift :-) as I've earned some money building storage systems based-on Swift for my customers in the past three years. | 05:28 |
| *** nadeem has joined #openstack-swift | 05:28 | |
| *** asettle has joined #openstack-swift | 05:31 | |
| *** asettle has quit IRC | 05:35 | |
| *** rcernin has joined #openstack-swift | 05:37 | |
| *** nakagawamsa has joined #openstack-swift | 05:42 | |
| notmyname | :-) | 05:42 |
| *** baojg has joined #openstack-swift | 05:45 | |
| *** nakagawamsa has quit IRC | 05:48 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwa_ has joined #openstack-swift | 06:01 | |
| *** asettle has joined #openstack-swift | 06:06 | |
| * mattoliverau is having one of those days where he's failing at debugging tests. Screw it its Friday arvo, I'm rage quitting for the day! | 06:09 | |
| mattoliverau | Tl:Dr : I'm calling it a day, have a great weekend all | 06:09 |
| notmyname | mattoliverau: you too :-) | 06:09 |
| openstackgerrit | Tim Burke proposed openstack/swift: Add Vary: headers for CORS responses https://review.openstack.org/274303 | 06:10 |
| *** asettle has quit IRC | 06:11 | |
| *** resker has joined #openstack-swift | 06:13 | |
| *** esker has quit IRC | 06:15 | |
| *** nadeem has quit IRC | 06:23 | |
| *** nadeem has joined #openstack-swift | 06:24 | |
| *** asettle has joined #openstack-swift | 06:29 | |
| *** resker has quit IRC | 06:30 | |
| *** asettle has quit IRC | 06:34 | |
| *** McMurlock has joined #openstack-swift | 06:35 | |
| peterlisak | peterli | 06:42 |
| notmyname | hello peterlisak | 06:43 |
| notmyname | timburke: I've got unreviewed swiftclient patches on the new dashboard now. and counted in the timings (although I think I've got a bug there). not yet on community reviews | 06:44 |
| timburke | huzzah! | 06:45 |
| notmyname | and swiftclient unreviewed are listed after swift unreviewed patches. don't have the info plumbed through yet to do the more correct ordering | 06:45 |
| *** McMurlock has quit IRC | 06:47 | |
| *** zaitcev has quit IRC | 06:47 | |
| *** trifon has joined #openstack-swift | 06:50 | |
| *** kei_yama has quit IRC | 06:52 | |
| timburke | notmyname: huh. i expected patch 184956 to be higher than it was. i guess it's ordered by time-since-last-update rather than time-since-creation | 06:53 |
| patchbot | timburke: https://review.openstack.org/#/c/184956/ - python-swiftclient - Accept gzip-encoded API responses | 06:53 |
| *** esker has joined #openstack-swift | 06:53 | |
| notmyname | timburke: note it's number 3rd oldest in the set of swiftclient patches | 06:54 |
| notmyname | timburke: but yeah, IIRC gerrit returns data in last-udated order | 06:55 |
| timburke | ...did you see when it was created? :P | 06:55 |
| notmyname | no ;-) | 06:55 |
| notmyname | timburke: I'll raise you https://review.openstack.org/#/c/103779/ | 06:56 |
| patchbot | notmyname: patch 103779 - swift - Add support for multiple container-reconciler | 06:56 |
| timburke | although that's actually gotten reviews ;-) | 06:57 |
| notmyname | right | 06:57 |
| *** haomaiwa_ has quit IRC | 07:01 | |
| *** 20WAABHNQ has joined #openstack-swift | 07:01 | |
| notmyname | timburke: oldest open patch (actually, longest active for both open an closed patches) in swift is https://review.openstack.org/#/c/62494/ | 07:10 |
| patchbot | notmyname: patch 62494 - swift - Add containeralias middleware | 07:10 |
| notmyname | nearly 800 days :-( | 07:10 |
| timburke | mighty old | 07:10 |
| notmyname | your https://review.openstack.org/#/c/184956/ patch is the oldest open patch for swiftclient (@ ~ 273 days | 07:12 |
| patchbot | notmyname: patch 184956 - python-swiftclient - Accept gzip-encoded API responses | 07:12 |
| timburke | since Vancouver! | 07:12 |
| notmyname | bed time for me. talk to everyone tomorrow | 07:15 |
| *** silor has joined #openstack-swift | 07:18 | |
| openstackgerrit | Merged openstack/swift: Add unit tests for checking responses in AccountController https://review.openstack.org/271142 | 07:23 |
| *** silor1 has joined #openstack-swift | 07:25 | |
| *** silor has quit IRC | 07:26 | |
| *** silor1 is now known as silor | 07:26 | |
| *** mingyu has quit IRC | 07:28 | |
| *** mingyu has joined #openstack-swift | 07:28 | |
| *** esker has quit IRC | 07:31 | |
| *** chlong_ has quit IRC | 07:41 | |
| *** pcaruana has joined #openstack-swift | 07:45 | |
| *** esker has joined #openstack-swift | 07:50 | |
| *** 20WAABHNQ has quit IRC | 08:01 | |
| *** haomaiwang has joined #openstack-swift | 08:01 | |
| *** nadeem has quit IRC | 08:14 | |
| *** klrmn has quit IRC | 08:21 | |
| *** ChanServ sets mode: +v cschwede | 08:22 | |
| *** asettle has joined #openstack-swift | 08:30 | |
| *** esker has quit IRC | 08:31 | |
| *** asettle has quit IRC | 08:35 | |
| *** esker has joined #openstack-swift | 08:41 | |
| *** jmccarthy has joined #openstack-swift | 08:43 | |
| *** acoles_ is now known as acoles | 08:51 | |
| *** davideagle_ has joined #openstack-swift | 08:52 | |
| *** haomaiwang has quit IRC | 09:01 | |
| *** haomaiwang has joined #openstack-swift | 09:01 | |
| acoles | good morning | 09:05 |
| *** mingyu has quit IRC | 09:20 | |
| *** mingyu has joined #openstack-swift | 09:20 | |
| *** jordanP has joined #openstack-swift | 09:20 | |
| mahatic | good morning | 09:29 |
| *** d0ugal has quit IRC | 09:30 | |
| *** esker has quit IRC | 09:30 | |
| *** d0ugal has joined #openstack-swift | 09:31 | |
| *** d0ugal has quit IRC | 09:31 | |
| *** d0ugal has joined #openstack-swift | 09:31 | |
| *** d0ugal has quit IRC | 09:31 | |
| *** d0ugal has joined #openstack-swift | 09:32 | |
| *** d0ugal has quit IRC | 09:32 | |
| *** d0ugal has joined #openstack-swift | 09:32 | |
| *** esker has joined #openstack-swift | 09:34 | |
| *** baojg has quit IRC | 09:44 | |
| *** baojg has joined #openstack-swift | 09:45 | |
| *** haomaiwang has quit IRC | 10:01 | |
| *** asettle has joined #openstack-swift | 10:01 | |
| *** davideagle_ has quit IRC | 10:01 | |
| *** haomaiwang has joined #openstack-swift | 10:01 | |
| *** StraubTW_ has joined #openstack-swift | 10:02 | |
| *** stantonnet has quit IRC | 10:03 | |
| *** stantonnet has joined #openstack-swift | 10:05 | |
| *** StraubTW_ has quit IRC | 10:07 | |
| *** davideagle_ has joined #openstack-swift | 10:09 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Silence SkipTest noise in in-process test teardown https://review.openstack.org/282261 | 10:16 |
| acoles | clayg: notmyname ^^ | 10:17 |
| *** mingyu has quit IRC | 10:34 | |
| *** mingyu has joined #openstack-swift | 10:34 | |
| *** asettle has quit IRC | 10:40 | |
| *** ho_ has quit IRC | 10:40 | |
| *** asettle has joined #openstack-swift | 10:40 | |
| *** remix_tj has joined #openstack-swift | 10:41 | |
| *** haypo has joined #openstack-swift | 10:47 | |
| acoles | tdasilva: thanks for fixing that commit message | 10:53 |
| *** mingyu has quit IRC | 10:54 | |
| *** mingyu has joined #openstack-swift | 10:54 | |
| haypo | cschwede: hi. i'm back for my hmac change/config parser issue. https://review.openstack.org/#/c/236998/ if i understood correctly, the main risk is that upgrading from py2 to py3 would break the world if py3 "reads" the hash prefix/suffix differently, right? | 10:55 |
| patchbot | haypo: patch 236998 - swift - Port get_hmac() and hash_path() to Python 3 | 10:55 |
| haypo | cschwede: i suggest to implement a guard to prevent any mistake: raise a strong exception if the hash prefix/suffix may be interpreted differently on py2/py3, for example if it contains '%' | 10:56 |
| haypo | cschwede: *and* add a new optional format for hash prefix/suffix to support arbitrary random bytes | 10:56 |
| haypo | i propose to add a base64 format, ex: "hash_prefix = base64:xxxx" | 10:57 |
| haypo | we can provide a tool or at least a recipe in swift doc to explain how to generate such base64-encoded config | 10:57 |
| cschwede | haypo: yep, exactly. i brought it up on Wednesdays community meeting, asking for more reviews. and there is an idea for config validation from clayg: https://review.openstack.org/#/c/281547/ | 10:57 |
| patchbot | cschwede: patch 281547 - swift - Better hash path config validation | 10:57 |
| *** baojg has quit IRC | 10:58 | |
| haypo | example: >>> print("hash_prefix = %s" % base64.b64encode(os.urandom(32)).decode('ascii')) | 10:58 |
| haypo | hash_prefix = p+2B2etSKYZJj7XCtZK+7MW+Q2Xu5m/Rwnqou4ekg5o= | 10:58 |
| acoles | ppai: hi, tdasilva jrichli and I have been working on some patches to allow copy middleware to go to left of slo, dlo in pipeline, and now I am making changes to patch 156923 - do you mind if I push changes over your review? | 10:58 |
| patchbot | acoles: https://review.openstack.org/#/c/156923/ - swift - Refactor server side copy as middleware | 10:58 |
| *** baojg has joined #openstack-swift | 10:58 | |
| haypo | cschwede: oh right, you had the same idea than me :) | 10:59 |
| haypo | cschwede: https://review.openstack.org/#/c/281547/ is the first part, but it will break existing deployement using a config which is now considered as "bad" (invalid) | 10:59 |
| patchbot | haypo: patch 281547 - swift - Better hash path config validation | 10:59 |
| haypo | cschwede: i suggest to also provide a way to keep the old config. otherwise, you cannot start your cluster anymore, right? | 10:59 |
| haypo | or is it possible to change a hash prefix/suffix? does it require expensive operations? | 11:00 |
| *** haomaiwang has quit IRC | 11:01 | |
| *** haomaiwang has joined #openstack-swift | 11:01 | |
| cschwede | haypo: you’re right, if the validation fails you won’t be able to start any swift service. changing the suffix/prefix is not possible; and making it changeable would require moving all data in the cluster… | 11:02 |
| cschwede | so that’s a nogo. | 11:02 |
| haypo | cschwede: ok. what do you think of my base64 format? | 11:02 |
| ppai | acoles, not at all | 11:03 |
| haypo | cschwede: it should be backward compatible. IMHO it's very unlikely that something used "base64:" prefix as a random hash prefix/suffix :) | 11:03 |
| *** baojg has quit IRC | 11:03 | |
| *** mingdang1 has quit IRC | 11:05 | |
| cschwede | haypo: imo it should be enough if operators simply use printable chars. i added a tiny note on that: https://review.openstack.org/#/c/281492/ | 11:05 |
| patchbot | cschwede: patch 281492 - swift - Add note on using printable chars for swift_hash_p... (MERGED) | 11:05 |
| cschwede | or even restricting it to chars and digits | 11:06 |
| cschwede | that should be enough; but it doesn’T solve the problem for possibly existing strange settings | 11:06 |
| haypo | cschwede: i'm talking about existing settings | 11:07 |
| haypo | cschwede: technically, does it matter that the hash prefix/suffix is "very" random? | 11:07 |
| haypo | i don't know the purpose of it | 11:07 |
| acoles | ppai: thanks - that's a great patch you did btw. | 11:07 |
| haypo | i'm asking because a base64 format would allow to generate high quality entropy | 11:08 |
| *** haomaiwang has quit IRC | 11:08 | |
| ppai | acoles: And sorry, I've been busy and couldn't give much attention to it. | 11:10 |
| cschwede | haypo: the purpose is to make the object location unguessable, so „just“ random should be enough (even a hexadecimal string with like 15-20 chars should be more than enough) | 11:11 |
| *** dabukalam has joined #openstack-swift | 11:12 | |
| acoles | ppai: np | 11:12 |
| haypo | cschwede: a long prefix/suffix slows down all hash computations. base64 encoding allows to get more entropy with shorter strings. in practice, i have no idea of the cost of hash computations on an overall swift request :) | 11:13 |
| haypo | cschwede: well, it's just a side note :) my concern is not to make swift faster, just to prevent users to have to change the hash suffix/prefix if swift starts to refuse to start because hash prefix/suffix is now seen as "invalid" | 11:15 |
| haypo | for my use case (porting swift to py3), clayg's change is enough to avoid bad surprises when upgrading from py2 to py3, since the behaviour will be the same | 11:16 |
| haypo | cschwede: oh. your change on the doc asks to use 'printable chars', but '%' is in string.printable, and my problem is on the '%' char | 11:17 |
| asettle | notmyname: sorry I missed your ping! Loving London :D | 11:18 |
| *** mingyu_ has joined #openstack-swift | 11:19 | |
| cschwede | haypo: yep, but using the raw parser in py3 makes it possible to still use „%“, right? | 11:21 |
| openstackgerrit | renminmin proposed openstack/swift: Account/Container quotas should allow object re-upload https://review.openstack.org/263227 | 11:21 |
| haypo | cschwede: i don't want swift to behave differently depending on the python version | 11:21 |
| *** mingyu has quit IRC | 11:22 | |
| cschwede | haypo: iirc, % was usable on py2? i need to have a look at my tests again… but i agree with you, it should behave the same on py2 and py3 | 11:23 |
| haypo | cschwede: Swift currently uses ConfigParser.ConfigParser, you *can* write hash_suffix = %(hash_prefix)s | 11:24 |
| *** linkmark has joined #openstack-swift | 11:28 | |
| *** d0ugal has quit IRC | 11:31 | |
| *** esker has quit IRC | 11:31 | |
| *** d0ugal has joined #openstack-swift | 11:31 | |
| *** d0ugal has quit IRC | 11:31 | |
| *** d0ugal has joined #openstack-swift | 11:31 | |
| haypo | cschwede: i commented https://review.openstack.org/#/c/281547/ to explain '%' issue and suggest base64 format | 11:36 |
| patchbot | haypo: patch 281547 - swift - Better hash path config validation | 11:36 |
| *** NM has joined #openstack-swift | 11:41 | |
| *** esker has joined #openstack-swift | 11:43 | |
| *** sanchitmalhotra has quit IRC | 11:43 | |
| *** mingdang1 has joined #openstack-swift | 11:55 | |
| *** davideagle_ has quit IRC | 11:59 | |
| *** mingyu_ has quit IRC | 12:00 | |
| *** mingyu has joined #openstack-swift | 12:01 | |
| *** chlong_ has joined #openstack-swift | 12:11 | |
| *** davideagle_ has joined #openstack-swift | 12:13 | |
| *** mingyu has quit IRC | 12:20 | |
| *** mingyu has joined #openstack-swift | 12:20 | |
| *** cdelatte has joined #openstack-swift | 12:22 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Add containeralias middleware https://review.openstack.org/62494 | 12:23 |
| *** davideagle_ has quit IRC | 12:27 | |
| *** esker has quit IRC | 12:27 | |
| *** ppai has quit IRC | 12:30 | |
| *** davideagle_ has joined #openstack-swift | 12:34 | |
| *** esker has joined #openstack-swift | 12:36 | |
| *** kota_ has joined #openstack-swift | 12:40 | |
| *** ChanServ sets mode: +v kota_ | 12:40 | |
| kota_ | yeah,I got resolved my bouncer issue finally. | 12:41 |
| kota_ | kota_: Test from mobile | 12:43 |
| kota_ | seems to work well | 12:43 |
| *** asettle has quit IRC | 12:44 | |
| peterlisak | clayg, notmyname , mattoliverau : Hi, could you review my ionice patch 238799? Thanks. I replaced psutil by own wrapper. | 12:47 |
| patchbot | peterlisak: https://review.openstack.org/#/c/238799/ - swift - Change schedule priority of daemon/server in config | 12:47 |
| *** kota_ has quit IRC | 12:51 | |
| *** kota_ has joined #openstack-swift | 12:51 | |
| *** ChanServ sets mode: +v kota_ | 12:51 | |
| openstackgerrit | Merged openstack/swift: Make write-only container access consistent https://review.openstack.org/264961 | 12:56 |
| *** asettle has joined #openstack-swift | 13:00 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: PUT/GET path for global ec cluster https://review.openstack.org/219165 | 13:02 |
| *** km has quit IRC | 13:09 | |
| *** fthiagogv__ has quit IRC | 13:12 | |
| *** SkyRocknRoll has quit IRC | 13:14 | |
| tdasilva | good morning | 13:14 |
| tdasilva | acoles: np from me | 13:15 |
| acoles | tdasilva: tests looking good with pipeline of copy middleware/dlo/slo/versioned_writes (all 3 patches applied) :) | 13:17 |
| acoles | kota_: :) | 13:18 |
| tdasilva | acoles: awesome, I was trying to do that last night but ran into some problems. Regarding the copy manifest between accounts, I think we are ok with responding with an error if the segments are not already there, correct? | 13:19 |
| acoles | tdasilva: yes. I changed the failing func test to verify that the slo manifest fails if the segments are not readable in the dest account, then create readable segments in dest, then assert the manifest copy succeeds | 13:21 |
| acoles | tdasilva: and I think timburke and I (and you?) agreed that was reasonable behaviour | 13:21 |
| tdasilva | acoles: yep, it makes sense to me...like timburke said, otherwise, if they do a get on the manifest, they would get an error anyway | 13:22 |
| *** ho_away has joined #openstack-swift | 13:24 | |
| *** lpabon has joined #openstack-swift | 13:33 | |
| *** davideagle_ has quit IRC | 13:42 | |
| *** StraubTW_ has joined #openstack-swift | 13:45 | |
| *** ndk has quit IRC | 13:47 | |
| *** portante has quit IRC | 13:47 | |
| *** mingdang1 has quit IRC | 13:52 | |
| *** mingdang1 has joined #openstack-swift | 13:53 | |
| *** davideagle_ has joined #openstack-swift | 13:53 | |
| *** StraubTW_ has quit IRC | 13:55 | |
| *** diogogmt has quit IRC | 14:01 | |
| *** fthiagogv has joined #openstack-swift | 14:02 | |
| *** diogogmt has joined #openstack-swift | 14:02 | |
| *** baojg has joined #openstack-swift | 14:07 | |
| *** mingyu has quit IRC | 14:10 | |
| *** mingyu has joined #openstack-swift | 14:10 | |
| *** mingyu has quit IRC | 14:12 | |
| *** mingyu has joined #openstack-swift | 14:12 | |
| *** mingyu has quit IRC | 14:15 | |
| *** mingyu has joined #openstack-swift | 14:16 | |
| openstackgerrit | Alexandra Settle proposed openstack/swift: WIP: New operational procedures guide https://review.openstack.org/278245 | 14:16 |
| *** mingdang1 has quit IRC | 14:18 | |
| *** SkyRocknRoll has joined #openstack-swift | 14:19 | |
| *** mingdang1 has joined #openstack-swift | 14:20 | |
| *** trifon has quit IRC | 14:21 | |
| *** mingdang1 has quit IRC | 14:25 | |
| *** esker has quit IRC | 14:26 | |
| *** ndk has joined #openstack-swift | 14:26 | |
| *** links has quit IRC | 14:27 | |
| *** StraubTW_ has joined #openstack-swift | 14:29 | |
| *** baojg has quit IRC | 14:34 | |
| mahatic | tdasilva: acoles yay, that's great to hear! looks like we're gonna have COPY middleware soon | 14:34 |
| *** diogogmt has quit IRC | 14:37 | |
| *** petertr7_away is now known as petertr7 | 14:37 | |
| *** portante has joined #openstack-swift | 14:41 | |
| *** openstackgerrit_ has quit IRC | 14:43 | |
| *** diogogmt has joined #openstack-swift | 14:49 | |
| openstackgerrit | Joel Wright proposed openstack/python-swiftclient: Do not reveal auth token in swiftclient log messages by default https://review.openstack.org/282363 | 14:54 |
| *** ho_away has quit IRC | 14:55 | |
| openstackgerrit | Béla Vancsics proposed openstack/swift: Reduce code duplication https://review.openstack.org/267767 | 14:57 |
| openstackgerrit | Joel Wright proposed openstack/python-swiftclient: Do not reveal auth token in swiftclient log messages by default https://review.openstack.org/282363 | 14:59 |
| *** openstackgerrit_ has joined #openstack-swift | 15:00 | |
| *** linkmark has quit IRC | 15:02 | |
| *** joeljwright has joined #openstack-swift | 15:06 | |
| *** ChanServ sets mode: +v joeljwright | 15:06 | |
| *** Jeffrey4l__ has quit IRC | 15:07 | |
| jrichli | acoles: yay! good to see movement on that one | 15:10 |
| *** davideagle_ has quit IRC | 15:12 | |
| *** links has joined #openstack-swift | 15:13 | |
| *** davideagle_ has joined #openstack-swift | 15:14 | |
| *** davideagle_ has quit IRC | 15:15 | |
| *** dhellmann has quit IRC | 15:16 | |
| *** dhellmann has joined #openstack-swift | 15:19 | |
| *** esker has joined #openstack-swift | 15:20 | |
| *** esker has quit IRC | 15:25 | |
| notmyname | good morning | 15:34 |
| joeljwright | notmyname: good morning | 15:34 |
| joeljwright | notmyname: I tried to combine the 2 patches regarding logging auth tokens (https://review.openstack.org/#/c/282363) | 15:36 |
| *** mvk has quit IRC | 15:36 | |
| notmyname | asettle: I had a docs-related question for you. | 15:37 |
| joeljwright | CLI —debug outputting complete curl commands is also maintained - would be nice if people could have a look | 15:37 |
| notmyname | asettle: why isn't https://github.com/openstack/python-swiftclient/commit/a3a78be87b88beca83a8cc0c96e209ab8e1a4189 reflected in http://docs.openstack.org/developer/python-swiftclient/ | 15:37 |
| notmyname | joeljwright: ah cool | 15:37 |
| asettle | That's a good question | 15:37 |
| joeljwright | does anyone know how I can contact someone on the ubuntu cloud archive team regarding swift packages? | 15:38 |
| * asettle blinks | 15:38 | |
| asettle | No. | 15:38 |
| notmyname | joeljwright: I don't remember who does that. zigo and onovy do swift in debian. zaitcev does swift in RDO. I don't remember who does swift in ubuntu | 15:39 |
| *** bjkeller has joined #openstack-swift | 15:40 | |
| zigo | notmyname: The changelog may be helpful: http://changelogs.ubuntu.com/changelogs/pool/main/s/swift/swift_2.5.0-0ubuntu1/changelog | 15:40 |
| notmyname | joeljwright: there you go :-) | 15:41 |
| joeljwright | zigo: thanks | 15:41 |
| zigo | (you get this by searching for http://packages.ubuntu.com/swift, then click on the latest release, then search for "changelog" in that page) | 15:41 |
| notmyname | heh. a good chunk of doing stuff in open source seems to be knowing who has what phrases highlighted in their IRC client ;-) | 15:42 |
| zigo | notmyname: BTW, any chance that the CVE for Swift in Jessie gets fixed? :) | 15:42 |
| zigo | (sorry, that's what you get when highlighting me... :P) | 15:42 |
| notmyname | zigo: that's the backport for juno, right? | 15:42 |
| zigo | Yup. | 15:42 |
| notmyname | zigo: I wrote it down on my todo list. but TBH it's not at the top. | 15:43 |
| * zigo clicks on the up arrow 10 times... | 15:44 | |
| asettle | acoles: yeah I will have to get around to that. At the moment I am just formatting the RST, i'll go through after and do exactly that soon too. Thank you for pointing it out :) | 15:45 |
| asettle | There's a lot of bonky crud in there, haha. | 15:45 |
| *** _JZ_ has joined #openstack-swift | 15:46 | |
| acoles | asettle: k | 15:46 |
| acoles | asettle: you are coming to Bristol right? | 15:46 |
| asettle | acoles: yep. Sitting in the London office currently. | 15:46 |
| acoles | oic | 15:46 |
| asettle | I'm trying to get this done before we get there. | 15:46 |
| acoles | asettle: the original authors of that hpe content will be here too for the hackathon | 15:46 |
| asettle | 2500 to go. | 15:46 |
| asettle | acoles: amazing, point them to me? | 15:47 |
| asettle | There's a lot of blanks which have understandably been redacted and would really appreciate them being filled in with *some* credible knowledge of what was tehre previously. | 15:47 |
| acoles | asettle: they don't hang out here, but i will make sure you get intro in bristol | 15:47 |
| asettle | acoles: thank you :) you coming? | 15:47 |
| acoles | asettle: i'm hosting :) | 15:48 |
| asettle | ... | 15:48 |
| asettle | Of course you are :p | 15:48 |
| asettle | Merp. | 15:48 |
| acoles | hehe | 15:48 |
| asettle | I'd love to say it's jet lag or something but I've been here long enough now. | 15:48 |
| acoles | bristol is 13mins west of london so you'll get another dose of jet lag when you move here :) | 15:49 |
| *** fthiagogv has quit IRC | 15:49 | |
| notmyname | asettle: are you in london permanently yet? or do you have another return to oz planned? | 15:49 |
| notmyname | acoles: 13 minutes? | 15:49 |
| asettle | That works for me! I'm actually hoping to come up the weekend beforehand and do my tourist thang. | 15:49 |
| *** fthiagogv has joined #openstack-swift | 15:49 | |
| asettle | notmyname: Nope, back in April permanently | 15:49 |
| asettle | But I have been here all this week, and will be obviously the next until the hackathon. | 15:49 |
| asettle | But I get to set up some of my things while I'm here which has been good. UK phone number, etc etc | 15:50 |
| acoles | asettle: so are you relocating to UK? | 15:50 |
| asettle | acoles: you bet. I've come for the sun. | 15:51 |
| acoles | LOL! | 15:51 |
| acoles | is that to escape mattoliverau ? :P | 15:51 |
| asettle | Never! <3 | 15:51 |
| asettle | It's because I'm so pale I just needed to find people to fit in with :p | 15:51 |
| notmyname | acoles: I think you logged off yesterday before I had a v1 of this. check out http://not.mn/swift/swift_community_dashboard.html | 15:52 |
| acoles | notmyname: yeah i saw it in scrollback and I have been busying applying my stars | 15:53 |
| notmyname | lol | 15:53 |
| acoles | cannot star own reviews tho :( | 15:53 |
| asettle | Also notmyname did you get my email regarding your plans on passing the content? | 15:53 |
| notmyname | acoles: really? | 15:54 |
| jrichli | acoles: interesting. I am able to star my own review | 15:54 |
| acoles | huh. | 15:54 |
| notmyname | asettle: on getting stuff merged at the hackathon? | 15:54 |
| * acoles will try again | 15:54 | |
| asettle | notmyname: yerp | 15:54 |
| asettle | Cause I'll probably want/need this template-esque situation to be merged before we get there: https://review.openstack.org/#/c/278248/2 | 15:55 |
| patchbot | asettle: patch 278248 - python-swiftclient - New python swiftclient doc | 15:55 |
| asettle | OH thanks patchbot | 15:55 |
| acoles | notmyname: jrichli ok must have been a gerrit or user error :) | 15:55 |
| acoles | this could be interesting...starring in progress | 15:55 |
| notmyname | asettle: yes. I think that is a reasonable goal. especially if it's phrased as "some docs that aren't terrible are better than no docs, so let's land these now if possible" | 15:55 |
| asettle | Basically. It's much easier to have something up and through, and go back later and fix it up, than have a bout 20 merge conflicts and some really unhappy patches | 15:56 |
| asettle | Do we have enough cores to help out? | 15:56 |
| *** fthiagogv has quit IRC | 15:56 | |
| acoles | asettle: no | 15:56 |
| asettle | acoles: super. | 15:56 |
| acoles | asettle: help with what? :) | 15:56 |
| acoles | asettle: jk | 15:56 |
| asettle | Haha well, traditionally when I've done documentation based swarms, it's generally easiest to get the content up, reviewed by cores, and passed en masse. | 15:57 |
| asettle | Following that, we then have me/anyone else going back over the top and reviewing the content separately. | 15:57 |
| notmyname | asettle: I agree. and as terrible as it may sound, docs are easier to land since they generally don't break prod ;-) | 15:57 |
| asettle | Pretty much, tbh | 15:58 |
| asettle | And as long as people are okay with following a few quick RST conventions I'll point them in the direction of... I'm good (with a few others?) going over hte top and checking for accuracy of content and spelling/grammar | 15:58 |
| asettle | Thoughts on that funciton? | 15:58 |
| joeljwright | I'll certainly be helping with the docs | 15:58 |
| *** fthiagogv has joined #openstack-swift | 15:58 | |
| asettle | joeljwright: are you core? If not, we probs don't want you aye :P | 15:58 |
| joeljwright | core for python-swiftclient | 15:58 |
| joeljwright | :) | 15:59 |
| joeljwright | happy to wield a +2 hammer | 15:59 |
| asettle | joeljwright: https://www.youtube.com/watch?v=7oKPYe53h78 | 15:59 |
| asettle | So relevant rn | 15:59 |
| acoles | notmyname: asettle i think when you have a large amount of new content you could get trapped in endless word-smithing, better to land something that is reasonable and then polish in follow-up patches | 15:59 |
| notmyname | acoles: right | 15:59 |
| asettle | THat's the plan, Stan | 15:59 |
| joeljwright | acoles: yup | 16:00 |
| acoles | asettle: and joeljwright is in UK too so you get great timezone overlap | 16:00 |
| asettle | I know, it's muchos exciting. | 16:00 |
| asettle | Oh John Travolta | 16:00 |
| joeljwright | acoles: but you know this means we'll be bugging you more for the remaining +2... | 16:00 |
| asettle | acoles: please. I'll give YOU a Grease song too if you say yes | 16:01 |
| asettle | I know how tempting that is | 16:01 |
| acoles | joeljwright: i'm moving to mattoliverau-land | 16:01 |
| acoles | can't stand the sum | 16:01 |
| acoles | sun* | 16:01 |
| asettle | ... Melbourne? | 16:02 |
| notmyname | acoles: perth | 16:02 |
| asettle | Where'd you get Perth from?! | 16:02 |
| notmyname | if we're all together at the hackathon, I'd imagine perfect spot to hide fromt he sun | 16:02 |
| notmyname | lol | 16:02 |
| notmyname | perth is a perfect spot to hide from the sun | 16:02 |
| notmyname | funny how those 2 mistakenly joined phrases sorta worked | 16:03 |
| *** cschwede has quit IRC | 16:04 | |
| *** cschwede has joined #openstack-swift | 16:05 | |
| acoles | notmyname: you lost me. but yeah perth isn't the sunniest place on the planet https://www.google.co.uk/maps/place/Perth | 16:05 |
| notmyname | bad joke badly delivered ;-) | 16:06 |
| acoles | the good news today is that the hackathon snacks are safely stashed under my desk... | 16:06 |
| notmyname | https://goo.gl/maps/QfbBEjJJy992 is rather sunny ;-) | 16:06 |
| acoles | ...the bad news is I have to go a week resisting easting them! | 16:06 |
| notmyname | acoles: do you have "topics"? | 16:07 |
| acoles | that would be telling ;) | 16:07 |
| acoles | they only get deployed if *my* topics are selected | 16:07 |
| jrichli | hehe | 16:07 |
| acoles | *other* people's topics get rice crackers | 16:08 |
| *** pcaruana has quit IRC | 16:10 | |
| joeljwright | :D | 16:12 |
| asettle | acoles: what kind of snacks?! | 16:12 |
| asettle | Will I like them | 16:12 |
| asettle | Being Australian I naturally only eat things that can kill me. | 16:13 |
| *** klrmn has joined #openstack-swift | 16:13 | |
| *** haypo has left #openstack-swift | 16:13 | |
| acoles | asettle: i also have an hp z640 workstation under my desk, that would probably kill you | 16:13 |
| *** arch-nemesis has joined #openstack-swift | 16:13 | |
| asettle | acoles: cool. I'll just hide under your desk and feel right at home. | 16:14 |
| acoles | they're mostly unhealthy | 16:14 |
| tdasilva | are beers being served with snacks? | 16:15 |
| notmyname | acoles: that looks pretty nice as a desktop workstation ;-) | 16:16 |
| acoles | it is. stacked with 128GB too. | 16:16 |
| notmyname | nice | 16:16 |
| acoles | i use it for email | 16:16 |
| acoles | dev on my laptop | 16:16 |
| acoles | :P | 16:16 |
| notmyname | lol | 16:16 |
| *** SkyRocknRoll has quit IRC | 16:17 | |
| asettle | Beers... what about wine? | 16:20 |
| notmyname | acoles: I just got some of these at the office http://www.intel.com/content/www/us/en/solid-state-drives/solid-state-drives-dc-p3500-series.html | 16:21 |
| *** diazjf has joined #openstack-swift | 16:26 | |
| *** nadeem has joined #openstack-swift | 16:28 | |
| *** nadeem has quit IRC | 16:28 | |
| *** nadeem has joined #openstack-swift | 16:29 | |
| acoles | notmyname: nice. i have a 1TB ssd, not sure of specs, doubt it matches those! | 16:33 |
| * acoles regrets mentioning the snacks | 16:34 | |
| *** arch-nemesis has quit IRC | 16:35 | |
| *** diazjf has quit IRC | 16:39 | |
| *** bjkeller has quit IRC | 16:45 | |
| *** arch-nemesis has joined #openstack-swift | 16:47 | |
| *** gyee has joined #openstack-swift | 16:48 | |
| *** trifon has joined #openstack-swift | 16:50 | |
| *** links has quit IRC | 16:53 | |
| *** bjkeller has joined #openstack-swift | 16:55 | |
| *** diazjf has joined #openstack-swift | 16:55 | |
| *** rcernin has quit IRC | 16:58 | |
| openstackgerrit | Joel Wright proposed openstack/python-swiftclient: Do not reveal auth token in swiftclient log messages by default https://review.openstack.org/282363 | 17:01 |
| Zyric_ | Good morning | 17:04 |
| gmmaha | ls | 17:04 |
| *** chsc has joined #openstack-swift | 17:05 | |
| *** esker has joined #openstack-swift | 17:08 | |
| *** asettle has quit IRC | 17:12 | |
| *** lyrrad has joined #openstack-swift | 17:14 | |
| *** klrmn has quit IRC | 17:17 | |
| *** garthb has joined #openstack-swift | 17:17 | |
| clayg | Owners are winning! | 17:23 |
| clayg | one thing I like there is that whenI am both a reviewer and an owner - I'M ALWAYS #WINNING | 17:24 |
| *** esker has quit IRC | 17:29 | |
| joeljwright | timburke: you around? | 17:29 |
| *** rcernin has joined #openstack-swift | 17:30 | |
| acoles | clayg: can I beg for a review? fast-post patch depends on patch 267788 which is not so huge, maybe we could progress that before the hackathon? | 17:34 |
| patchbot | acoles: https://review.openstack.org/#/c/267788/ - swift - Fix inconsistent suffix hashes after ssync of tomb... | 17:34 |
| *** mingyu has quit IRC | 17:37 | |
| *** mingyu has joined #openstack-swift | 17:38 | |
| *** haomaiwang has joined #openstack-swift | 17:52 | |
| tdasilva | notmyname: re the dashboard, any plans to show the stats as a graph? like a trend.. | 17:55 |
| *** petertr7 is now known as petertr7_away | 17:55 | |
| *** jordanP has quit IRC | 17:56 | |
| *** haomaiwang has quit IRC | 17:56 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Refactor server side copy as middleware https://review.openstack.org/156923 | 17:57 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Re-format the SLO manifest file on new multipart-manifest GET call https://review.openstack.org/263902 | 17:57 |
| openstackgerrit | Alistair Coles proposed openstack/swift: decouple versioned writes from COPY https://review.openstack.org/260179 | 17:57 |
| acoles | tdasilva: jrichli ^^ the three patches are now in a chain (as listed above). server_side_copy to left of slo, dlo and versioned_writes, no more copy_hooks | 17:59 |
| *** Zyric_ has quit IRC | 18:00 | |
| *** zhill has joined #openstack-swift | 18:00 | |
| tdasilva | acoles: sweeeet! | 18:01 |
| notmyname | clayg: you're always a winner in my book | 18:05 |
| notmyname | tdasilva: I've thought about graphing it. but that brings in time-series stuff and a lot more complexity. right now it's running in a cron job on my laptop | 18:06 |
| *** asettle has joined #openstack-swift | 18:07 | |
| tdasilva | notmyname: got it | 18:07 |
| *** Gordoni has joined #openstack-swift | 18:08 | |
| joeljwright | clayg: new patch for redacting tokens in swiftclient logs here https://review.openstack.org/#/c/282363 | 18:11 |
| *** asettle has quit IRC | 18:12 | |
| joeljwright | clayg: CLI —debug can still give you the full tokens and it's a combination of the 2 previous patches | 18:12 |
| notmyname | acoles: is the only way in the new gerrit interface to see patch dependencies to look at the "same topic" section in the top right? | 18:13 |
| notmyname | joeljwright: yeah, I want to review that one today | 18:13 |
| joeljwright | notmyname: I'll check in on it later tonight and tomorrow in case there are required fixes | 18:14 |
| timburke | joeljwright: am now; was still catching up on scrollback | 18:14 |
| acoles | notmyname: I also see a Related-Changes tab at top right when there is a dependency | 18:14 |
| acoles | notmyname: but don't know any other way | 18:14 |
| joeljwright | timburke: cool, I combined the 2 auth token logging patches here - https://review.openstack.org/#/c/282363 | 18:14 |
| notmyname | acoles: any idea how it determines "related"? | 18:15 |
| timburke | i saw that; was going to aim for a review after a meeting this morning | 18:15 |
| notmyname | I like the old way of explicitly showing the patch chain /curmudgeon | 18:15 |
| acoles | notmyname: in my experience that have always been dependencies | 18:15 |
| *** klrmn has joined #openstack-swift | 18:16 | |
| acoles | notmyname: there's lots of old ways i preferred. like to change patch set now requires moving mouse to top right which is a long way on a big screen! | 18:16 |
| *** chlong_ has quit IRC | 18:18 | |
| *** Gordoni has quit IRC | 18:20 | |
| *** bjkeller has quit IRC | 18:22 | |
| timburke | acoles: agreed! and conversely, it doesn't display very well windowed, either! | 18:22 |
| *** remix_tj has quit IRC | 18:25 | |
| *** tonyb has quit IRC | 18:25 | |
| *** ejat has quit IRC | 18:25 | |
| *** balajir has quit IRC | 18:25 | |
| *** ctennis has quit IRC | 18:25 | |
| *** glange has quit IRC | 18:25 | |
| *** then3rd has quit IRC | 18:25 | |
| *** urth has quit IRC | 18:25 | |
| *** remix_tj has joined #openstack-swift | 18:25 | |
| *** then3rd has joined #openstack-swift | 18:26 | |
| *** glange has joined #openstack-swift | 18:26 | |
| *** ChanServ sets mode: +v glange | 18:26 | |
| *** ejat has joined #openstack-swift | 18:26 | |
| *** ejat has quit IRC | 18:26 | |
| *** ejat has joined #openstack-swift | 18:26 | |
| *** ctennis has joined #openstack-swift | 18:26 | |
| clayg | acoles: yeah I want to see the hash suffix stuff fixed - been on my list - but not doing a ton of reviews | 18:27 |
| *** hrou has quit IRC | 18:27 | |
| *** ajiang has quit IRC | 18:27 | |
| *** raginbajin has quit IRC | 18:27 | |
| *** redbo has quit IRC | 18:27 | |
| *** hurricanerix has quit IRC | 18:27 | |
| *** a1|away has quit IRC | 18:27 | |
| *** number80 has quit IRC | 18:27 | |
| *** jrichli has quit IRC | 18:27 | |
| *** lifeless has quit IRC | 18:27 | |
| *** early has quit IRC | 18:27 | |
| *** balajir has joined #openstack-swift | 18:27 | |
| *** redbo has joined #openstack-swift | 18:27 | |
| *** ChanServ sets mode: +v redbo | 18:27 | |
| *** raginbaj- has joined #openstack-swift | 18:27 | |
| *** ajiang has joined #openstack-swift | 18:27 | |
| *** lifeless has joined #openstack-swift | 18:27 | |
| *** a1|away has joined #openstack-swift | 18:27 | |
| *** jrichli has joined #openstack-swift | 18:27 | |
| *** early has joined #openstack-swift | 18:27 | |
| *** david-lyle has joined #openstack-swift | 18:27 | |
| *** hurricanerix_ has joined #openstack-swift | 18:28 | |
| *** StraubT__ has joined #openstack-swift | 18:28 | |
| *** StraubTW_ has quit IRC | 18:28 | |
| *** hrou has joined #openstack-swift | 18:28 | |
| acoles | clayg: yeah i hate to ask cos i know you're busy | 18:28 |
| *** urth has joined #openstack-swift | 18:28 | |
| *** number80 has joined #openstack-swift | 18:28 | |
| *** raginbaj- is now known as raginbajin | 18:29 | |
| *** flaper87 has quit IRC | 18:29 | |
| *** flaper87 has joined #openstack-swift | 18:29 | |
| acoles | timburke: the other one that bugs me is the distance from 'add reviewer' to 'add me' | 18:29 |
| *** wanghua has quit IRC | 18:30 | |
| *** Zyric has joined #openstack-swift | 18:30 | |
| *** tonyb has joined #openstack-swift | 18:31 | |
| *** jrichli has quit IRC | 18:31 | |
| *** jrichli has joined #openstack-swift | 18:31 | |
| *** hrou has quit IRC | 18:31 | |
| *** hrou has joined #openstack-swift | 18:31 | |
| *** wanghua has joined #openstack-swift | 18:36 | |
| acoles | have a good weekend everyone | 18:36 |
| joeljwright | acoles: you too! | 18:37 |
| *** joeljwright has quit IRC | 18:38 | |
| *** Gordoni has joined #openstack-swift | 18:41 | |
| *** acoles is now known as acoles_ | 18:42 | |
| *** petertr7_away is now known as petertr7 | 18:47 | |
| *** Gordoni has quit IRC | 18:47 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 18:48 |
| *** Gordoni has joined #openstack-swift | 18:49 | |
| *** haomaiwang has joined #openstack-swift | 18:53 | |
| *** nadeem has quit IRC | 18:55 | |
| *** fthiagogv has quit IRC | 18:56 | |
| *** rcernin has quit IRC | 18:56 | |
| *** haomaiwang has quit IRC | 18:58 | |
| *** Gordoni has quit IRC | 18:58 | |
| jrichli | acoles: yay! no more copy-hooks! the entire COPY as mw patch chain is all ready for review in that case. nice! | 18:58 |
| *** petertr7 is now known as petertr7_away | 19:02 | |
| *** Gordoni has joined #openstack-swift | 19:05 | |
| clayg | acoles_: no no you have to ashk! | 19:12 |
| clayg | jrichli: holy cow - awesome! | 19:12 |
| *** StraubT__ has quit IRC | 19:13 | |
| jrichli | clayg: I know! Having all functests pass for crypto is near. Now we just gotta pin you down at hackathon to chat about that whole timestamps / sysmeta /metadata thing again ;-) it will be a fun chat. | 19:14 |
| *** esker has joined #openstack-swift | 19:15 | |
| clayg | acoles_: well I'm worried about how using only the files that would construct a valid fileset might effect EC in a situation with multiple frag_archive's | 19:16 |
| clayg | but I'll thinkin through it and try to come up with a probe scenario by cribing from some of your stuff | 19:17 |
| *** StraubTW_ has joined #openstack-swift | 19:19 | |
| *** StraubTW_ has quit IRC | 19:21 | |
| *** StraubTW_ has joined #openstack-swift | 19:21 | |
| *** rcernin has joined #openstack-swift | 19:23 | |
| *** diogogmt has quit IRC | 19:24 | |
| *** diogogmt has joined #openstack-swift | 19:25 | |
| *** Zyric has quit IRC | 19:26 | |
| *** petertr7_away is now known as petertr7 | 19:26 | |
| *** Zyric has joined #openstack-swift | 19:27 | |
| *** Zyric_ has joined #openstack-swift | 19:27 | |
| *** Zyric_ has quit IRC | 19:27 | |
| *** Zyric has quit IRC | 19:30 | |
| *** jmccarthy has quit IRC | 19:32 | |
| *** silor has quit IRC | 19:37 | |
| *** bjkeller has joined #openstack-swift | 19:39 | |
| *** Zyric has joined #openstack-swift | 19:39 | |
| *** diogogmt has quit IRC | 19:40 | |
| *** asettle has joined #openstack-swift | 19:44 | |
| clayg | jrichli: no moar types of metdata! | 19:44 |
| *** rcernin has quit IRC | 19:45 | |
| *** diogogmt has joined #openstack-swift | 19:46 | |
| *** asettle has quit IRC | 19:48 | |
| *** asettle has joined #openstack-swift | 19:51 | |
| *** asettle has quit IRC | 19:51 | |
| *** ChubYann has joined #openstack-swift | 19:52 | |
| jrichli | clayg: I know :-) acoles_ has an idea. maybe it will be more palatable than the other routes we have thought of thus far. | 19:54 |
| *** trifon has quit IRC | 19:54 | |
| clayg | jrichli: acoles_ is so smart | 19:55 |
| jrichli | +1 | 19:55 |
| jrichli | +2 | 19:55 |
| *** zhill has quit IRC | 19:57 | |
| *** zhill has joined #openstack-swift | 19:57 | |
| *** ChubYann has quit IRC | 19:58 | |
| *** ChubYann_ has joined #openstack-swift | 19:58 | |
| *** asettle has joined #openstack-swift | 19:59 | |
| bjkeller | The tempest gate checks keep failing for patch 274906 saying the etag header does not exist. Does anyone know what the issue could be? | 20:05 |
| patchbot | bjkeller: https://review.openstack.org/#/c/274906/ - swift - Headers now use a case incensitive dictionary | 20:05 |
| siva_krishnan | Just had question about "swift-recon -u" command. when I was looking through recon code, I saw it was adding mounted drives to unmounted list as well. Can someone please confirm this ? | 20:08 |
| *** jmccarthy1 has quit IRC | 20:11 | |
| *** diazjf has quit IRC | 20:11 | |
| *** jmccarthy has joined #openstack-swift | 20:11 | |
| *** zhill has quit IRC | 20:16 | |
| *** mwheckmann has joined #openstack-swift | 20:29 | |
| *** asettle has quit IRC | 20:32 | |
| *** NM has quit IRC | 20:34 | |
| *** _JZ_ has quit IRC | 20:34 | |
| mwheckmann | Hello. Anyone familiar with intermittent timeouts connecting to memcached? I keep seeing the following under load: "proxy-server: STDERR: ERROR:root:Timeout talking to memcached: 172.16.0.17:11211" | 20:39 |
| mwheckmann | I have 2 separate memcache servers setup (on the proxies) and two proxies. | 20:39 |
| mwheckmann | This happens no matter what config value I set for "memcache_max_connections" | 20:40 |
| mwheckmann | memcached's are configured for 4k connections max, but I don't get anywhere near that. | 20:40 |
| *** haomaiwa_ has joined #openstack-swift | 20:41 | |
| clayg | siva_krishnan: it looks like it wants to return only unmounted devices? is using the utils.check_mount to set the boolean flag - and only the mounted is not True is appended to the mountlist that's retruend? | 20:41 |
| *** diazjf has joined #openstack-swift | 20:42 | |
| clayg | mwheckmann: I've seen network issues cause memcache timeouts - and sometimes memory pressure on the nodes - but nothing systemic | 20:42 |
| mwheckmann | I'm actually suspecting CPU contention with the proxy processes | 20:43 |
| clayg | mwheckmann: newer versions of swift allow some tuning via config for the very agressive timeout value to try and debug how slow the connections acctually are - but I've never since had an operational scearnio that required testing them | 20:43 |
| clayg | how many cores/procs do you have on the proxies? | 20:43 |
| mwheckmann | 24 threads | 20:43 |
| mwheckmann | 2x proc (6 core each) | 20:44 |
| mwheckmann | 48 proxy processes | 20:44 |
| mwheckmann | clayg: those timeouts documented somewhere? I'm running 2.5.0 | 20:45 |
| clayg | mwheckmann: oh like HT? and you have 48 workers on.... 12 physical cores? | 20:45 |
| mwheckmann | clayg: correct | 20:45 |
| clayg | memcache.conf-sample? | 20:45 |
| mwheckmann | my benchmarking shows that 48 is the sweet spot. | 20:45 |
| siva_krishnan | clayg When I was trying to write unit test for check_mount and gave [{u'device': u'sdb4', u'mounted': False}, {u'device': u'sdb8', u'mounted': True}] as input . It added both sdb4 and sdb8 in the unmounted list | 20:45 |
| mwheckmann | my stuff is in proxy-server.conf under the cache section. My understanding is that it's the same? | 20:46 |
| *** haomaiwa_ has quit IRC | 20:46 | |
| clayg | mwheckmann: well maybe :\ I mean that seems like the kernel is going to be doing a lot of process juggling - I've seen proxies with a lot more physical cores than 12 run more workers than that tho - so maybe you're fine? | 20:46 |
| mwheckmann | memcache_servers = 172.16.0.17:11211,172.16.0.18:11211 | 20:46 |
| mwheckmann | memcache_max_connections = 2 | 20:47 |
| mwheckmann | but I've gone up to 8 or more for the latter setting with no diff | 20:47 |
| clayg | siva_krishnan: I probably don't follow - the get_unmouted code rolls over all the devices in the listdir and calls checkmount and builds the mpoint dict - but the inclusion into the returned mountlist (of dicts) is conditional on mounted is not True | 20:47 |
| *** diazjf has quit IRC | 20:48 | |
| mwheckmann | I'm going to try re-nicing the memcached processes as a quick 'n dirty test to see if that makes it go away. That should prove contention. | 20:48 |
| clayg | mwheckmann: yeah I'm thinking [memcahce] connect_timeout might be worth playing with | 20:48 |
| clayg | mwheckmann: ok, good luck | 20:49 |
| *** proteusguy__ has quit IRC | 20:49 | |
| clayg | mwheckmann: oh sorry, yeah you should be able to set connect_timeout in the filter:cache section of the proxy-config just the same as the seperate memcache config | 20:49 |
| mwheckmann | clayg: cool. thanks | 20:50 |
| *** nadeem has joined #openstack-swift | 20:51 | |
| *** nadeem has quit IRC | 20:51 | |
| *** nadeem has joined #openstack-swift | 20:52 | |
| *** diazjf has joined #openstack-swift | 20:54 | |
| siva_krishnan | clayg sorry I coudn't confirm if we are checking for mounted is true or false in umount_check method in cli/recon.py. I was referring to this one http://paste.openstack.org/show/48762 | 20:58 |
| *** asettle has joined #openstack-swift | 20:58 | |
| *** asettle has quit IRC | 20:58 | |
| clayg | siva_krishnan: no, that's the wrong paste | 20:59 |
| siva_krishnan | http://paste.openstack.org/show/487627/ | 20:59 |
| siva_krishnan | clayg sorry for that | 20:59 |
| *** mingdang1 has joined #openstack-swift | 21:00 | |
| clayg | siva_krishnan: ok, so what's the question there? | 21:00 |
| *** proteusguy__ has joined #openstack-swift | 21:02 | |
| mwheckmann | clayg: so re-nice'ing memcached to -20 (including all of it's threads) makes the problem go away. So contention was the issue when the proxies are being pushed to the max. | 21:03 |
| mwheckmann | I'm just going to leave things as they are. | 21:03 |
| siva_krishnan | http://127.0.0.1:6010/recon/unmounted: [{u'device': u'sdb1', u'mounted': False}, {u'device': u'sdb5', u'mounted': True}]. If this is given as input. unmounted drives should be retruned as 1 (sdb1) right | 21:04 |
| siva_krishnan | but it was including sdb5 in the list as well | 21:04 |
| *** mingdang1 has quit IRC | 21:05 | |
| *** rickyrem has joined #openstack-swift | 21:05 | |
| *** _JZ_ has joined #openstack-swift | 21:07 | |
| clayg | siva_krishnan: yeah I don't think that method expects the recon data you're stubbing in? | 21:14 |
| clayg | siva_krishnan: a real object server won't return the mounted true device in that call | 21:14 |
| clayg | siva_krishnan: it might return some sort of non-boolean error string, or False - basically it expects everything coming back from the response deserves some kind of output | 21:15 |
| clayg | siva_krishnan: i'm sure you could *change* that? but I don't think your test is doing a good just creating a scearnio that cli tool is *actually* going to run into? | 21:16 |
| *** Gordoni has quit IRC | 21:17 | |
| siva_krishnan | ah ok got you! I was under the impression that it would return true if devices are mounted. thanks for that info. would remove that input as well | 21:19 |
| *** mingyu has quit IRC | 21:21 | |
| *** trifon has joined #openstack-swift | 21:27 | |
| clayg | siva_krishnan: i mean you can curl http://storage.node:port/recon/unmounted and see what's up if you want? | 21:30 |
| siva_krishnan | yup I already saw it | 21:30 |
| *** garthb_ has joined #openstack-swift | 21:30 | |
| clayg | a'ight | 21:31 |
| siva_krishnan | every drive was showing as unmounted. | 21:31 |
| clayg | was this on a dev/vm? | 21:32 |
| siva_krishnan | yup | 21:32 |
| siva_krishnan | on dev | 21:32 |
| siva_krishnan | SAIO setup | 21:32 |
| clayg | siva_krishnan: i was noticing that there's some mount_check option plumbing that's missing from recon - might be interesting useful to support the configuration option for dir-check vs mount-point-check in the mount_check utils command | 21:32 |
| *** garthb has quit IRC | 21:33 | |
| clayg | siva_krishnan: I'm not sure how all the plumbing would work out - but yeah I also noticed my dev was (correctly!) telling me that none of the "devices" I have in the ring were real physical mounted devices | 21:33 |
| clayg | siva_krishnan: i also have a notion that it would be useful to update vsaio to carve up the loopback device with lvm - so that I can have real mount-points and one less difference between dev and real deployments | 21:34 |
| ahale | recon saying all drives unmounted sounds like a real deployment | 21:35 |
| notmyname | heh | 21:36 |
| siva_krishnan | sure clayg. will look into SAIO docs once I am done with this | 21:36 |
| notmyname | I know this has been asked before, but has anyone looked at swift under pypy recently? | 21:38 |
| * notmyname asked, because he has | 21:38 | |
| clayg | notmyname: not recently | 21:38 |
| pdardeau | siva_krishnan: i think clayg was referring to vagrant setup scripts for saio | 21:39 |
| clayg | pdardeau: well I mean we could document an lvm setup too - or maybe there's a way to do multiple loopback devices - or you could use virtualbox storage volumes | 21:40 |
| siva_krishnan | pdardeau noted :) | 21:42 |
| notmyname | initial pypy results were pretty good in some tests I ran this week. just changing the proxy server to run under pypy, and I got 230% puts-per-second | 21:47 |
| notmyname | further tests are required, of course, but the initial results are pretty promising | 21:52 |
| pdardeau | notmyname: interesting. any idea how it compares to some of the go stuff that rax has? | 21:53 |
| notmyname | (of course that's the first question) ;-) | 21:53 |
| notmyname | pdardeau: no. I haven't run any of the golang object server perf tests (but I also need a lot of the stuff that isn't implemented in hummingbird) | 21:54 |
| openstackgerrit | Sivasathurappan Radhakrishnan proposed openstack/swift: Added unit test cases for cli/recon.py https://review.openstack.org/282547 | 22:01 |
| notmyname | pdardeau: do you have any relative perf numbers on that? | 22:03 |
| *** asettle has joined #openstack-swift | 22:07 | |
| notmyname | pdardeau: also, I went from 800+ms time to first byte with cpython to 500ms ttfb on pypy (for reads) | 22:08 |
| pdardeau | notmyname: no idea on go implementation improvement numbers | 22:08 |
| pdardeau | notmyname: but vaguely recall scotticus saying that it was significant for their workload | 22:09 |
| pdardeau | notmyname: interesting on the pypy speedup | 22:09 |
| portante | notmyname: what did it take to run the proxy under pypy? | 22:10 |
| portante | what it is a large change? | 22:10 |
| *** csmart has quit IRC | 22:10 | |
| portante | easy or hard to deploy? | 22:10 |
| *** csmart has joined #openstack-swift | 22:11 | |
| *** rickyrem has quit IRC | 22:11 | |
| *** asettle has quit IRC | 22:11 | |
| *** changbl has quit IRC | 22:12 | |
| *** petertr7 is now known as petertr7_away | 22:13 | |
| notmyname | portante: really pretty easy, actually. | 22:16 |
| notmyname | portante: basically, there are a couple of c modules that had to be recompiled (or relinked) for pypy. and then it's a matter of calling pypy instead of cpython and making sure the PATH elements are correct | 22:17 |
| notmyname | no code changes in swift | 22:18 |
| notmyname | however, there should be one small patch coming next week that would be required for any prod pypy use. pypy and cpython do GC differently, and there's one place we don't do quite the right thing with a socket and pypy catches it while it works fine in cpython | 22:19 |
| notmyname | portante: speaking of performance stuff, have you done much with cpu frequency locking? if so, have you had any problems with locking each core at max? eg with heat/reliability/etc | 22:21 |
| openstackgerrit | Merged openstack/swift: Fixed log format in object auditor https://review.openstack.org/281947 | 22:21 |
| *** garthb__ has joined #openstack-swift | 22:29 | |
| *** haomaiwang has joined #openstack-swift | 22:29 | |
| *** garthb_ has quit IRC | 22:32 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 22:32 |
| *** haomaiwang has quit IRC | 22:35 | |
| *** diazjf1 has joined #openstack-swift | 22:36 | |
| *** diazjf has quit IRC | 22:38 | |
| *** changbl has joined #openstack-swift | 22:40 | |
| *** StraubTW_ has quit IRC | 22:50 | |
| *** david-lyle has quit IRC | 22:50 | |
| *** david-lyle has joined #openstack-swift | 22:51 | |
| *** bjkeller has left #openstack-swift | 22:51 | |
| *** david-lyle has quit IRC | 22:54 | |
| *** david-lyle has joined #openstack-swift | 22:54 | |
| *** haomaiwang has joined #openstack-swift | 23:09 | |
| *** haomaiwang has quit IRC | 23:14 | |
| mattoliverau | acoles_: yeah you should move to mattoliverau-land :) | 23:25 |
| clayg | mattoliverau: oh goodness - delusions of gradeur much? | 23:25 |
| clayg | mattoliverau: oh - did I read you right you won't make it to bristol?! | 23:26 |
| mattoliverau | clayg: nope :( Baby's due around then, and remaining married is important to me. | 23:26 |
| * clayg 's heart breaks | 23:26 | |
| clayg | what about *our* love matt!? | 23:27 |
| mattoliverau | clayg: wouldn't it be funny if you came to Oz and found out I had my own province. :P | 23:27 |
| clayg | mattoliverau: lol, *yes* | 23:27 |
| mattoliverau | clayg: sorry man, our love is important, but you ain't pregnant with my child :P | 23:28 |
| clayg | yeah I can't take this joke any further | 23:28 |
| clayg | i'll miss you bro | 23:28 |
| mattoliverau | clayg: yeah, getting close to some HR line I'm sure :P | 23:28 |
| *** nadeem has quit IRC | 23:29 | |
| *** david-lyle_ has joined #openstack-swift | 23:29 | |
| mattoliverau | clayg: yeah annoyed I can't make it. My bro lives in the UK and was going to be the perfect chance to go meet up with him | 23:29 |
| *** zaitcev has joined #openstack-swift | 23:30 | |
| *** ChanServ sets mode: +v zaitcev | 23:30 | |
| *** david-lyle has quit IRC | 23:30 | |
| *** lpabon has quit IRC | 23:32 | |
| *** mingdang1 has joined #openstack-swift | 23:36 | |
| *** mingdang1 has quit IRC | 23:40 | |
| openstackgerrit | Matthew Oliver proposed openstack/swift: Add concurrent reads option to proxy https://review.openstack.org/117710 | 23:50 |
| mattoliverau | ^^ new concurrecnt gets patch | 23:51 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!