| *** Suyi_ has quit IRC | 00:05 | |
| *** zul has quit IRC | 00:08 | |
| *** vint_bra has joined #openstack-swift | 00:13 | |
| *** tqtran has quit IRC | 00:25 | |
| *** vint_bra has quit IRC | 00:30 | |
| *** vint_bra has joined #openstack-swift | 00:31 | |
| *** itlinux has quit IRC | 00:40 | |
| *** itlinux has joined #openstack-swift | 00:41 | |
| *** m_kazuhiro has joined #openstack-swift | 00:55 | |
| *** itlinux has quit IRC | 00:56 | |
| *** asettle has joined #openstack-swift | 00:59 | |
| *** itlinux has joined #openstack-swift | 01:00 | |
| *** trananhkma has joined #openstack-swift | 01:02 | |
| *** gyee has quit IRC | 01:03 | |
| *** asettle has quit IRC | 01:04 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: make object server/replicator work with swift-init https://review.openstack.org/369765 | 01:05 |
|---|---|---|
| kota_ | good morning | 01:08 |
| clayg | redbo: can you tell nadeem I need to talk about https://review.openstack.org/#/c/369765/1 when he has some time? | 01:11 |
| patchbot | patch 369765 - swift (feature/hummingbird) - make object server/replicator work with swift-init | 01:11 |
| mattoliverau | kota_: morning | 01:15 |
| kota_ | acoles: thanks for fixing the optimistic-GET patch, will review again soon. | 01:15 |
| kota_ | mattoliverau: morning! | 01:15 |
| *** zhengyin has joined #openstack-swift | 01:18 | |
| *** jamielennox is now known as jamielennox|away | 01:26 | |
| *** mingyu has joined #openstack-swift | 01:32 | |
| *** mingyu has quit IRC | 01:36 | |
| zhengyin | good morning | 01:36 |
| kota_ | zhengyin: good morning | 01:38 |
| zhengyin | kota_: morning | 01:38 |
| *** itlinux has quit IRC | 01:39 | |
| mattoliverau | zhengyin: morning | 01:41 |
| zhengyin | mattoliverau: morning | 01:41 |
| *** itlinux has joined #openstack-swift | 01:46 | |
| *** itlinux has quit IRC | 01:56 | |
| *** zul has joined #openstack-swift | 02:27 | |
| *** zul has quit IRC | 02:28 | |
| *** zul has joined #openstack-swift | 02:28 | |
| *** klrmn has quit IRC | 02:48 | |
| *** stevemar_ is now known as stevemar | 02:49 | |
| charz | morning | 02:56 |
| *** itlinux has joined #openstack-swift | 02:57 | |
| mattoliverau | charz: o/ | 02:59 |
| *** zaitcev has joined #openstack-swift | 03:00 | |
| *** ChanServ sets mode: +v zaitcev | 03:00 | |
| *** zul has quit IRC | 03:05 | |
| *** itlinux has quit IRC | 03:07 | |
| *** mingyu has joined #openstack-swift | 03:08 | |
| *** tqtran has joined #openstack-swift | 03:22 | |
| *** tqtran has quit IRC | 03:26 | |
| *** chsc has joined #openstack-swift | 03:32 | |
| *** chsc has joined #openstack-swift | 03:32 | |
| clayg | made. my. day. https://github.com/gfredericks/quinedb | 03:44 |
| clayg | why *shouldn't* your database be a quine!? | 03:45 |
| *** itlinux has joined #openstack-swift | 03:46 | |
| *** itlinux has quit IRC | 03:56 | |
| *** chsc has quit IRC | 03:58 | |
| *** zaitcev has quit IRC | 04:04 | |
| *** m_kazuhiro has quit IRC | 04:07 | |
| *** itlinux has joined #openstack-swift | 04:37 | |
| *** psachin has joined #openstack-swift | 04:44 | |
| *** itlinux has quit IRC | 04:47 | |
| openstackgerrit | Pallavi proposed openstack/swift: Use ConfigParser instead of SafeConfigParser https://review.openstack.org/369814 | 04:49 |
| *** mingyu has quit IRC | 04:50 | |
| openstackgerrit | Pallavi proposed openstack/python-swiftclient: Use ConfigParser instead of SafeConfigParser https://review.openstack.org/369819 | 04:59 |
| *** Jeffrey4l_ has joined #openstack-swift | 05:00 | |
| *** Jeffrey4l has quit IRC | 05:04 | |
| *** silor has joined #openstack-swift | 05:14 | |
| *** klrmn has joined #openstack-swift | 05:22 | |
| *** silor1 has joined #openstack-swift | 05:23 | |
| *** silor has quit IRC | 05:25 | |
| *** silor1 is now known as silor | 05:25 | |
| *** rcernin has joined #openstack-swift | 05:33 | |
| *** jamielennox|away is now known as jamielennox | 05:40 | |
| *** ppai has joined #openstack-swift | 05:46 | |
| *** itlinux has joined #openstack-swift | 05:47 | |
| *** ppai has quit IRC | 05:48 | |
| *** klrmn has quit IRC | 05:57 | |
| *** ppai has joined #openstack-swift | 05:59 | |
| *** itlinux has quit IRC | 06:01 | |
| *** ChubYann has quit IRC | 06:03 | |
| *** mingyu has joined #openstack-swift | 06:07 | |
| *** mingyu has quit IRC | 06:12 | |
| *** ChubYann has joined #openstack-swift | 06:18 | |
| *** 1JTAA9HZM is now known as mohit | 06:29 | |
| *** mohit is now known as Guest59079 | 06:29 | |
| *** pcaruana has joined #openstack-swift | 06:32 | |
| *** aswadr_ has joined #openstack-swift | 06:34 | |
| *** itlinux has joined #openstack-swift | 06:35 | |
| *** silor has quit IRC | 06:39 | |
| *** itlinux has quit IRC | 06:46 | |
| *** hseipp has joined #openstack-swift | 06:47 | |
| *** tesseract- has joined #openstack-swift | 06:57 | |
| *** jistr is now known as jistr|mtgs | 07:09 | |
| *** jistr|mtgs is now known as jistr|mtg | 07:09 | |
| *** mingyu has joined #openstack-swift | 07:20 | |
| *** rledisez has joined #openstack-swift | 07:20 | |
| *** tqtran has joined #openstack-swift | 07:26 | |
| *** tqtran has quit IRC | 07:31 | |
| *** oshritf has joined #openstack-swift | 07:36 | |
| *** oshritf has quit IRC | 07:41 | |
| *** sams-gleb has joined #openstack-swift | 07:44 | |
| *** oshritf has joined #openstack-swift | 07:46 | |
| *** oshritf_ has joined #openstack-swift | 07:46 | |
| *** oshritf has left #openstack-swift | 07:47 | |
| *** onovy has quit IRC | 07:51 | |
| *** peterlisak has quit IRC | 07:52 | |
| *** itlinux has joined #openstack-swift | 07:55 | |
| *** oshritf_ has quit IRC | 08:02 | |
| *** permalac has joined #openstack-swift | 08:03 | |
| *** ntata has joined #openstack-swift | 08:04 | |
| *** itlinux has quit IRC | 08:05 | |
| *** tries_ has joined #openstack-swift | 08:17 | |
| *** manous has joined #openstack-swift | 08:23 | |
| *** asettle has joined #openstack-swift | 08:29 | |
| *** joeljwright has joined #openstack-swift | 08:29 | |
| *** ChanServ sets mode: +v joeljwright | 08:29 | |
| *** peterlisak has joined #openstack-swift | 08:30 | |
| *** onovy has joined #openstack-swift | 08:30 | |
| *** takashi has joined #openstack-swift | 08:33 | |
| openstackgerrit | Donagh McCabe proposed openstack/swift: Corrections for the API specifications in api-ref https://review.openstack.org/354767 | 08:50 |
| *** acoles is now known as acoles_ | 09:02 | |
| *** kei_yama has quit IRC | 09:02 | |
| *** itlinux has joined #openstack-swift | 09:06 | |
| *** donagh has joined #openstack-swift | 09:11 | |
| *** itlinux has quit IRC | 09:16 | |
| openstackgerrit | Donagh McCabe proposed openstack/swift: Corrections for the API specifications in api-ref https://review.openstack.org/354767 | 09:16 |
| *** mvk has quit IRC | 09:19 | |
| *** tqtran has joined #openstack-swift | 09:28 | |
| *** acoles_ is now known as acoles | 09:28 | |
| *** peterlisak has quit IRC | 09:29 | |
| *** onovy has quit IRC | 09:29 | |
| *** tqtran has quit IRC | 09:32 | |
| *** stack_ has joined #openstack-swift | 09:35 | |
| *** permalac has quit IRC | 09:38 | |
| *** mingyu_ has joined #openstack-swift | 09:41 | |
| *** stack__ has joined #openstack-swift | 09:45 | |
| *** mingyu has quit IRC | 09:45 | |
| *** trananhkma has quit IRC | 09:45 | |
| *** stack_ has quit IRC | 09:45 | |
| *** stack__ has quit IRC | 09:46 | |
| *** trananhkma has joined #openstack-swift | 09:46 | |
| *** daemontool has joined #openstack-swift | 09:46 | |
| *** mvk has joined #openstack-swift | 09:52 | |
| *** trananhkma has quit IRC | 09:58 | |
| *** bikmak has quit IRC | 10:03 | |
| *** takashi has quit IRC | 10:11 | |
| *** silor has joined #openstack-swift | 10:16 | |
| *** itlinux has joined #openstack-swift | 10:17 | |
| *** mingyu_ has quit IRC | 10:20 | |
| *** silor has quit IRC | 10:25 | |
| *** itlinux has quit IRC | 10:28 | |
| *** cbartz has joined #openstack-swift | 10:43 | |
| *** silor has joined #openstack-swift | 10:46 | |
| *** silor has quit IRC | 10:54 | |
| *** silor has joined #openstack-swift | 10:55 | |
| *** itlinux has joined #openstack-swift | 10:58 | |
| *** cdelatte has joined #openstack-swift | 11:00 | |
| *** itlinux has quit IRC | 11:09 | |
| *** oshritf has joined #openstack-swift | 11:09 | |
| *** oshritf_ has joined #openstack-swift | 11:09 | |
| *** thebloggu has joined #openstack-swift | 11:12 | |
| *** oshritf_ has left #openstack-swift | 11:18 | |
| *** oshritf__ has joined #openstack-swift | 11:26 | |
| *** oshritf has quit IRC | 11:27 | |
| *** tqtran has joined #openstack-swift | 11:29 | |
| *** tqtran has quit IRC | 11:34 | |
| *** openstackgerrit has quit IRC | 11:34 | |
| *** openstackgerrit has joined #openstack-swift | 11:35 | |
| *** silor has quit IRC | 11:40 | |
| *** asettle has quit IRC | 11:42 | |
| *** asettle has joined #openstack-swift | 11:43 | |
| *** acoles is now known as acoles_ | 11:45 | |
| *** asettle has quit IRC | 11:47 | |
| *** acoles_ is now known as acoles | 11:53 | |
| *** manous has quit IRC | 11:54 | |
| *** asettle has joined #openstack-swift | 12:07 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: EC Fragment Duplication - Foundational Global EC Cluster Support https://review.openstack.org/219165 | 12:08 |
| kota_ | rebased | 12:08 |
| *** itlinux has joined #openstack-swift | 12:09 | |
| *** ppai has quit IRC | 12:11 | |
| *** psachin has quit IRC | 12:13 | |
| *** psachin has joined #openstack-swift | 12:18 | |
| *** thebloggu has quit IRC | 12:19 | |
| *** itlinux has quit IRC | 12:20 | |
| *** gabor_antal has quit IRC | 12:43 | |
| *** dabukalam_ is now known as dabukalam | 12:48 | |
| *** MVenesio has joined #openstack-swift | 12:54 | |
| *** MVenesio has quit IRC | 12:55 | |
| *** StraubTW has joined #openstack-swift | 13:09 | |
| *** peterlisak has joined #openstack-swift | 13:12 | |
| *** onovy has joined #openstack-swift | 13:12 | |
| *** itlinux has joined #openstack-swift | 13:20 | |
| *** jistr|mtg is now known as jistr | 13:24 | |
| *** tqtran has joined #openstack-swift | 13:31 | |
| *** itlinux has quit IRC | 13:34 | |
| *** zhugaoxiao has quit IRC | 13:35 | |
| *** tqtran has quit IRC | 13:35 | |
| *** vint_bra has joined #openstack-swift | 13:42 | |
| *** tongli has joined #openstack-swift | 13:43 | |
| *** jistr is now known as jistr|mtg | 13:49 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: EC Fragment Duplication - Foundational Global EC Cluster Support https://review.openstack.org/219165 | 13:50 |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: EC Fragment Duplication - Foundational Global EC Cluster Support https://review.openstack.org/219165 | 13:59 |
| *** chsc has joined #openstack-swift | 14:01 | |
| *** chsc has joined #openstack-swift | 14:01 | |
| *** CaioBrentano has quit IRC | 14:01 | |
| *** CaioBrentano has joined #openstack-swift | 14:01 | |
| *** mingyu has joined #openstack-swift | 14:30 | |
| *** hseipp has quit IRC | 14:33 | |
| *** itlinux has joined #openstack-swift | 14:35 | |
| *** sgundur_ has joined #openstack-swift | 14:39 | |
| *** tgtanya has joined #openstack-swift | 14:39 | |
| *** _JZ_ has joined #openstack-swift | 14:41 | |
| *** hseipp has joined #openstack-swift | 14:47 | |
| *** mingyu has quit IRC | 14:53 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Make container sync copy SLO manifests https://review.openstack.org/348495 | 14:53 |
| *** mingyu has joined #openstack-swift | 14:54 | |
| *** nadeem has joined #openstack-swift | 15:00 | |
| *** jistr|mtg is now known as jistr | 15:01 | |
| pdardeau | clayg: nadeem is out today iirc | 15:04 |
| pdardeau | clayg: sorry, he was out Monday and Tuesday | 15:05 |
| nadeem | I have extended my vacation till today as my brother is visiting me from out of town. | 15:06 |
| nadeem | clayg we can talk about it tomorrow | 15:06 |
| *** tgtanya has quit IRC | 15:16 | |
| *** nadeem has quit IRC | 15:19 | |
| notmyname | good morning | 15:21 |
| notmyname | mattoliverau: you're talking at LCA, too? cool! | 15:22 |
| openstackgerrit | Alistair Coles proposed openstack/python-swiftclient: Make tempurl command check for valid object path https://review.openstack.org/367377 | 15:24 |
| *** psachin has quit IRC | 15:30 | |
| *** sams-gleb has quit IRC | 15:31 | |
| notmyname | hmm...that can't be right. I don't have hundreds on unread messages in the openstack ML from overnight | 15:34 |
| *** rcernin has quit IRC | 15:34 | |
| *** ChanServ sets mode: +v tdasilva | 15:34 | |
| notmyname | ah, there. kicking the mail client in the pants helped. | 15:35 |
| notmyname | hmmm... this looks interesting on cross-project summit planning. but the linked etherpad seems to imply (well, explicitly state, in bold) that only the TC can express any preference on it http://lists.openstack.org/pipermail/openstack-dev/2016-September/103657.html | 15:40 |
| *** pcaruana has quit IRC | 15:40 | |
| notmyname | ehh, maybe not. the TC chooses, but I suppose anyone can comment | 15:43 |
| *** daemontool_ has joined #openstack-swift | 15:45 | |
| notmyname | acoles: I'm going through review emails now. I know several of your patches are being reviewed right now. anything specifically I should pay attention to? | 15:46 |
| *** daemontool has quit IRC | 15:48 | |
| *** sgundur_ has quit IRC | 15:58 | |
| acoles | notmyname: from those on priority review page, https://review.openstack.org/#/c/348495/ has a +2 from donagh and maybe clayg will complete on that one | 16:00 |
| patchbot | patch 348495 - swift - Make container sync copy SLO manifests | 16:00 |
| notmyname | ok, based on your bug report, that's kinda the most important one | 16:00 |
| acoles | notmyname: https://review.openstack.org/#/c/215276/ - kota is up to speed on that and timburke seems to have started :) | 16:01 |
| patchbot | patch 215276 - swift - Enable object server to return non-durable data | 16:01 |
| *** zul has joined #openstack-swift | 16:01 | |
| acoles | notmyname: https://review.openstack.org/#/c/355958/ has mattoliverau thiago and mahatic looking at it | 16:02 |
| patchbot | patch 355958 - swift - EC - eliminate .durable files | 16:02 |
| notmyname | acoles: ok, thanks for the summary | 16:03 |
| acoles | notmyname: I have a hand in this one but there are no reviews on it yet https://review.openstack.org/#/c/210099/ | 16:03 |
| patchbot | patch 210099 - swift - Add process level concurrency to container sync | 16:03 |
| acoles | notmyname: summary is yay for teamwork | 16:03 |
| notmyname | :-) | 16:03 |
| acoles | I'll try to keep up with the reviewers but I am outgunned | 16:03 |
| notmyname | yeah, those top 3 ones are all you ;-) | 16:04 |
| notmyname | what help do you need to respond to reviews? anything the rest of us can do? | 16:04 |
| clayg | acoles: single container async update - MIND BLOWN | 16:04 |
| acoles | notmyname: yeah, how did that happen, and worse one is over a year old so how am I supposed to remember what it does :O | 16:05 |
| clayg | acoles: if I say I still disagree the leak goes the right way (but agree can fix later) - will do that thing where you say "it's ok, we're all wrong sometimes" without clarifying which of us is wrong!? | 16:06 |
| clayg | wait... we care about container sync now? | 16:07 |
| acoles | clayg: lol | 16:07 |
| clayg | what happened to no one uses container sync?! do we use container sync now!? | 16:07 |
| notmyname | clayg: /me *whispers* no | 16:08 |
| notmyname | shhh | 16:08 |
| clayg | phew | 16:09 |
| acoles | clayg: is there a "right" way for a leak to go, we have to argue about which is "least wrong" ;) | 16:09 |
| *** rledisez has quit IRC | 16:10 | |
| acoles | clayg: so i can allow you to be less wrong than me without allowing you to be right, right? ;) | 16:10 |
| *** chsc has quit IRC | 16:11 | |
| acoles | notmyname: I'm thinking on your question, not ignoring | 16:11 |
| *** sgundur_ has joined #openstack-swift | 16:14 | |
| clayg | clayg: when he's not right - he's least wrong | 16:17 |
| clayg | i like that | 16:17 |
| clayg | torgomatic_: when you type SLO is the article 'a' or 'an' - is it "an ess-ell-oh" or "a slow" | 16:22 |
| jrichli | I care about container sync. I need to start digging into it ... on my list | 16:22 |
| clayg | notmyname: jrichli just signed up to review mutli-process container sync | 16:22 |
| jrichli | lol. | 16:22 |
| jrichli | it really is on my list | 16:22 |
| notmyname | :-) | 16:22 |
| jrichli | along with keymasterV2, symlinks, revisiting some SLO and metadata stuff, docs, ... <fades> | 16:23 |
| jrichli | oh yeah, and notifications | 16:23 |
| jrichli | +1 for "an ess-ell-oh" | 16:24 |
| clayg | ahhhh BOOO | 16:25 |
| *** zaitcev has joined #openstack-swift | 16:25 | |
| *** ChanServ sets mode: +v zaitcev | 16:25 | |
| clayg | So when talking a static large object with segment ranges you don't call it a slornage? | 16:25 |
| clayg | *slorange | 16:26 |
| jrichli | clayg: internally, yes :-) but i dont think we have in docs | 16:26 |
| *** cbartz has quit IRC | 16:29 | |
| mahatic | I like and call it "an ess-ell-oh" or "an ess-ell-oh range". "a slow" is lazy :P | 16:32 |
| clayg | rofl | 16:32 |
| jrichli | mahatic: for me, its gotta be ess-ell-oh because of dlo : how do you say dlo without having to say d-lo or d-l-o. | 16:33 |
| jrichli | so ess-lo woudl be fine | 16:33 |
| clayg | keep it on the d-low | 16:33 |
| jrichli | ha! | 16:33 |
| *** rcernin has joined #openstack-swift | 16:34 | |
| mahatic | fwiw when I first heard "slow" from clayg in a conf, it took me a sec to translate it to "s-l-o". of course the context helped ;) | 16:35 |
| *** tesseract- has quit IRC | 16:35 | |
| clayg | gha! UNCLE! ;) | 16:37 |
| mahatic | UNCLE? is that a nickname for a new feature/middleware? ;) | 16:37 |
| clayg | https://en.wikipedia.org/wiki/Say_Uncle | 16:38 |
| MooingLemur | is that the 503-generating middleware? | 16:38 |
| MooingLemur | :) | 16:38 |
| MooingLemur | cluster is saying uncle | 16:38 |
| mahatic | clayg: oo! TIL! thanks | 16:38 |
| clayg | i love y'all so much | 16:39 |
| jrichli | looking forward to Barcelona! MooingLemur: will you be there? | 16:43 |
| MooingLemur | Probably not. Several reasons, including PCI-DSS audit week :/ | 16:47 |
| jrichli | MooingLemur: sorry to hear that | 16:47 |
| *** mvk has quit IRC | 16:47 | |
| *** hseipp has quit IRC | 16:48 | |
| *** diogogmt has joined #openstack-swift | 16:49 | |
| *** tsg has joined #openstack-swift | 16:51 | |
| acoles | jrichli: welcome to core-land, where todo lists have no end :) | 16:54 |
| *** takashi has joined #openstack-swift | 16:54 | |
| acoles | is a bug in SLO a sloworm? | 16:56 |
| jrichli | acoles: :-) I didn't even list the stuff I just want to look at out of my own interest. Its a good feeling though - to have a lot of fun choices | 16:56 |
| asettle | acoles: I came in here simply to lol at that, hahahah nice. | 16:57 |
| acoles | jrichli: did they not tell you, when you become a core you promise not to have any interests of your own, you are a servant to the community | 16:57 |
| notmyname | mmotiani: around today? I may have a question for you in a little bit. need to do some reading first.. | 16:57 |
| ntata | s | 16:57 |
| jrichli | acoles: ic! I dont think that was in the intro to core talk. | 16:58 |
| notmyname | jrichli: bah! I'll have to redo the training materials | 16:58 |
| notmyname | make some more handouts | 16:58 |
| jrichli | that sounds fun | 16:59 |
| acoles | of course it could be neither "ess-ell-oh" nor "slow", there's slo as in sloppy | 16:59 |
| jrichli | that sort of makes me think of pho, although a different sound | 17:00 |
| * jrichli is gonna go get lunch | 17:01 | |
| timburke | dammit! got in not long ago, just catching up with the overnight backlog, and now i'm hungry :P | 17:03 |
| acoles | notmyname: ok, here's an answer...if someone has time, take the optimistic gets patch 215276 and run up proxy and object servers with and without the patch, and vice versa, then run func tests to check there is no upgrade issue | 17:04 |
| patchbot | https://review.openstack.org/#/c/215276/ - swift - Enable object server to return non-durable data | 17:04 |
| clayg | acoles: i can't help but think about how much more terrible this would all have been if container sync was sending x-object-sysmeta-static-large-object: true | 17:04 |
| notmyname | acoles: +1 | 17:05 |
| acoles | clayg: ugh, yeah, swift.gatekeeper_override wouldn't quite cut it whichever way you leaked it | 17:07 |
| notmyname | acoles: thanks. If I get to it, I'll do that, but I wrote that down on the priority reviews page | 17:07 |
| *** klrmn has joined #openstack-swift | 17:07 | |
| acoles | notmyname: ok. I may get to it myself but it seemed a self-contained task. | 17:08 |
| *** timur_ is now known as timur | 17:11 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Assert metadata of SLO PUT from container sync https://review.openstack.org/370316 | 17:11 |
| openstackgerrit | Shashirekha Gundur proposed openstack/swift: Invalidate cached tokens api https://review.openstack.org/370319 | 17:12 |
| openstackgerrit | Clay Gerrard proposed openstack/swift: DRY out probe.common https://review.openstack.org/370321 | 17:14 |
| *** asettle has quit IRC | 17:17 | |
| clayg | you know - i think i have yet to read rfc 7230-7235 in their entirety | 17:17 |
| *** asettle has joined #openstack-swift | 17:18 | |
| acoles | clayg: let's do it together one evening in Barcelona, on the beach with some beers | 17:18 |
| acoles | clayg: *read the rfcs* I mean | 17:18 |
| * acoles blushes | 17:18 | |
| openstackgerrit | Alistair Coles proposed openstack/python-swiftclient: Make tempurl subcommand insist on whole number seconds https://review.openstack.org/367915 | 17:19 |
| *** Suyi_ has joined #openstack-swift | 17:20 | |
| redbo | wow | 17:20 |
| clayg | acoles: k, i'll print 'em out | 17:20 |
| *** daemontool_ has quit IRC | 17:21 | |
| *** asettle has quit IRC | 17:26 | |
| *** oshritf__ has quit IRC | 17:28 | |
| *** sgundur_ has quit IRC | 17:29 | |
| *** raba has joined #openstack-swift | 17:31 | |
| jrichli | that's one for the list of quotes | 17:31 |
| *** tqtran has joined #openstack-swift | 17:34 | |
| *** tqtran has quit IRC | 17:38 | |
| *** itlinux has quit IRC | 17:39 | |
| *** sgundur_ has joined #openstack-swift | 17:39 | |
| clayg | boom - what's next!? | 17:39 |
| *** mingyu has quit IRC | 17:41 | |
| clayg | 500 lines in disk file and 400 lines in the obj controller - i've already reviewed it half a dozen times over the past 13 months - patch 215276 is no joke :\ | 17:46 |
| patchbot | https://review.openstack.org/#/c/215276/ - swift - Enable object server to return non-durable data | 17:46 |
| timburke | clayg: what's next? https://tools.ietf.org/html/rfc7236 of course! we could change our use of `X-Auth-Token: <token>` to be `Authorization: Bearer <token>` :-P | 17:51 |
| *** tqtran has joined #openstack-swift | 17:54 | |
| clayg | timburke: does it never end? | 17:56 |
| clayg | 7240? | 17:56 |
| *** takashi has quit IRC | 18:01 | |
| clayg | wait - which one needs the upgrade testing? | 18:01 |
| notmyname | the one you just mentioned | 18:02 |
| clayg | the loss of .durable files seems like the more dangerous change? can't rollback at least - hopefully the upgrade is smooth | 18:02 |
| acoles | clayg: 215276 | 18:02 |
| clayg | notmyname: you mean like mixed-mode upgrade? where some diskfiles are stupid and the proxy is already smart? | 18:02 |
| acoles | clayg: even I am struggling to review 215276 :/ | 18:02 |
| clayg | ... so is the remove .durables gunna get pushed out? | 18:03 |
| *** mvk has joined #openstack-swift | 18:03 | |
| *** acoles is now known as acoles_ | 18:04 | |
| *** tsg has quit IRC | 18:09 | |
| timburke | clayg: 7686, when last i pulled down RFC-all.tar.gz | 18:10 |
| *** joeljwright has quit IRC | 18:15 | |
| *** chsc has joined #openstack-swift | 18:17 | |
| *** chsc has joined #openstack-swift | 18:17 | |
| *** tsg has joined #openstack-swift | 18:21 | |
| *** sams-gleb has joined #openstack-swift | 18:24 | |
| Guest59079 | notmyname: Hi | 18:30 |
| *** sams-gleb has quit IRC | 18:30 | |
| *** sgundur_ has quit IRC | 18:41 | |
| *** tsg has quit IRC | 18:42 | |
| *** zul_ has quit IRC | 18:55 | |
| *** zul has quit IRC | 18:59 | |
| *** Mype_ has joined #openstack-swift | 18:59 | |
| Mype_ | Hello, i have a problem with the OpenStack PHP SDK (https://github.com/php-opencloud/openstack) | 18:59 |
| Mype_ | Unpossible to connect to the api | 18:59 |
| Mype_ | http://pastebin.com/QbQrBCqX | 19:00 |
| Mype_ | That returns "Endpoint URL could not be found in the catalog for this service. Name: swift Type: object-store Region: GRA1 URL type: public" | 19:00 |
| Mype_ | Thank you in advance! | 19:00 |
| *** sgundur_ has joined #openstack-swift | 19:01 | |
| notmyname | Mype_: I think that means that you don't have the right swift storage URL (or any swift endpoint?) in your keystone catalog. | 19:02 |
| clayg | timburke: how many times did you have to type `swift info --json | jq .swift.policies[].name` before you could get it on the first try? | 19:03 |
| clayg | i'm down to like ~7 tries after like ~3 attempts? i am seeing improvement tho - i'll best you jq | 19:04 |
| *** zul_ has joined #openstack-swift | 19:06 | |
| *** pcaruana has joined #openstack-swift | 19:07 | |
| Mype_ | notmyname: But when I use python client for swift, there's no problem, i can see all my files and interact with them | 19:09 |
| notmyname | Mype_: oh, interesting. might be a auth v3 issue in the php sdk then. | 19:09 |
| notmyname | Mype_: have you found any docs on that? or maybe you can talk to OVH with v2 auth? | 19:10 |
| Mype_ | OVH has both v2 and v3 auth API, but i don't know if the PHP SDK that i'm using is compatible with v2.0 | 19:11 |
| notmyname | I don't either. what do the docs say? | 19:12 |
| timburke | Mype_: i'm guessing the issue is in how the php library is trying to select the endpoint to use -- it's probably worth looking at what the service catalog coming out of OVH looks like, then figuring out how to make the library use some non-default selection criteria | 19:12 |
| *** oshritf__ has joined #openstack-swift | 19:14 | |
| timburke | fwiw, it looks like that library *does* support v2 auth, though i don't yet know how to change between them | 19:14 |
| Mype_ | notmyname: OVH Docs ? Just how to use Python client, they allow us to download a preconfig file | 19:15 |
| Mype_ | timburke: Thanks a lot ! Do you think i have to add a new argument in the auth php code like the endpointUrl ? Or also use the v2 api if available ? | 19:17 |
| timburke | Mype_: looking at https://github.com/php-opencloud/openstack/blob/master/src/Identity/v3/Service.php#L30-L47 i'm guessing you'll need to pass one or more of catalogName, catalogType, or interface. you've already got region | 19:19 |
| *** oshritf__ has quit IRC | 19:19 | |
| *** oshritf__ has joined #openstack-swift | 19:21 | |
| Mype_ | Thanks i'll try to add these arguments. I also watched at the PHP code (beginner) and i found that the error i get was here: https://github.com/php-opencloud/openstack/blob/b5ace79795558e2ad6dcc03ac1f91cb60c3d84a7/src/Identity/v3/Models/Catalog.php#L47 | 19:23 |
| openstackgerrit | Michael Barton proposed openstack/swift: go: container server and replicator https://review.openstack.org/370395 | 19:26 |
| Mype_ | timburke: Thanks i'll try to add these arguments. I also watched at the PHP code (beginner) and i found that the error i get was here: https://github.com/php-opencloud/openstack/blob/b5ace79795558e2ad6dcc03ac1f91cb60c3d84a7/src/Identity/v3/Models/Catalog.php#L47 (repost) | 19:27 |
| timburke | ooh! i knew i'd seen something that could help with this once! https://wiki.openstack.org/wiki/API_Working_Group/Current_Design/Service_Catalog#OVH | 19:31 |
| timburke | only that "Name: swift Type: object-store Region: GRA1 URL type: public" looks right :-/ | 19:31 |
| timburke | at least, for the v2 example catalog that was listed there | 19:31 |
| *** aswadr_ has quit IRC | 19:32 | |
| *** dfg_ has joined #openstack-swift | 19:32 | |
| *** sgundur_ has quit IRC | 19:33 | |
| Mype_ | i set for the v3 : 'authUrl' => 'https://auth.cloud.ovh.net/v3'. With the v2 URL the result was an 404 error. The SDK uses by default the v3 | 19:35 |
| Mype_ | Do you think I should contact OVH for this ? | 19:36 |
| notmyname | just got the note that our tshirts are ready to be picked up :-) | 19:39 |
| *** asettle has joined #openstack-swift | 19:43 | |
| timburke | Mype_: maybe. they've likely helped people use that library before (or a similar one). if you wanted to take a look at the service catalog, though, something like https://gist.github.com/tipabu/c945dc146d45b9bd5f636c528b2cca75 should help (it's using the same, surely-redacted credentials as your original pastebin, so it'll need some alterations) | 19:44 |
| timburke | the response should include a catalog element that has a list of service dicts; i think the type is fairly consistent, so you might want to look for object-store | 19:46 |
| Mype_ | thanks! | 19:51 |
| Mype_ | I found the object-store in the catalog | 19:51 |
| *** hoonetorg has quit IRC | 19:53 | |
| timburke | Mype_: did it have "swift" as the name? and did it have an endpoint that had a region of "GRA1" and interface of "public"? | 19:54 |
| *** can8dnSix has joined #openstack-swift | 19:55 | |
| Mype_ | It has no name | 19:55 |
| Mype_ | just a type http://pastebin.com/ZHVfYgyd | 19:55 |
| Mype_ | swift was not in the output | 19:56 |
| Mype_ | For the GRA1 all seems good | 19:57 |
| timburke | ...and the library doesn't seem to like that https://github.com/php-opencloud/openstack/blob/v2.0.5/src/Identity/v3/Models/Service.php#L77 :-/ | 19:58 |
| *** pcaruana has quit IRC | 20:00 | |
| Mype_ | So the SDK is not compatible with OVH or it's a mistake from the SDK's dev ? Is there any alternative to use Swift in a PHP ? | 20:02 |
| *** zul has joined #openstack-swift | 20:02 | |
| Mype_ | PHP website | 20:02 |
| Mype_ | * | 20:02 |
| MooingLemur | Coming from 1 region and two zones, slowly adding a single-zone 2nd region, there's more shuffling of partitions around between the two old zones than I'd expect on rebalance. Any tricks anyone can imagine that might minimize that? | 20:03 |
| MooingLemur | I suppose I should ensure my swift-ring-builder is the most modern | 20:03 |
| timburke | Mype_: IMHO, it seems like the SDK is making too many assumptions. i'm sure there are other php libraries out there, but i've never looked into them myself | 20:04 |
| timburke | i also just noticed that the catalog is sending back a region_id but no region, so it's probably going to fall down around https://github.com/php-opencloud/openstack/blob/v2.0.5/src/Identity/v3/Models/Endpoint.php#L70 too | 20:05 |
| timburke | Mype_: it seems like talking to OVH is the way to go; they're the people most likely to know what php SDKs will definitely work with them | 20:08 |
| *** hoonetorg has joined #openstack-swift | 20:08 | |
| *** can8dnSix has quit IRC | 20:08 | |
| Mype_ | I will contact them. Do you think it's easy to create my own "librairie" working with the API, I just need to create and delete object, nothing more. With curl ? | 20:11 |
| *** zul has quit IRC | 20:11 | |
| *** zul has joined #openstack-swift | 20:12 | |
| *** tsg has joined #openstack-swift | 20:12 | |
| *** can8dnSix has joined #openstack-swift | 20:13 | |
| timburke | Mype_: i don't think it should be too bad, especially if you use tempurls. i'd wager most of the devs here primarily use curl to poke at swift :-) | 20:20 |
| *** joeljwright has joined #openstack-swift | 20:21 | |
| *** ChanServ sets mode: +v joeljwright | 20:21 | |
| Mype_ | Thanks a lot for your help ! It's very nice of you ! | 20:22 |
| clayg | for i in {1..10}; do nosetests swift/test/unit/common/test_utils.py:TestUnlinkOlder.test_unlink_mixed; if [ $? -ne 0 ]; then break; fi; done <- this breaks for me on master? | 20:25 |
| notmyname | clayg: womm | 20:28 |
| clayg | hmph - git blame says I wrote the damn test? | 20:28 |
| *** sgundur_ has joined #openstack-swift | 20:29 | |
| clayg | ahhhhhhhnd now it's working :'( | 20:30 |
| clayg | i hate computers | 20:30 |
| timburke | i see ~1/100 fail, or ~7/1000 | 20:30 |
| *** niknakpaddywak has joined #openstack-swift | 20:33 | |
| openstackgerrit | Merged openstack/swift: Make container sync copy SLO manifests https://review.openstack.org/348495 | 20:33 |
| clayg | ok, looking at the code I guess subsequent calls to time.time can return the same value :\ | 20:33 |
| clayg | WHO KNEW! | 20:33 |
| clayg | i hate computers | 20:34 |
| timburke | clayg: they're all jerks | 20:34 |
| clayg | when you need them to be fast - too slow; when you need them to be slow - TOO FAST! | 20:34 |
| *** rcernin has quit IRC | 20:36 | |
| *** tsg has quit IRC | 20:37 | |
| *** tongli has quit IRC | 20:38 | |
| clayg | lp bug #1623655 | 20:39 |
| openstack | Launchpad bug 1623655 in OpenStack Object Storage (swift) "TestUnlinkOlder can fail if python is too fast" [Wishlist,Confirmed] https://launchpad.net/bugs/1623655 | 20:39 |
| *** rcernin has joined #openstack-swift | 20:40 | |
| notmyname | hmm...so the fact that it never failed for me is just because my computer is slower than yours? | 20:40 |
| *** Mype_ has quit IRC | 20:40 | |
| clayg | notmyname: timburke's computer is more *optimized* | 20:41 |
| timburke | notmyname: does the swift community need to pitch in to get you a better laptop? | 20:41 |
| *** kei_yama has joined #openstack-swift | 20:41 | |
| *** rcernin has quit IRC | 20:41 | |
| *** rcernin has joined #openstack-swift | 20:41 | |
| *** rcernin has quit IRC | 20:42 | |
| *** rcernin has joined #openstack-swift | 20:42 | |
| openstackgerrit | Merged openstack/python-swiftclient: Make tempurl command check for valid object path https://review.openstack.org/367377 | 20:45 |
| openstackgerrit | Merged openstack/python-swiftclient: Make object option dependent on container in usage strings https://review.openstack.org/367261 | 20:46 |
| openstackgerrit | Merged openstack/python-swiftclient: Fix intermittent test failure https://review.openstack.org/361537 | 20:46 |
| mattoliverau | Morning | 20:50 |
| openstackgerrit | Thai Tran proposed openstack/swift: Supporting Account ACL in keystoneauth https://review.openstack.org/356715 | 20:50 |
| *** m_kazuhiro has joined #openstack-swift | 20:53 | |
| kota_ | good morning | 20:57 |
| mattoliverau | kota_: o/ | 20:58 |
| notmyname | hello | 20:58 |
| notmyname | meeting in 2 minutes in #openstack-meeting | 20:58 |
| kota_ | mattoliverau, notmyname: o/ | 20:58 |
| *** can8dnSix has quit IRC | 20:59 | |
| *** acoles_ is now known as acoles | 21:01 | |
| *** tsg has joined #openstack-swift | 21:02 | |
| openstackgerrit | Thai Tran proposed openstack/swift: Refactored test_tempauth account acl tests https://review.openstack.org/370436 | 21:12 |
| openstackgerrit | Travis McPeak proposed openstack/swift: Updating Bandit config file https://review.openstack.org/365065 | 21:13 |
| openstackgerrit | Thai Tran proposed openstack/swift: Refactored test_tempauth account acl tests https://review.openstack.org/370436 | 21:13 |
| *** dmorita has joined #openstack-swift | 21:17 | |
| *** oshritf__ has quit IRC | 21:18 | |
| *** sgundur_ has quit IRC | 21:23 | |
| *** asettle has quit IRC | 21:25 | |
| *** sgundur_ has joined #openstack-swift | 21:26 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: EC Fragment Duplication - Foundational Global EC Cluster Support https://review.openstack.org/219165 | 21:32 |
| kota_ | update a missing part in the rebasing | 21:34 |
| *** tsg has quit IRC | 21:34 | |
| kota_ | it may still have something, it has been affected by io nice change.... | 21:36 |
| kota_ | which i've not catch up well. | 21:36 |
| *** dfg_ has quit IRC | 21:39 | |
| *** tsg has joined #openstack-swift | 21:42 | |
| clayg | wee that was fun | 21:51 |
| zaitcev | poor inodes | 21:51 |
| clayg | acoles: did you know about the thing with the pessemistic reconstructor not rebuilding fragments? | 21:51 |
| mattoliverau | zaitcev: yeah poor inodes :) | 21:52 |
| * mattoliverau is off to find some breakfast | 21:52 | |
| *** asettle has joined #openstack-swift | 21:53 | |
| acoles | clayg: no? serious? oh, do you mean when will we make reconstructor be optimistic too? | 21:53 |
| tdasilva | clayg: are you talking about this bug: https://bugs.launchpad.net/swift/+bug/1510342 | 21:53 |
| openstack | Launchpad bug 1510342 in OpenStack Object Storage (swift) "Reconstructor does not restore a fragment to a handoff node" [High,Confirmed] - Assigned to Bill Huber (wbhuber) | 21:53 |
| acoles | kota_: timburke clayg I'm at a day conference thing tomorrow so may be a little slow responding on patch 215276 | 21:54 |
| patchbot | https://review.openstack.org/#/c/215276/ - swift - Enable object server to return non-durable data | 21:54 |
| clayg | tdasilva: not sure - my comments on that bug sound like I was thinking about a differet thing | 21:55 |
| zaitcev | I wrote a little stub /info for hummingbird, and suddenly stopped working. Returns 401 now. | 21:55 |
| clayg | acoles: maybe one of us can do a rev so we don't loose a whole rotation of the day star? | 21:55 |
| kota_ | acoles: ok | 21:56 |
| clayg | zaitcev: you forgot to recompile | 21:56 |
| clayg | best part of hummingbird - all current and future bugs can be blamed on "you forgot to recompile" | 21:56 |
| zaitcev | wait, really? | 21:56 |
| clayg | zaitcev: i don't know? | 21:56 |
| zaitcev | I thought there were no object files saved at any point, and everything was rebuild from scratch every time | 21:57 |
| acoles | clayg: sure. I'll get something done tomorrow, I am on site, but likely not all. I'll push whatever I can and handover the baton | 21:57 |
| zaitcev | LOL you totally got me going | 21:57 |
| *** sgundur_ has quit IRC | 21:59 | |
| * kota_ is off for breakfast as well as mattolvierau | 22:00 | |
| *** m_kazuhiro has quit IRC | 22:01 | |
| *** tsg has quit IRC | 22:02 | |
| *** acoles is now known as acoles_ | 22:05 | |
| openstackgerrit | Merged openstack/swift: Make load_crypto_meta more py3-friendly https://review.openstack.org/336721 | 22:06 |
| *** _JZ_ has quit IRC | 22:07 | |
| *** tsg has joined #openstack-swift | 22:07 | |
| *** StraubTW has quit IRC | 22:09 | |
| *** asettle has quit IRC | 22:12 | |
| *** joeljwright has quit IRC | 22:17 | |
| *** patchbot has quit IRC | 22:19 | |
| *** patchbot has joined #openstack-swift | 22:19 | |
| *** patchbot has quit IRC | 22:20 | |
| *** patchbot has joined #openstack-swift | 22:21 | |
| *** rcernin has quit IRC | 22:43 | |
| *** zul has quit IRC | 22:44 | |
| *** neha__ has joined #openstack-swift | 22:44 | |
| *** diogogmt has quit IRC | 22:47 | |
| clayg | it's edgy-edge-cases like this that make it fun to work on swift -> https://gist.github.com/clayg/f6db6b41ef8b5c409297c19de94b3394 | 23:00 |
| clayg | so *most* of the time a 4+2 can recover from two missing frags and a missing durable - but if you put it *just* the right order - no rebuildzy | 23:02 |
| kota_ | clayg: interesting | 23:02 |
| kota_ | will try that probe | 23:03 |
| clayg | object-6020: Unable to get enough responses (3/4) to reconstruct 127.0.0.1:6040/sdb8/584/AUTH_test/container-a0684b76-539c-47b1-a784-2f18bb2a2627/object-2d3aeb2d-cc2f-4505-8f09-5c0895420135 policy#1 frag#5 with ETag 6196a6a3747f858b39574c3631244260 | 23:03 |
| clayg | boooooo | 23:03 |
| kota_ | snipets that we need to optimistic-Reconstrution? | 23:04 |
| clayg | kota_: minimal fix based on Al's work -> https://gist.github.com/clayg/0b7e223fa2b276b273c683f1772541b7 | 23:05 |
| timburke | clayg: wait, (0, 1, 2) works?! | 23:06 |
| kota_ | with optimistic-GET and all fragments should have same timestamp i think. | 23:06 |
| clayg | timburke: yeah i run reconstrutor like three times - so eventually 2 gets his durable back (from 3 i guess?) then it can rebuild | 23:06 |
| clayg | but the guy in the middle never gets his durable - he's trapped by two missing frags that will never rebuild | 23:07 |
| clayg | timburke: I have a little picture -> | 23:07 |
| *** tsg has quit IRC | 23:15 | |
| clayg | https://goo.gl/K7ZKKt | 23:17 |
| clayg | the numbers on the inner circle are just the devices I happen to have in this particular failure | 23:18 |
| clayg | anyway - i guess i'm not tracking if there's a super good reason that we shouldn't have the object server default to action - and how many of problems would just be "fixed" by returning a data file if it's there despite not being durable? idk | 23:21 |
| timburke | clayg: what's a webp file? | 23:21 |
| *** jamielennox is now known as jamielennox|away | 23:22 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: EC Fragment Duplication - Foundational Global EC Cluster Support https://review.openstack.org/219165 | 23:23 |
| clayg | timburke: i don't know? https://en.wikipedia.org/wiki/WebP | 23:23 |
| *** chsc has quit IRC | 23:23 | |
| *** Suyi_ has quit IRC | 23:25 | |
| kota_ | ah, may i got it the failure | 23:27 |
| kota_ | possible solution might be reconstructor can make a commit if the rebulding fragment got success and the local fragment is not durable. | 23:28 |
| kota_ | maybe? | 23:28 |
| kota_ | or other reconstructor should know the non-durable file to push the commit | 23:29 |
| kota_ | ah, optimistic-GET can also help it because since that object-server can report durable frag set so that... | 23:31 |
| kota_ | wrong? (or hard?) anyway, it seems to need more hack on reconstructor/ssync_sender-ish. | 23:32 |
| kota_ | oh, i probably was wrong. | 23:37 |
| kota_ | the durable can be back if a neighbor is durable | 23:39 |
| kota_ | s/the durable/the durable file/ | 23:39 |
| *** takashi has joined #openstack-swift | 23:41 | |
| kota_ | so https://gist.github.com/clayg/0b7e223fa2b276b273c683f1772541b7 can help to make .durable (in fact, also the object) in the neighbors (i.e. node 0, 2) and finally either of those push the .durable file to node 1 | 23:41 |
| kota_ | make sense. | 23:42 |
| openstackgerrit | Merged openstack/swift: Corrections for the API specifications in api-ref https://review.openstack.org/354767 | 23:43 |
| *** nadeem has joined #openstack-swift | 23:43 | |
| clayg | kota_: sounds perfectly reasonable to me | 23:46 |
| kota_ | clayg: ok, thanks! | 23:47 |
| *** takashi has quit IRC | 23:48 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!