| *** wbhuber has quit IRC | 00:01 | |
| openstackgerrit | Hisashi Osanai proposed openstack/swift: Add metadata size tests on the border https://review.openstack.org/187489 | 00:02 |
|---|---|---|
| *** jkugel has joined #openstack-swift | 00:17 | |
| *** gyee has quit IRC | 00:23 | |
| portante | swifterdarrell: yes, I can appreciate those goals. I'd be interested to know how that turns out. | 00:39 |
| *** breitz1 has quit IRC | 00:45 | |
| *** breitz has joined #openstack-swift | 00:45 | |
| notmyname | hello | 00:59 |
| notmyname | sorry about the last minute meeting canceling. thanks clayg for telling everyone | 00:59 |
| mattoliverau | notmyname: No probs, looks like your been busy bug triaging | 01:09 |
| notmyname | just this morning before my laptop ran out of juice | 01:09 |
| notmyname | I hope that will become a normal thing, though. depends how much I'll get bothered at the office, though ;-) | 01:10 |
| notmyname | new plan: go to conferences so I have some uninterrupted time to work ;-) | 01:10 |
| mattoliverau | Lol | 01:11 |
| mattoliverau | Good plan | 01:11 |
| notmyname | mattoliverau: speaking of which, LCA CFP opened up! | 01:13 |
| *** openstack has joined #openstack-swift | 01:23 | |
| *** openstack has joined #openstack-swift | 01:38 | |
| mattoliverau | notmyname: have fun | 01:45 |
| *** openstack has quit IRC | 01:52 | |
| *** openstack has joined #openstack-swift | 01:54 | |
| *** blmartin_ has quit IRC | 01:55 | |
| *** wbhuber has joined #openstack-swift | 02:05 | |
| *** kota_ has joined #openstack-swift | 02:08 | |
| *** bill_az has quit IRC | 02:11 | |
| *** jkugel has left #openstack-swift | 02:16 | |
| wbhuber | kota_: do you happen to have status for this bug: https://bugs.launchpad.net/swift/+bug/1457691 | 02:20 |
| openstack | Launchpad bug 1457691 in OpenStack Object Storage (swift) "Insufficient Fragments occurred on EC GET object" [Undecided,New] | 02:20 |
| wbhuber | kota_: if not i can go take a crack at it | 02:21 |
| kota_ | wbhuber: hi, wait a second | 02:22 |
| wbhuber | kota_: absolutely. | 02:23 |
| kota_ | wbhuber: I didn't start to work about it yet. If you could take it, it seems good to me :) | 02:24 |
| wbhuber | kota_: okay, i might need some guidance on the problem recreation though. | 02:27 |
| kota_ | wbhuber: Thanks! If you need a help, please feel free to ask me ;) | 02:28 |
| wbhuber | kota_: ok, i will leave a comment in the bug report that i'm starting investigation on it | 02:30 |
| *** zaitcev has quit IRC | 03:12 | |
| *** wbhuber has quit IRC | 03:22 | |
| *** asettle-afk is now known as asettle | 03:45 | |
| *** EmilienM|off has quit IRC | 03:55 | |
| *** clyps___1 has quit IRC | 03:55 | |
| *** ppai has joined #openstack-swift | 03:58 | |
| *** EmilienM has joined #openstack-swift | 04:00 | |
| *** clyps___1 has joined #openstack-swift | 04:02 | |
| *** sc has quit IRC | 04:20 | |
| *** mordred has quit IRC | 04:21 | |
| *** sc has joined #openstack-swift | 04:23 | |
| *** mordred has joined #openstack-swift | 04:25 | |
| *** sorrison_laptop has quit IRC | 04:29 | |
| *** ChanServ sets mode: +v kota_ | 04:38 | |
| notmyname | I'd like to welcome kota_ as a new swift core reviewer | 04:42 |
| kota_ | yey, much appreciated! | 04:42 |
| dmorita | Congrats! kota_! | 04:43 |
| kota_ | dmorita: Thanks :) | 04:43 |
| *** sorrison_laptop has joined #openstack-swift | 04:50 | |
| openstackgerrit | Janie Richling proposed openstack/swift: WIP - working on the encryption feature. https://review.openstack.org/157907 | 04:57 |
| jrichli | kota: congrats! | 04:57 |
| jrichli | kota_ ^^ | 04:57 |
| kota_ | jrichli: Thanks :) | 04:57 |
| *** jrichli has quit IRC | 05:00 | |
| *** nadeem has joined #openstack-swift | 05:08 | |
| *** nadeem has quit IRC | 05:08 | |
| *** [1]dimasot has joined #openstack-swift | 05:32 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:40 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:40 | |
| *** [2]dimasot has joined #openstack-swift | 05:47 | |
| *** ppai has quit IRC | 05:48 | |
| *** [1]dimasot has quit IRC | 05:49 | |
| *** Triveni has joined #openstack-swift | 05:55 | |
| *** ppai has joined #openstack-swift | 06:02 | |
| *** [1]dimasot has joined #openstack-swift | 06:21 | |
| *** [2]dimasot has quit IRC | 06:24 | |
| *** flaper87 has quit IRC | 06:33 | |
| *** flaper87 has joined #openstack-swift | 06:33 | |
| mattoliverau | kota_: Woo!! Congrats! | 06:40 |
| kota_ | mattoliverau: Thanks! I'm very happy! | 06:41 |
| mattoliverau | kota_: Next panel, you'll be the new core guy ;) | 06:42 |
| kota_ | mattoliverau: absolutely, I can introduce myself as so :) | 06:43 |
| *** chenhuayi has joined #openstack-swift | 06:55 | |
| *** chenhuayi has quit IRC | 07:09 | |
| *** holahola has joined #openstack-swift | 07:11 | |
| holahola | a question. Is it possible to backup openstack swift data with TSM ? http://docs.openstack.org/juno/config-reference/content/swift-backup-driver.html | 07:12 |
| holahola | i find a general driver. and a tsm driver for cinder volumes | 07:12 |
| cschwede | Good Morning! | 07:20 |
| cschwede | kota_: congrats :D | 07:20 |
| kota_ | cschwede: good morning and thanks :D | 07:21 |
| *** ppai has quit IRC | 07:21 | |
| *** ppai has joined #openstack-swift | 07:35 | |
| *** chlong has quit IRC | 07:44 | |
| *** jistr has joined #openstack-swift | 07:50 | |
| *** jistr is now known as jistr|mt | 07:54 | |
| *** jistr|mt is now known as jistr|mtg | 07:54 | |
| *** dimasot has joined #openstack-swift | 08:00 | |
| *** [1]dimasot has quit IRC | 08:01 | |
| *** aix has joined #openstack-swift | 08:03 | |
| *** km has quit IRC | 08:06 | |
| *** km has joined #openstack-swift | 08:07 | |
| ho | kota_: Congrats! | 08:08 |
| *** acoles_away is now known as acoles | 08:11 | |
| acoles | morning | 08:11 |
| acoles | kota_: welcome and congratulations! | 08:12 |
| kota_ | acoles, ho: thank you so much! | 08:20 |
| ho | kota_: thanks for your first +2 :-) | 08:21 |
| acoles | ho kota_ : heh, quick to get going :) | 08:22 |
| *** mmcardle has joined #openstack-swift | 08:24 | |
| *** jordanP has joined #openstack-swift | 08:27 | |
| *** mmcardle has quit IRC | 08:34 | |
| *** mmcardle has joined #openstack-swift | 08:45 | |
| *** km has quit IRC | 08:59 | |
| *** km has joined #openstack-swift | 08:59 | |
| acoles | torgomatic: heh, i spent several years working on a project called 'sup' so did a double take when scrolling back | 09:01 |
| *** [1]dimasot has joined #openstack-swift | 09:07 | |
| *** mandarine has joined #openstack-swift | 09:08 | |
| *** dimasot has quit IRC | 09:11 | |
| *** [1]dimasot is now known as dimasot | 09:11 | |
| *** ho has quit IRC | 09:19 | |
| *** jistr|mtg is now known as jistr | 09:19 | |
| *** dmorita has quit IRC | 09:28 | |
| *** ppai has quit IRC | 09:36 | |
| *** Trixboxer has joined #openstack-swift | 09:43 | |
| *** sorrison_laptop has quit IRC | 09:45 | |
| *** geaaru has joined #openstack-swift | 09:50 | |
| *** ppai has joined #openstack-swift | 09:50 | |
| *** kota_ has quit IRC | 09:51 | |
| *** ahale has joined #openstack-swift | 10:03 | |
| *** sorrison_laptop has joined #openstack-swift | 10:15 | |
| *** ppai has quit IRC | 10:29 | |
| holahola | is there anyone who know when tivoli storage manager will start to support to store data on Swift ? | 10:33 |
| holahola | tried cloudberry, that was easy and working very well. | 10:33 |
| cschwede | holahola: most likely someone from IBM can answer that. I’m curious, are there plans to add Swift support to TSM? | 10:40 |
| *** sorrison_laptop has quit IRC | 10:40 | |
| *** ppai has joined #openstack-swift | 10:42 | |
| *** sorriso__ has joined #openstack-swift | 10:43 | |
| *** sorriso__ has quit IRC | 10:44 | |
| *** marzif_ has joined #openstack-swift | 10:47 | |
| *** marzif_ has quit IRC | 10:50 | |
| *** marzif_ has joined #openstack-swift | 10:50 | |
| holahola | rumors say so | 10:57 |
| holahola | a version in late fall | 10:57 |
| holahola | this yar | 10:57 |
| holahola | this year | 10:57 |
| holahola | 1q 2016 will tsm have s3 support as well | 10:57 |
| *** csmart has quit IRC | 11:05 | |
| *** csmart has joined #openstack-swift | 11:07 | |
| *** aix has quit IRC | 11:30 | |
| *** Triveni has quit IRC | 11:33 | |
| *** ppai has quit IRC | 11:35 | |
| *** tellesnobrega has joined #openstack-swift | 11:47 | |
| *** dimasot has quit IRC | 11:48 | |
| *** ppai has joined #openstack-swift | 11:49 | |
| *** mmcardle has quit IRC | 11:58 | |
| *** mmcardle has joined #openstack-swift | 11:58 | |
| *** sorrison_laptop has joined #openstack-swift | 11:59 | |
| *** mmcardle has quit IRC | 12:03 | |
| openstackgerrit | Merged openstack/python-swiftclient: Allow reading from object body on download https://review.openstack.org/155291 | 12:08 |
| *** km has quit IRC | 12:10 | |
| *** chlong has joined #openstack-swift | 12:19 | |
| openstackgerrit | Merged openstack/swift: EC: support multiple ranges for GET requests https://review.openstack.org/173497 | 12:24 |
| *** thurloat_isgone is now known as thurloat | 12:28 | |
| *** aix has joined #openstack-swift | 12:30 | |
| *** ppai has quit IRC | 12:36 | |
| openstackgerrit | Christian Schwede proposed openstack/python-swiftclient: Run functional tests using keystone auth options https://review.openstack.org/123514 | 12:42 |
| *** holahola has quit IRC | 12:43 | |
| *** jkugel has joined #openstack-swift | 12:44 | |
| *** marzif_ has quit IRC | 12:44 | |
| *** fanyaohong has joined #openstack-swift | 12:44 | |
| *** jkugel1 has joined #openstack-swift | 12:45 | |
| *** sorrison_laptop has quit IRC | 12:47 | |
| *** kei_yama has quit IRC | 12:48 | |
| *** ppai has joined #openstack-swift | 12:48 | |
| *** jkugel has quit IRC | 12:49 | |
| acoles | cschwede: thx for rebase. actually, when i saw that both func test runs were running all 36 tests was when i realised we need the passenv trick :) so we should now see one run with 24 skips and one with no skips. | 12:51 |
| openstackgerrit | Christian Schwede proposed openstack/python-swiftclient: Run functional tests using keystone auth options https://review.openstack.org/123514 | 12:55 |
| cschwede | acoles: you’re welcome. same for me, i was looking at the code, then at the logs - and realized something is missing there :) | 12:56 |
| cschwede | i just realized that i rebased only against master, not against the outstanding tox.ini patch. so i fixed that now | 12:56 |
| acoles | cschwede: yes ^^ thats makes sense :) | 12:56 |
| acoles | was just typing the same | 12:56 |
| *** delatte has joined #openstack-swift | 12:58 | |
| *** cdelatte has quit IRC | 12:59 | |
| *** delattec has quit IRC | 12:59 | |
| *** delattec has joined #openstack-swift | 13:00 | |
| *** delatte has quit IRC | 13:02 | |
| *** petertr7_away is now known as petertr7 | 13:15 | |
| *** nileshb has joined #openstack-swift | 13:20 | |
| *** Triveni has joined #openstack-swift | 13:24 | |
| *** Triveni has quit IRC | 13:29 | |
| *** odsail has joined #openstack-swift | 13:49 | |
| *** nileshb has quit IRC | 13:57 | |
| *** blmartin has joined #openstack-swift | 14:01 | |
| *** blmartin_ has joined #openstack-swift | 14:01 | |
| openstackgerrit | pradeep kumar singh proposed openstack/python-swiftclient: SwiftClient object upload beginning with / or "./" https://review.openstack.org/185269 | 14:02 |
| *** mandarine has left #openstack-swift | 14:05 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 14:05 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 14:05 |
| *** odsail has quit IRC | 14:08 | |
| *** wbhuber has joined #openstack-swift | 14:19 | |
| *** SkyRocknRoll has quit IRC | 14:20 | |
| *** acampbel11 has joined #openstack-swift | 14:21 | |
| *** acampbell has joined #openstack-swift | 14:21 | |
| *** jrichli has joined #openstack-swift | 14:27 | |
| *** mwheckmann has joined #openstack-swift | 14:31 | |
| *** jrichli has quit IRC | 14:32 | |
| *** odsail has joined #openstack-swift | 14:35 | |
| *** acampbel11 has quit IRC | 14:37 | |
| *** acampbell has quit IRC | 14:37 | |
| openstackgerrit | Darrell Bishop proposed openstack/swift: Allow 1+ object-servers-per-disk deployment https://review.openstack.org/184189 | 14:43 |
| swifterdarrell | clayg: notmyname: cschwede: acoles: tdasilva: portante: zaitcev: "not any worse and slightly better when healthy" benchmarks for servers_per_port patch: https://gist.github.com/dbishop/c214f89ca708a6b1624a#file-summary-md | 14:46 |
| *** chlong has quit IRC | 14:47 | |
| *** chenhuayi has joined #openstack-swift | 14:47 | |
| swifterdarrell | clayg: notmyname: cschwede: acoles: tdasilva: portante: zaitcev: going to investigate any difference with 2 of 30 disks (7%) hammered with I/O (one from an unleashed object-auditor and 1 from some kind of synthetic filesystem benchmark program) | 14:49 |
| *** SkyRocknRoll has joined #openstack-swift | 14:52 | |
| *** fanyaohong has quit IRC | 14:53 | |
| *** mwheckmann has quit IRC | 14:55 | |
| *** lpabon has joined #openstack-swift | 15:00 | |
| *** SkyRocknRoll has quit IRC | 15:00 | |
| *** mwheckmann has joined #openstack-swift | 15:01 | |
| *** chenhuayi has quit IRC | 15:02 | |
| *** SkyRocknRoll has joined #openstack-swift | 15:04 | |
| *** SkyRocknRoll has joined #openstack-swift | 15:04 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add metadata size tests on the border https://review.openstack.org/187489 | 15:08 |
| *** jrichli has joined #openstack-swift | 15:09 | |
| *** annegentle has joined #openstack-swift | 15:13 | |
| *** silor has joined #openstack-swift | 15:13 | |
| portante | swifterdarrell: nice, that is a good result | 15:19 |
| *** minwoob has joined #openstack-swift | 15:19 | |
| portante | I am guessing that the 90 worker count will track well against the servers-per-port work, but the 30 will be "off" | 15:23 |
| *** elmiko has joined #openstack-swift | 15:28 | |
| elmiko | notmyname: hey, might you have a minute or two to talk about the api-wg? | 15:28 |
| peluse | swifterdarrell, you there? | 15:31 |
| swifterdarrell | peluse: ya! | 15:31 |
| swifterdarrell | peluse: what's up? | 15:31 |
| peluse | wondering about your patch just now, you know ctennis is out here in phoenix all week and we're just about ready to confirm good baseline 3x nubers on a 14 node cluster | 15:32 |
| peluse | not sure how much time we'll have but does it make sense to try and check your patch on this setup? | 15:32 |
| peluse | (we have to do EC first though) | 15:32 |
| swifterdarrell | peluse: I'd consider it a nice-to-have, but IIRC, you guys were short on time relative to what you wanted to benchmark | 15:33 |
| *** gyee has joined #openstack-swift | 15:33 | |
| swifterdarrell | peluse: so I wouldn't consider benchmarking my patch a high priority | 15:33 |
| * portante thinks so, but just increasing worker counts would lend credence to swifterdarrell's work | 15:33 | |
| peluse | yeah, kinda what I figured but since I saw the patch I figured I'd ask :) | 15:33 |
| swifterdarrell | peluse: *nod* thanks! So if you have time and want to fiddle with it, that'd be great, but definitely do your primary objectives first | 15:34 |
| peluse | rock n roll | 15:34 |
| swifterdarrell | portante: I'm excited about my while-disks-are-beat-up testing; I have a bonnie++ killing one disk per node now, and just setting up the ssbench stuff | 15:35 |
| portante | kewl | 15:36 |
| peluse | swifterdarrell, so another question - do you recall when threads_per_disk was introduced (I think by torgomatic) I thought the default was non-zero and the intent was to prevent things from getting backed up from a single slow/hosed disk. Def is now 0 - am I remembering wrong or did just experience/perf work move us to 0? | 15:36 |
| portante | how long until we can see your results ya think? ;) | 15:36 |
| peluse | I ask because we have a node here with a few wonky disks and they were really jacking up cluster performance | 15:37 |
| swifterdarrell | peluse: defautl for threads-per-disk was always 0 (which is good) | 15:37 |
| portante | peluse: try ramping up worker counts to alleviate those effects | 15:37 |
| portante | I think you'll see that by doing so you'll see improvements | 15:37 |
| swifterdarrell | peluse: it's tricky keeping the thread count low (it's the product of disk_count * workers * threads_per_disk); when that gets near or over 1000, everything goes to shit; even before that, threads-per-disk has huge overhead | 15:38 |
| peluse | portante, hey! Well, we were hoping today but we came in this morinng to a small screwball something so need to figure out what happened with the overnight gremlins then we'll be off to the races. If all goes well 3x and EC by EOW (maybe not fully written up but baseline numbers collected) | 15:38 |
| portante | nice | 15:38 |
| portante | torgomatic: did we ever get the splice work into swift? | 15:39 |
| swifterdarrell | portante: eta on next batch of results is probably a couple of days? I am going to use shorter run times and a little more automation 'cause I can't do this forever | 15:39 |
| swifterdarrell | portante: zero-copy (splice) for read path got in, IIRC but not for writes? | 15:39 |
| peluse | portante, I think the GET splice made it in but not PUT | 15:40 |
| swifterdarrell | portante: RHEL 6.x doesn't have the required crypto-socket thing in its kernel, tho | 15:40 |
| portante | swifterdarrell: so that is in play with your testing, right? | 15:40 |
| peluse | yeah, what swifterdarrell said :) | 15:40 |
| portante | you guys should try out RHEL 7, nice performance gains in that environment | 15:41 |
| swifterdarrell | peluse: re: the bad disks screwing you up, that would be a good test bed for servers_per_disk, but as portante said, he and I have seen good results with just cranking the worker count up; I would *not* recommend threads-per-disk due to its overhead | 15:42 |
| swifterdarrell | portante: not using zero-copy stuff atm | 15:42 |
| portante | swifterdarrell, peluse, torgomatic, I ask about the splice thing because another possibility would be to hand off a port to a thread to drive all the I/O instead of passing it through the eventlet hub. | 15:43 |
| swifterdarrell | portante: re: RHEL 7, we do what the customers do, and RHEL customers, almost by definition, are slow-moving ;) we'll obvs get there, but in proportion to customer demand/adoption | 15:43 |
| peluse | swifterdarrell, thanks. yeah this morn that was the first thing I wanted to try but one of guys came in last night to get a few more nodes added and things went haywire (temporarily I hope)... we will likely have to punt on investigating the slow disk thing and just get HW replaced so we can move fwd on BM | 15:43 |
| *** thurloat is now known as thurloat_isgone | 15:43 | |
| peluse | benchmarking that is, not bowel movement :) | 15:44 |
| portante | swifterdarrell: understood about RHEL 7 adoption rates | 15:44 |
| * portante smiles quietly to himself | 15:44 | |
| *** zaitcev has joined #openstack-swift | 15:45 | |
| *** ChanServ sets mode: +v zaitcev | 15:45 | |
| *** jordanP has quit IRC | 15:45 | |
| portante | zaitcev: see above about zero-copy work in read path not in play on RHEL 6 due to missing crypto thingy | 15:46 |
| zaitcev | portante: "above"? | 15:46 |
| portante | oh, you don't use znc? | 15:46 |
| *** nadeem has joined #openstack-swift | 15:48 | |
| *** evanjfraser has quit IRC | 15:53 | |
| *** evanjfraser has joined #openstack-swift | 15:53 | |
| *** nadeem has quit IRC | 15:54 | |
| *** nadeem has joined #openstack-swift | 15:54 | |
| *** nadeem has quit IRC | 15:54 | |
| *** jistr has quit IRC | 16:01 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/174343 | 16:05 |
| *** annegentle has quit IRC | 16:08 | |
| *** jordanP has joined #openstack-swift | 16:18 | |
| *** jordanP has quit IRC | 16:19 | |
| notmyname | good morning | 16:26 |
| notmyname | elmiko: sure, but give me a little bit to get settled. I just got to the office | 16:27 |
| elmiko | notmyname: no worries, i've got a couple meetings atm | 16:27 |
| openstackgerrit | Alistair Coles proposed openstack/swift-specs: Updates to encryption spec https://review.openstack.org/154318 | 16:27 |
| *** silor has quit IRC | 16:28 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 16:30 | |
| peluse | morning | 16:32 |
| *** aix has quit IRC | 16:33 | |
| *** petertr7 is now known as petertr7_away | 16:33 | |
| openstackgerrit | Alistair Coles proposed openstack/swift-specs: Updates to encryption spec https://review.openstack.org/154318 | 16:35 |
| wbhuber | morning | 16:36 |
| openstackgerrit | Alistair Coles proposed openstack/swift-specs: Updates to encryption spec https://review.openstack.org/154318 | 16:36 |
| openstackgerrit | Alistair Coles proposed openstack/swift-specs: Add keymaster discussion to encryption spec https://review.openstack.org/183014 | 16:36 |
| *** marzif_ has joined #openstack-swift | 16:37 | |
| *** breitz has quit IRC | 16:37 | |
| *** breitz has joined #openstack-swift | 16:38 | |
| *** odsail has quit IRC | 16:59 | |
| acoles | torgomatic: any chance you could cast an eye over the spec update in patch 154318, its been hanging around for a while and maybe good enough to land then patch as we iterate on the details? | 17:00 |
| patchbot | acoles: https://review.openstack.org/#/c/154318/ | 17:00 |
| *** marzif_ has quit IRC | 17:00 | |
| *** odsail has joined #openstack-swift | 17:05 | |
| *** ppai has quit IRC | 17:10 | |
| *** marzif_ has joined #openstack-swift | 17:11 | |
| notmyname | oh, cool. I just saw that the multi-range get for EC patch just landed | 17:17 |
| notmyname | new keyboard at work now. yay | 17:18 |
| acoles | notmyname: what do you have? | 17:18 |
| lcurtis | Hello all...is erasure coding fully 'supported' in latest swift release or should it still be considered beta | 17:18 |
| notmyname | acoles: WASD code keyboard http://www.wasdkeyboards.com/index.php/products/code-keyboard/code-87-key-mechanical-keyboard.html | 17:18 |
| notmyname | acoles: I got one at home a while back, and just got a 2nd for work | 17:18 |
| *** haomaiwang has joined #openstack-swift | 17:19 | |
| notmyname | lcurtis: beta | 17:19 |
| lcurtis | thank you notmyname...so best not to run production workloads with it at the moment? | 17:19 |
| *** haomaiw__ has quit IRC | 17:20 | |
| notmyname | lcurtis: I wouldn't. there have been a few patches that have landed since the last release that you'd need in prod, and there is still various testing going on to understand what it does at larger scales and wrt performance | 17:20 |
| acoles | notmyname: nice. i was looking recently - the clear keys the quietest right? | 17:21 |
| notmyname | lcurtis: that being said, I feel a lot better about EC on master right now | 17:21 |
| notmyname | acoles: this one isn't too loud. certainly not silent, but it's not "clicky" | 17:21 |
| lcurtis | notmyname...that is fantastic news | 17:21 |
| lcurtis | keep up the great work! | 17:21 |
| notmyname | hi acoles. i'm making a video now of typing on my new keyboard | 17:22 |
| acoles | notmyname: maybe i should ask your colleagues for a loudness assessment after a day or two ;) | 17:22 |
| notmyname | :-) | 17:22 |
| notmyname | let me figure out how to upload this | 17:23 |
| acoles | i've been using tenkeyless for years but not of a great quality | 17:24 |
| notmyname | acoles: http://d.not.mn/typing.m4v | 17:24 |
| acoles | wow thats quiet! | 17:25 |
| lcurtis | notmyname will EC be an in-place upgrade? | 17:25 |
| acoles | oh wait, let me un-mute... | 17:25 |
| *** Viswanath has joined #openstack-swift | 17:25 | |
| notmyname | lcurtis: what do you mean? | 17:25 |
| notmyname | acoles: lol | 17:25 |
| *** petertr7_away is now known as petertr7 | 17:26 | |
| lcurtis | is it something that can be enabled and replicas will be chunked automatically? | 17:26 |
| notmyname | lcurtis: it's a storage policy. so you can have either (or both) EC and replication | 17:27 |
| notmyname | but it's set on a per-container level | 17:27 |
| notmyname | at the time the container is created | 17:27 |
| lcurtis | ah..so cannot retroactively perform on container? | 17:27 |
| *** acampbel11 has joined #openstack-swift | 17:28 | |
| *** acampbell has joined #openstack-swift | 17:28 | |
| *** Viswanath has quit IRC | 17:30 | |
| *** SkyRocknRoll_ has quit IRC | 17:30 | |
| *** doxavore has joined #openstack-swift | 17:30 | |
| acoles | notmyname: i thought about taking on the challenge of a ninja https://www.diatec.co.jp/en/det.php?prod_c=775 but don't think my touch typing is quite there yet ;) | 17:30 |
| notmyname | check out the wasd keyboards site. they've got a ton of configuration options | 17:33 |
| *** SkyRocknRoll_ has joined #openstack-swift | 17:33 | |
| acoles | yeah and i read about retrofit collars you can fit to reduce travel etc | 17:34 |
| *** delattec has quit IRC | 17:34 | |
| *** cdelatte has joined #openstack-swift | 17:34 | |
| *** geaaru has quit IRC | 17:35 | |
| notmyname | yeah, the o-rings | 17:35 |
| notmyname | I ended up getting the clear switches because they are rated at the same force as the keyboards I was used to (macbook/apple) | 17:35 |
| notmyname | 65N | 17:36 |
| acoles | peluse: wow swift-dispersion-report was totally broken if default wasn't policy 0 | 17:38 |
| acoles | notmyname: makes sense | 17:38 |
| acoles | peluse: (just looking at your patch) | 17:38 |
| *** haomaiwa_ has joined #openstack-swift | 17:39 | |
| elmiko | notmyname: hey, finally got free. can you spare a few minutes? | 17:39 |
| *** cdelatte has quit IRC | 17:39 | |
| *** haomaiwang has quit IRC | 17:41 | |
| notmyname | elmiko: sure. what's up? | 17:43 |
| notmyname | elmiko: do you want to talk in here? or do you want to video chat? phone? | 17:45 |
| elmiko | notmyname: we can chat here, i just wanted to share a few links | 17:46 |
| notmyname | ok | 17:46 |
| elmiko | so, i'm reaching out to all the listed liaisons | 17:46 |
| elmiko | i want to share this list of proposed liaison responsibilities from the nova team, https://wiki.openstack.org/wiki/Nova/APIWGLiaisons | 17:47 |
| elmiko | and also, this merge process review we are working on, https://review.openstack.org/#/c/186836 | 17:47 |
| *** dontalton has joined #openstack-swift | 17:47 | |
| elmiko | i imagine over the next few weeks we'll be sending some emails to the ML with proposed guidelines that are going into freeze before merge | 17:47 |
| elmiko | no specific action is required from the liaisons, but we'd like to help improve communcation about the upcoming guidelines | 17:48 |
| elmiko | also, if you have any questions/comments/concerns we usually hang out in openstack-api | 17:48 |
| *** gyee has quit IRC | 17:49 | |
| elmiko | not sure if we've met at summit before or not, but my name is michael mccune =) | 17:49 |
| notmyname | I lurk in the -api channel | 17:49 |
| notmyname | I'm looking over the process doc | 17:49 |
| elmiko | cool, thanks! | 17:49 |
| elmiko | yea, i know you've got the global regex for "swift" on freenode ;) | 17:49 |
| notmyname | I do think that keeping the patches in front of people is a good thing | 17:50 |
| notmyname | heh, not quite | 17:50 |
| elmiko | yea, we're hoping that getting more visibility will help | 17:50 |
| elmiko | anyways, that's about all for my song and dance, thanks for your time =) | 17:51 |
| notmyname | elmiko: thanks for the info | 17:51 |
| notmyname | elmiko: what's the current process for the api-wg for when a proposal doesn't match existing APIs? or isn't compatible with an existing API? | 17:52 |
| notmyname | from what I can tell, the wg pretty much says "well this is just guidance" and then nothing happens | 17:52 |
| notmyname | so far :-) | 17:52 |
| elmiko | yea, we try to set out the guidelines in a way that won't obstruct projects that might be doing things differently | 17:53 |
| *** SkyRocknRoll_ has quit IRC | 17:53 | |
| elmiko | i think our preferred engagement method is to help on reviews for new api content | 17:53 |
| elmiko | and we'd like to become a resource that projects can call on if there are questions | 17:53 |
| elmiko | maybe about new versions, or revamps to old code | 17:53 |
| elmiko | we're going with more of the carrot than the stick ;) | 17:53 |
| notmyname | so far the feedback I've gotten is "swift is doing it wrong" | 17:54 |
| elmiko | lol, yea. there's a lot of that going around. i don't necessarily agree | 17:54 |
| elmiko | i mean, you guys have some reasoning as to why you've crafted the api the way it is. | 17:55 |
| elmiko | if the project is open to suggestions, then i'm sure there would be folks in the wg who would want to help, but | 17:55 |
| *** marzif_ has quit IRC | 17:55 | |
| elmiko | if you'd like to keep developing in the manner that's been working, imo we need to respect that | 17:56 |
| notmyname | that's good to hear | 17:56 |
| *** marzif_ has joined #openstack-swift | 17:56 | |
| elmiko | the overarching goal for the wg is help bring the overall apis into a more consistent state, but that does not mean beating projects into submission | 17:56 |
| elmiko | some projects will always have reasons for going against the guidelines, that's fine imo. | 17:57 |
| notmyname | my concern that I've tried to share with the wg is that many things in openstack have started as "here's some helpful guidance that you can choose to use or not" and then turned in to "why aren't you doing things the OpenStack Way(tm)?!" | 17:57 |
| notmyname | which is where the root of my concern about the api-wg comes from | 17:57 |
| notmyname | in general, I think the idea of consistent APIs is wonderful! I have no argument there | 17:58 |
| elmiko | i can't predict how things will go in the future, but i know there are several voices within the wg (mine included) who are against becoming the api police | 17:59 |
| notmyname | and that being said, I'm happy to work with the api-wg as new things are proposed. and we'll use those in swift where we can (ie not gratuitous changes from existing behavior or breaking changes) | 17:59 |
| elmiko | cool, i know we'll be glad to have another voice =) | 18:00 |
| *** annegentle has joined #openstack-swift | 18:01 | |
| *** Viswanath has joined #openstack-swift | 18:01 | |
| *** barra204 has joined #openstack-swift | 18:01 | |
| *** jrichli has quit IRC | 18:03 | |
| elmiko | thanks again notmyname, take care | 18:03 |
| *** Viswanath has quit IRC | 18:04 | |
| *** tellesnobrega_ has joined #openstack-swift | 18:06 | |
| *** barra204 has quit IRC | 18:07 | |
| *** jrichli has joined #openstack-swift | 18:07 | |
| portante | notmyname: can't resist ... who is saying "swift is doing it wrong"? | 18:08 |
| redbo | people who think POST should be the only verb that can create things? | 18:09 |
| *** haomai___ has joined #openstack-swift | 18:09 | |
| *** haomaiwa_ has quit IRC | 18:10 | |
| portante | oh boy, the REST police to the rescue | 18:10 |
| notmyname | lol | 18:11 |
| redbo | To be fair, there are some things swift does wrong. But we still work with clients written like 6.5 years ago, which seems worth keeping a few warts to me. | 18:12 |
| notmyname | portante: redbo: people who don't like us using headers that start with X-. or that want us to have a different path URL for metadata | 18:12 |
| notmyname | redbo: yes, that. exactly | 18:12 |
| *** Fin1te has joined #openstack-swift | 18:12 | |
| portante | sounds like these folks have not developed, deployed, run for a while and maintained real software in production | 18:13 |
| notmyname | I think that might be a little unfair ;-) | 18:13 |
| notmyname | (or ad-hominem) | 18:13 |
| sweeper | strawmen are the funnest | 18:14 |
| portante | if the target of that is the sentiments listed above then I agree, but if that is referring to the comment about real software ... | 18:14 |
| portante | "that" in quotes above ... | 18:14 |
| portante | that is | 18:14 |
| redbo | yeah, who wants to be fair? | 18:14 |
| notmyname | so far, there's a few things that the api-wg have approved that are ok, and there are a few that IMO are pretty bad (at least from swift's perspective). if there is something net-new that we add to swift, the api-wg can offer some guidance on how others also have implemented a similar problem. but the ones that don't work at all, I don't think we should worry about | 18:19 |
| *** annegentle has quit IRC | 18:21 | |
| *** annegentle has joined #openstack-swift | 18:22 | |
| *** kr4zy has joined #openstack-swift | 18:22 | |
| *** MVenesio has joined #openstack-swift | 18:23 | |
| *** acoles is now known as acoles_away | 18:24 | |
| *** silor has joined #openstack-swift | 18:27 | |
| clayg | lol @ redbo | 18:28 |
| *** tellesnobrega_ has quit IRC | 18:29 | |
| *** tellesnobrega_ has joined #openstack-swift | 18:29 | |
| *** tellesnobrega_ has quit IRC | 18:29 | |
| *** tellesnobrega_ has joined #openstack-swift | 18:32 | |
| *** tellesnobrega_ has quit IRC | 18:32 | |
| *** friedchicken has joined #openstack-swift | 18:36 | |
| *** dimasot has joined #openstack-swift | 18:38 | |
| *** openstackstatus has joined #openstack-swift | 18:40 | |
| *** ChanServ sets mode: +v openstackstatus | 18:40 | |
| *** tellesnobrega_ has joined #openstack-swift | 18:41 | |
| -openstackstatus- NOTICE: Gerrit has been restarted to clear an issue with its event stream. Any change events between 17:25 and 18:38 UTC should be rechecked or have their approvals reapplied to initiate testing. | 18:42 | |
| *** tellesnobrega_ has quit IRC | 18:43 | |
| *** silor has quit IRC | 18:45 | |
| *** tellesnobrega_ has joined #openstack-swift | 18:45 | |
| *** marzif_ has quit IRC | 18:47 | |
| *** Fin1te has quit IRC | 18:51 | |
| *** elmiko is now known as _elmiko | 18:53 | |
| *** SkyRocknRoll has quit IRC | 18:53 | |
| *** kr4zy has quit IRC | 18:54 | |
| *** mmcardle has joined #openstack-swift | 18:56 | |
| *** Fin1te has joined #openstack-swift | 18:57 | |
| *** Fin1te has quit IRC | 19:00 | |
| *** marzif_ has joined #openstack-swift | 19:04 | |
| *** Fin1te has joined #openstack-swift | 19:05 | |
| *** annegentle has quit IRC | 19:07 | |
| *** petertr7 is now known as petertr7_away | 19:16 | |
| *** Viswanath has joined #openstack-swift | 19:18 | |
| *** mjseger has quit IRC | 19:20 | |
| *** petertr7_away is now known as petertr7 | 19:21 | |
| *** Viswanath has quit IRC | 19:21 | |
| *** kr4zy has joined #openstack-swift | 19:21 | |
| kr4zy | For 1G NICs server, is ~1.62/sec for object-replicator a normal speed? | 19:22 |
| *** Fin1te has quit IRC | 19:24 | |
| *** mmcardle has quit IRC | 19:31 | |
| notmyname | kr4zy: there's not a "normal" since it's so dependent on hardware. however, while the replication network speed does matter, normally replication is limited by disk IO | 19:32 |
| jrichli | torgomatic: glad to see the +2 on the spec! Thanks for taking a look | 19:33 |
| notmyname | kr4zy: so the answer is really that you should look at what is normal to you instead of normal for anyone else | 19:33 |
| *** csmart has quit IRC | 19:35 | |
| *** csmart has joined #openstack-swift | 19:38 | |
| *** csmart has quit IRC | 19:43 | |
| *** csmart has joined #openstack-swift | 19:45 | |
| *** Fin1te has joined #openstack-swift | 19:47 | |
| *** mwheckmann has quit IRC | 19:53 | |
| *** Viswanath has joined #openstack-swift | 19:55 | |
| *** dimasot has quit IRC | 19:58 | |
| *** odsail has quit IRC | 20:00 | |
| *** Viswanath has quit IRC | 20:00 | |
| *** marzif_ has quit IRC | 20:00 | |
| *** mwheckmann has joined #openstack-swift | 20:01 | |
| *** csmart has quit IRC | 20:05 | |
| *** csmart has joined #openstack-swift | 20:05 | |
| *** Viswanath has joined #openstack-swift | 20:07 | |
| *** Viswanath has quit IRC | 20:10 | |
| *** csmart has quit IRC | 20:10 | |
| *** Viswanath has joined #openstack-swift | 20:16 | |
| *** csmart has joined #openstack-swift | 20:16 | |
| clayg | did the guy yesterday with the ssl cert thing ever figure out what keystone was doing? | 20:19 |
| clayg | jlk: <- this guy right? | 20:19 |
| *** Viswanath has quit IRC | 20:20 | |
| jlk | clayg: I haven't yet, I was distracted with other things. | 20:35 |
| *** Viswanath has joined #openstack-swift | 20:44 | |
| peluse | notmyname, can you repost the lcoation of the EC data you showed in vancouver? | 20:45 |
| *** mmcardle has joined #openstack-swift | 20:47 | |
| *** Viswanath has quit IRC | 20:47 | |
| notmyname | peluse: yeah, let me find it. | 20:58 |
| notmyname | also, I need to look and see if there's any newer data | 20:58 |
| notmyname | peluse: http://d.not.mn/20150512_run2_15fullness.tgz | 20:59 |
| notmyname | it's about 200MB | 20:59 |
| *** Fin1te has quit IRC | 21:01 | |
| peluse | thanks | 21:01 |
| *** operator99 is now known as gyee | 21:02 | |
| *** zhill_ has joined #openstack-swift | 21:02 | |
| *** blmartin has quit IRC | 21:06 | |
| *** blmartin_ has quit IRC | 21:07 | |
| *** petertr7 is now known as petertr7_away | 21:09 | |
| *** jrichli has quit IRC | 21:10 | |
| *** nadeem has joined #openstack-swift | 21:16 | |
| *** doxavore has quit IRC | 21:16 | |
| *** tellesnobrega_ has quit IRC | 21:21 | |
| lcurtis | does anyone happen to know unit of measure used in statsd middleware metrics for stats.SERVER.proxy-server.object.PUT.201.xfer | 21:22 |
| lcurtis | trying to figure actual throughput into my cluster | 21:23 |
| lcurtis | or perhaps I can use swift-recon divided by number of replicas? | 21:24 |
| *** acampbel11 has quit IRC | 21:33 | |
| *** acampbell has quit IRC | 21:33 | |
| *** wbhuber has quit IRC | 21:36 | |
| *** lpabon has quit IRC | 21:36 | |
| *** wbhuber has joined #openstack-swift | 21:36 | |
| *** MVenesio has quit IRC | 21:39 | |
| lcurtis | found it..its bytes | 21:42 |
| lcurtis | http://docs.openstack.org/developer/swift/admin_guide.html#reporting-metrics-to-statsd | 21:42 |
| *** mmcardle has quit IRC | 21:46 | |
| *** _elmiko is now known as elmiko | 21:47 | |
| *** elmiko has left #openstack-swift | 21:47 | |
| *** friedchicken has quit IRC | 21:51 | |
| *** kr4zy has quit IRC | 21:56 | |
| *** jkugel1 has quit IRC | 22:03 | |
| *** Fin1te has joined #openstack-swift | 22:07 | |
| *** wbhuber_ has joined #openstack-swift | 22:20 | |
| *** wbhuber has quit IRC | 22:24 | |
| *** Fin1te has quit IRC | 22:25 | |
| *** ndk has quit IRC | 22:27 | |
| clayg | so what name are we likeing for M? | 22:27 |
| *** ndk has joined #openstack-swift | 22:27 | |
| *** sorrison_laptop has joined #openstack-swift | 22:28 | |
| torgomatic | Mongoose! | 22:34 |
| *** dontalton has quit IRC | 22:35 | |
| *** annegentle has joined #openstack-swift | 22:35 | |
| clayg | torgomatic: not on the list -> https://wiki.openstack.org/wiki/Release_Naming/M_Proposals | 22:36 |
| torgomatic | well, it darn well should be :p | 22:36 |
| *** zaitcev has quit IRC | 22:37 | |
| notmyname | I like Metropolis | 22:37 |
| clayg | I like the Monkey -> http://en.wikipedia.org/wiki/Japanese_macaque | 22:38 |
| clayg | how do you pronounce "Macque" | 22:38 |
| notmyname | yeah, that one too :-) | 22:38 |
| notmyname | clayg: click the speaker link on http://www.oxforddictionaries.com/us/definition/american_english/macaque | 22:39 |
| clayg | wow - is it really muh-kawk? | 22:39 |
| clayg | yeah this release should toally be all 'bout muh-kwak | 22:40 |
| clayg | not *your* kawk mind you | 22:40 |
| *** Guest47317 has quit IRC | 22:40 | |
| *** annegentle has quit IRC | 22:40 | |
| *** sorrison_laptop has quit IRC | 22:41 | |
| *** minwoob has quit IRC | 22:41 | |
| *** scott has joined #openstack-swift | 22:41 | |
| *** scott is now known as Guest48719 | 22:41 | |
| *** annegentle has joined #openstack-swift | 22:42 | |
| mattoliverau | Morning | 22:43 |
| *** zhill_ has quit IRC | 22:45 | |
| torgomatic | welp, the monkey has my vote | 22:45 |
| *** zhill_ has joined #openstack-swift | 22:56 | |
| *** km has joined #openstack-swift | 22:58 | |
| *** zhill_ has quit IRC | 23:01 | |
| *** annegentle has quit IRC | 23:08 | |
| clayg | yeah macque - how do we vote | 23:15 |
| *** sorrison_laptop has joined #openstack-swift | 23:16 | |
| *** wbhuber_ has quit IRC | 23:16 | |
| *** zhill_ has joined #openstack-swift | 23:20 | |
| *** kei_yama has joined #openstack-swift | 23:22 | |
| *** zhill_ has quit IRC | 23:24 | |
| *** chlong has joined #openstack-swift | 23:27 | |
| *** odsail has joined #openstack-swift | 23:34 | |
| *** nadeem has quit IRC | 23:35 | |
| *** odsail has quit IRC | 23:37 | |
| *** dmorita has joined #openstack-swift | 23:37 | |
| *** asettle has quit IRC | 23:46 | |
| *** asettle has joined #openstack-swift | 23:48 | |
| *** ho has joined #openstack-swift | 23:52 | |
| *** sorrison_laptop has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!