| *** NM has quit IRC | 00:00 | |
| *** lyrrad has quit IRC | 00:06 | |
| *** garthb has quit IRC | 00:07 | |
| *** zaitcev has joined #openstack-swift | 00:13 | |
| *** ChanServ sets mode: +v zaitcev | 00:13 | |
| *** tqtran has quit IRC | 00:15 | |
| openstackgerrit | John Dickinson proposed openstack/swift: Rework the contributor docs https://review.openstack.org/301871 | 00:24 |
|---|---|---|
| *** asettle has quit IRC | 00:32 | |
| *** baojg has joined #openstack-swift | 00:37 | |
| *** Jeffrey4l has joined #openstack-swift | 00:48 | |
| *** mingdang1 has joined #openstack-swift | 00:49 | |
| *** NM has joined #openstack-swift | 00:50 | |
| *** NM has quit IRC | 00:54 | |
| dmorita | Hi, account/container-replicator sends 'local' metadata info to 'remote' in sync process, then update_metadata is done at remote. And, response from remote includes metadata info, but update_metadata function is not executed in local replicator. Is there someone who knows the reason why update_metadata is not executed in local? If there is no big reason of current architecture, I would like to fix DB replicator to | 01:10 |
| dmorita | execute update_metadata in local side. | 01:10 |
| *** tqtran has joined #openstack-swift | 01:11 | |
| *** klrmn has quit IRC | 01:12 | |
| *** tqtran has quit IRC | 01:16 | |
| *** mingdang1 has quit IRC | 01:16 | |
| *** sgundur has left #openstack-swift | 01:16 | |
| *** asettle has joined #openstack-swift | 01:22 | |
| *** asettle has quit IRC | 01:28 | |
| openstackgerrit | Jethro Sun proposed openstack/swift: Fix the errors in swift/common/ring/builder.py and swift/cli/ringbuilder.py https://review.openstack.org/300710 | 01:30 |
| *** NM has joined #openstack-swift | 01:34 | |
| *** NM has quit IRC | 01:39 | |
| *** vinsh has quit IRC | 01:40 | |
| *** vinsh_ has joined #openstack-swift | 01:40 | |
| *** sgundur has joined #openstack-swift | 01:43 | |
| *** sgundur has left #openstack-swift | 01:44 | |
| *** rickyrem has joined #openstack-swift | 01:57 | |
| zaitcev | Collecting .[keystone] | 01:59 |
| zaitcev | Could not find any downloads that satisfy the requirement .[keystone] | 01:59 |
| zaitcev | Tiiiiiiiiiiim | 01:59 |
| zaitcev | No, wait. | 02:00 |
| zaitcev | It's some kind of funky syntax in tox.ini. | 02:01 |
| openstackgerrit | Daisuke Morita proposed openstack/swift: Metadata update in local replicator site https://review.openstack.org/302494 | 02:07 |
| *** wuhg has joined #openstack-swift | 02:12 | |
| *** mwheckmann has joined #openstack-swift | 02:18 | |
| *** dmorita has quit IRC | 02:20 | |
| *** asettle has joined #openstack-swift | 02:29 | |
| stevemar | zaitcev: update your tox or pbr maybe? | 02:35 |
| zaitcev | stevemar: maybe pbr | 02:35 |
| stevemar | zaitcev: if y'all are doing optional dependecies then you need a newer pbr | 02:36 |
| zaitcev | stevemar: what's your version of pbr? | 02:38 |
| zaitcev | I have 1.8.1 | 02:38 |
| stevemar | pbr>=1.6 # Apache-2.0 | 02:38 |
| stevemar | hmm | 02:39 |
| stevemar | weird! | 02:39 |
| clarkb | it is saying install the keystone optional deps for the current package | 02:39 |
| *** gsaps has joined #openstack-swift | 02:39 | |
| stevemar | clarkb: so you're implying that pbr is smart enough to know the funky syntax | 02:41 |
| stevemar | which it should just look for in setup.cfg's entrypoints | 02:41 |
| clarkb | its actually setuptools iirc | 02:41 |
| stevemar | true, it is | 02:41 |
| clarkb | . means package in current dir. and [keystone] is optionals | 02:42 |
| stevemar | i'm assuming zaitcev is talking about this line: https://github.com/openstack/python-swiftclient/blob/master/tox.ini#L13-L15 | 02:42 |
| clarkb | and https://github.com/openstack/python-swiftclient/blob/master/setup.cfg#L35 is where they are defined | 02:43 |
| zaitcev | stevemar: yeah, it comes from the commit that moves the dependency into setup.cfg. | 02:43 |
| clarkb | if I had to guess its setuptools or pip complaining, you may need to upgrade virtualenv (which bundles both) | 02:45 |
| zaitcev | clarkb: thanks, it was virtualenv | 02:49 |
| *** mdnadeem has joined #openstack-swift | 03:00 | |
| *** openstackstatus has quit IRC | 03:01 | |
| *** M4urice has quit IRC | 03:02 | |
| *** M4urice has joined #openstack-swift | 03:05 | |
| *** Raymii has quit IRC | 03:05 | |
| *** Raymii has joined #openstack-swift | 03:06 | |
| *** nadeem has joined #openstack-swift | 03:11 | |
| *** dfg_ has quit IRC | 03:32 | |
| *** dfg has joined #openstack-swift | 03:32 | |
| *** sheel has joined #openstack-swift | 03:41 | |
| *** ChubYann has quit IRC | 03:41 | |
| *** sekrit has quit IRC | 03:51 | |
| *** ChubYann has joined #openstack-swift | 03:57 | |
| *** hosanai has joined #openstack-swift | 03:59 | |
| *** ChanServ sets mode: +v hosanai | 03:59 | |
| *** mwheckmann has quit IRC | 04:00 | |
| hosanai | hello | 04:01 |
| *** sekrit has joined #openstack-swift | 04:05 | |
| *** nadeem has quit IRC | 04:10 | |
| *** rickyrem has quit IRC | 04:10 | |
| *** links has joined #openstack-swift | 04:12 | |
| *** asettle has quit IRC | 04:17 | |
| *** asettle has joined #openstack-swift | 04:18 | |
| *** dmorita has joined #openstack-swift | 04:21 | |
| *** dmorita has quit IRC | 04:25 | |
| *** nakagawamsa has joined #openstack-swift | 04:48 | |
| *** nakagawamsa has quit IRC | 04:48 | |
| *** SkyRocknRoll has joined #openstack-swift | 04:52 | |
| *** gsaps has quit IRC | 04:57 | |
| *** asettle has quit IRC | 05:02 | |
| *** asettle has joined #openstack-swift | 05:03 | |
| *** ankur has quit IRC | 05:21 | |
| *** wer has quit IRC | 05:22 | |
| *** trifon has joined #openstack-swift | 05:25 | |
| *** ppai has joined #openstack-swift | 05:32 | |
| *** ChubYann has quit IRC | 05:44 | |
| *** rcernin has joined #openstack-swift | 05:45 | |
| *** mingdang1 has joined #openstack-swift | 05:58 | |
| *** zaitcev has quit IRC | 06:12 | |
| *** asettle has quit IRC | 06:20 | |
| *** pcaruana has joined #openstack-swift | 06:26 | |
| *** asettle has joined #openstack-swift | 06:33 | |
| openstackgerrit | Nguyen Hung Phuong proposed openstack/swift: Fix typos in Swift files https://review.openstack.org/302533 | 06:33 |
| *** asettle has quit IRC | 06:37 | |
| *** tqtran has joined #openstack-swift | 06:43 | |
| *** tesseract has joined #openstack-swift | 06:45 | |
| *** tesseract is now known as Guest12876 | 06:45 | |
| *** tqtran has quit IRC | 06:47 | |
| *** joeljwright has joined #openstack-swift | 06:58 | |
| *** ChanServ sets mode: +v joeljwright | 06:58 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Imported Translations from Zanata https://review.openstack.org/302542 | 07:00 |
| *** nakagawamsa has joined #openstack-swift | 07:01 | |
| *** mingdang1 has quit IRC | 07:04 | |
| *** joeljwright has quit IRC | 07:14 | |
| *** nakagawamsa has quit IRC | 07:14 | |
| *** dmorita has joined #openstack-swift | 07:16 | |
| *** joeljwright has joined #openstack-swift | 07:20 | |
| *** ChanServ sets mode: +v joeljwright | 07:20 | |
| *** dmorita has quit IRC | 07:21 | |
| *** rledisez has joined #openstack-swift | 07:21 | |
| *** wer has joined #openstack-swift | 07:23 | |
| *** baojg has quit IRC | 07:25 | |
| *** baojg has joined #openstack-swift | 07:25 | |
| *** daemontool has joined #openstack-swift | 07:32 | |
| *** geaaru has joined #openstack-swift | 07:38 | |
| *** jmccarthy has quit IRC | 07:41 | |
| *** jmccarthy has joined #openstack-swift | 07:42 | |
| *** joeljwright has quit IRC | 07:42 | |
| *** asettle has joined #openstack-swift | 07:42 | |
| *** jamielennox is now known as jamielennox|away | 07:45 | |
| *** asettle has quit IRC | 07:45 | |
| *** joeljwright has joined #openstack-swift | 08:02 | |
| *** ChanServ sets mode: +v joeljwright | 08:02 | |
| *** jordanP has joined #openstack-swift | 08:06 | |
| *** mmcardle has joined #openstack-swift | 08:08 | |
| *** joeljwright1 has joined #openstack-swift | 08:11 | |
| *** joeljwright1 has quit IRC | 08:11 | |
| *** joeljwright1 has joined #openstack-swift | 08:12 | |
| *** joeljwright1 has quit IRC | 08:13 | |
| *** acoles_ is now known as acoles | 08:13 | |
| *** joeljwright has quit IRC | 08:14 | |
| *** Jeffrey4l has quit IRC | 08:16 | |
| *** Jeffrey4l has joined #openstack-swift | 08:22 | |
| *** openstackstatus has joined #openstack-swift | 08:30 | |
| *** ChanServ sets mode: +v openstackstatus | 08:30 | |
| *** jistr has joined #openstack-swift | 08:30 | |
| *** mmcardle has quit IRC | 08:33 | |
| -openstackstatus- NOTICE: jobs depending on npm are now working again | 08:35 | |
| *** daemontool has quit IRC | 08:41 | |
| *** mmcardle has joined #openstack-swift | 08:47 | |
| openstackgerrit | Merged openstack/swift: Fix typos in Swift files https://review.openstack.org/302533 | 08:50 |
| *** joeljwright has joined #openstack-swift | 08:50 | |
| *** ChanServ sets mode: +v joeljwright | 08:50 | |
| *** stantonnet has quit IRC | 09:02 | |
| *** stantonnet has joined #openstack-swift | 09:06 | |
| openstackgerrit | Merged openstack/swift: Imported Translations from Zanata https://review.openstack.org/302542 | 09:06 |
| *** mmcardle has quit IRC | 09:09 | |
| *** links has quit IRC | 09:31 | |
| *** mvk has joined #openstack-swift | 09:41 | |
| *** mkrcmari__ has quit IRC | 09:43 | |
| *** links has joined #openstack-swift | 09:44 | |
| *** mvk_ has joined #openstack-swift | 09:49 | |
| *** mvk has quit IRC | 09:53 | |
| *** mingdang1 has joined #openstack-swift | 10:05 | |
| *** mmcardle has joined #openstack-swift | 10:05 | |
| *** baojg has quit IRC | 10:09 | |
| *** mingdang_ has joined #openstack-swift | 10:10 | |
| *** mingdang1 has quit IRC | 10:14 | |
| *** hosanai has quit IRC | 10:18 | |
| *** daemontool has joined #openstack-swift | 10:22 | |
| *** arch-nemesis has joined #openstack-swift | 10:26 | |
| *** arch-nemesis has joined #openstack-swift | 10:26 | |
| *** McMurlock1 has joined #openstack-swift | 10:29 | |
| *** links has quit IRC | 10:39 | |
| *** silor has joined #openstack-swift | 10:45 | |
| *** arch-nemesis has quit IRC | 10:46 | |
| *** Guest12876 is now known as tesseract | 10:47 | |
| *** mvk_ has quit IRC | 10:47 | |
| *** tesseract is now known as Guest74036 | 10:48 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 10:54 |
| *** links has joined #openstack-swift | 10:55 | |
| *** silor has quit IRC | 10:58 | |
| *** mmcardle has quit IRC | 10:58 | |
| *** arch-nemesis has joined #openstack-swift | 11:00 | |
| *** joeljwright has quit IRC | 11:02 | |
| *** McMurlock1 has quit IRC | 11:03 | |
| *** mmcardle has joined #openstack-swift | 11:05 | |
| openstackgerrit | Merged openstack/swift: remove unused current_status method https://review.openstack.org/302437 | 11:11 |
| *** mingyu has joined #openstack-swift | 11:14 | |
| *** arch-nemesis has quit IRC | 11:15 | |
| *** mvk_ has joined #openstack-swift | 11:16 | |
| *** joeljwright has joined #openstack-swift | 11:23 | |
| *** ChanServ sets mode: +v joeljwright | 11:23 | |
| *** mingyu has quit IRC | 11:26 | |
| *** john_bar has quit IRC | 11:29 | |
| *** mingyu has joined #openstack-swift | 11:30 | |
| tdasilva | good morning | 11:49 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Assert that ChunkWriteTimouts are not raised https://review.openstack.org/302294 | 11:53 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Assert that ChunkWriteTimouts are not raised https://review.openstack.org/302294 | 11:55 |
| *** mmcardle has quit IRC | 11:55 | |
| *** joeljwright has quit IRC | 11:55 | |
| tdasilva | timburke: I know it's very late to bring this up, but re patch 299686, didn't this bug show up because the rolling upgrade was done in the order of proxy first, then storage nodes? I thought the recommendation was to always upgrade storage nodes first, then the proxy | 11:55 |
| patchbot | tdasilva: https://review.openstack.org/#/c/299686/ - swift - Fix upgrade bug in versioned_writes (MERGED) | 11:55 |
| *** mmcardle has joined #openstack-swift | 11:56 | |
| *** jhesketh has joined #openstack-swift | 11:57 | |
| *** ChanServ changes topic to "Let's talk! We're nice | Review Dashboard: https://goo.gl/mtEv1C | Hackathon: https://etherpad.openstack.org/p/swift-hackathon-feb-2016 | Logs: http://eavesdrop.openstack.org/irclogs/%23openstack-swift/ | Meetings: https://wiki.openstack.org/wiki/Meetings/Swift | Community Dashboard: http://not.mn/swift/swift_community_dashboard.html" | 11:57 | |
| *** NM has joined #openstack-swift | 12:05 | |
| *** ppai has quit IRC | 12:06 | |
| *** jhesketh has quit IRC | 12:07 | |
| *** mmcardle has quit IRC | 12:07 | |
| *** jhesketh has joined #openstack-swift | 12:07 | |
| *** tamizh_geek has quit IRC | 12:11 | |
| *** tamizh_geek has joined #openstack-swift | 12:12 | |
| *** ppai has joined #openstack-swift | 12:19 | |
| *** mingyu has quit IRC | 12:27 | |
| *** wuhg has quit IRC | 12:29 | |
| *** Guest87820 has quit IRC | 12:29 | |
| *** cdelatte has quit IRC | 12:30 | |
| *** gsaps has joined #openstack-swift | 12:30 | |
| acoles | tdasilva: good morning. continuing the conversation form yesterday's meeting - there are a few places in versioned_writes that deal with receiving a COPY request. Those could/should be removed in patch 156923 | 12:32 |
| patchbot | acoles: https://review.openstack.org/#/c/156923/ - swift - Refactor server side copy as middleware | 12:32 |
| acoles | e.g. https://review.openstack.org/#/c/156923/35/swift/common/middleware/versioned_writes.py | 12:32 |
| patchbot | acoles: patch 156923 - swift - Refactor server side copy as middleware | 12:32 |
| acoles | tdasilva: nothing to right of server_side_copy should see a COPY request, right? | 12:33 |
| acoles | so long as we are confident that versioned_writes is auto-placed to the right of copy | 12:33 |
| *** ppai has quit IRC | 12:34 | |
| acoles | then we can remove all reference to COPY from versioned_writes in 156923, but NOT in 260179, because of the ordering of the patch chain | 12:34 |
| acoles | There are also some tests in test_versioned_writes that test handling of copies - not sure if they get removed or need updating to use an app for the test with server_side_copy plus versioned_writes - we don't want to lose test coverage of copying versioned objects. | 12:36 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Support for http footers - Replication and EC https://review.openstack.org/237393 | 12:42 |
| *** mingyu has joined #openstack-swift | 12:45 | |
| acoles | tdasilva: ^^ thanks for your +2, i fixed couple of things you spotted and made the callback name change, will merge | 12:45 |
| *** cdelatte has joined #openstack-swift | 12:47 | |
| *** silor has joined #openstack-swift | 12:51 | |
| *** delattec has joined #openstack-swift | 12:53 | |
| *** cdelatte has quit IRC | 12:56 | |
| *** diogogmt has quit IRC | 12:58 | |
| *** joeljwright has joined #openstack-swift | 12:59 | |
| *** ChanServ sets mode: +v joeljwright | 12:59 | |
| *** diogogmt has joined #openstack-swift | 12:59 | |
| *** mmcardle has joined #openstack-swift | 13:02 | |
| tdasilva | acoles: very good points about versioned writes. I was thinking the same about where to remove what and reached basically the same conclusion. Had not thought of the test coverage issue but that is very valid point. | 13:03 |
| tdasilva | acoles: http footers patch looks great! | 13:03 |
| *** diogogmt has quit IRC | 13:08 | |
| *** shakamunyi has quit IRC | 13:10 | |
| *** mvk_ has quit IRC | 13:11 | |
| *** shakamunyi has joined #openstack-swift | 13:12 | |
| *** diogogmt has joined #openstack-swift | 13:12 | |
| *** mvk has joined #openstack-swift | 13:13 | |
| *** esker has quit IRC | 13:17 | |
| *** daemontool has quit IRC | 13:17 | |
| *** SkyRocknRoll has quit IRC | 13:20 | |
| *** arch-nemesis has joined #openstack-swift | 13:23 | |
| *** NM has quit IRC | 13:23 | |
| *** NM has joined #openstack-swift | 13:23 | |
| *** esker has joined #openstack-swift | 13:26 | |
| *** daemontool has joined #openstack-swift | 13:28 | |
| *** kei_yama has quit IRC | 13:31 | |
| *** esker has quit IRC | 13:36 | |
| *** esker has joined #openstack-swift | 13:36 | |
| *** pauloewerton has joined #openstack-swift | 13:39 | |
| acoles | tdasilva: sorry been on a call | 13:41 |
| tdasilva | acoles: no worries :) | 13:42 |
| acoles | tdasilva: so do you think we should make those changes to patch 156923? or leave to a follow on. I am concerned it might cause confusion for other reviewers so wonder if we should bite the bullet and do it. | 13:43 |
| patchbot | acoles: https://review.openstack.org/#/c/156923/ - swift - Refactor server side copy as middleware | 13:43 |
| *** mingdang_ has quit IRC | 13:44 | |
| tdasilva | acoles: yeah, I thought we could make it there too. Hopefully that won't be too confusing to people. I don't think the change itself it big. | 13:44 |
| tdasilva | acoles: thoughts? | 13:45 |
| acoles | tdasilva: "make it there" == in 156923? | 13:45 |
| tdasilva | yes | 13:46 |
| tdasilva | sorry | 13:46 |
| acoles | np :) yes lets do it | 13:46 |
| acoles | tdasilva: you or me? i have plenty to do but you probably do too! | 13:46 |
| *** links has quit IRC | 13:46 | |
| tdasilva | acoles: yeah, i have some swiftonfile issues to work on for the first part of my day. I could tackle patch 156923 after I'm done with swiftonfile work | 13:48 |
| patchbot | tdasilva: https://review.openstack.org/#/c/156923/ - swift - Refactor server side copy as middleware | 13:48 |
| acoles | tdasilva: ok, leave a note here if/when you get to it and i'll do the same if i get to it | 13:50 |
| tdasilva | acoles: perfect | 13:50 |
| *** ametts has joined #openstack-swift | 13:56 | |
| *** sgundur has joined #openstack-swift | 13:58 | |
| *** esker has quit IRC | 14:06 | |
| *** diogogmt has quit IRC | 14:09 | |
| *** dmorita has joined #openstack-swift | 14:22 | |
| *** dmorita has quit IRC | 14:26 | |
| *** mwheckmann has joined #openstack-swift | 14:29 | |
| jrichli | yay! http footers merged. thanks acoles and tdasilva! | 14:32 |
| jrichli | btw - on callback namespace, I realized this morning that we are using swift.crypto.fetch_crypto_keys right now. so, I am assuming we just want to replace it with callback namespace? | 14:34 |
| jrichli | i mean, replace the '.cyrpto.' part with '.callback.' | 14:35 |
| *** ametts has quit IRC | 14:48 | |
| *** ametts has joined #openstack-swift | 14:49 | |
| tdasilva | jrichli: sounds good to me | 14:51 |
| jrichli | k | 14:51 |
| *** ametts has quit IRC | 14:57 | |
| *** ametts has joined #openstack-swift | 14:58 | |
| *** _erick0zcr has quit IRC | 15:00 | |
| *** j_king_ is now known as j_king | 15:07 | |
| *** mmcardle has quit IRC | 15:09 | |
| *** diogogmt has joined #openstack-swift | 15:09 | |
| *** Guest74036 has quit IRC | 15:10 | |
| *** geaaru has quit IRC | 15:12 | |
| *** garthb has joined #openstack-swift | 15:16 | |
| *** silor1 has joined #openstack-swift | 15:27 | |
| *** silor1 has quit IRC | 15:30 | |
| *** silor has quit IRC | 15:30 | |
| *** silor has joined #openstack-swift | 15:30 | |
| *** links has joined #openstack-swift | 15:36 | |
| *** silor1 has joined #openstack-swift | 15:36 | |
| *** silor has quit IRC | 15:37 | |
| *** silor1 is now known as silor | 15:37 | |
| acoles | tdasilva: i'm on removing copy from versions in patch 156923 | 15:37 |
| patchbot | acoles: https://review.openstack.org/#/c/156923/ - swift - Refactor server side copy as middleware | 15:37 |
| tdasilva | acoles: ok, sorry i couldn't catch up to it...still reviewing some other code | 15:38 |
| timburke | tdasilva: i'm not actually familiar with the recommended upgrade procedure. i know that for our clusters, we make a single node-by-node pass: shutdown services, upgrade swift, restart services. if there are only PACO nodes, we're guaranteed to have both some proxy-server processes upgraded before some account/container/object-server processes and vice versa | 15:38 |
| acoles | tdasilva: np | 15:38 |
| tdasilva | timburke: ok, I was drawing that recommendation from here: https://swiftstack.com/blog/2013/12/20/upgrade-openstack-swift-no-downtime/ | 15:40 |
| *** garthb has quit IRC | 15:40 | |
| tdasilva | I guess my point was that in the case of that versioned writes bug, the problem only showed up because the proxy was upgraded before the container server, right? | 15:40 |
| timburke | yes. if the container servers are *all* upgraded before any proxies (and any container servers that may be temporarily down will be upgraded before reintegrating with the cluster), the problem should never manifest | 15:45 |
| openstackgerrit | Merged openstack/swift: Support for http footers - Replication and EC https://review.openstack.org/237393 | 15:47 |
| *** links has quit IRC | 15:48 | |
| timburke | and yeah, we don't exactly do that, although the approach is sound. the trouble comes down to (1) we don't require that our users split things up as proxy-only and storage-only nodes and (2) given that, there's no good way to ensure storage processes are upgraded before proxy processes. this is particularly true when you consider unexpected restarts during a rolling upgrade | 15:49 |
| timburke | i suppose i should read more of our own blog posts... | 15:49 |
| *** pcaruana has quit IRC | 15:53 | |
| tdasilva | timburke: you still could if you had different packaging from each service, right? | 15:54 |
| *** shakamunyi has quit IRC | 15:56 | |
| tdasilva | timburke: but it doesn't matter, if the recommendation is not an official deployment recommendation then we need to support both ways | 15:56 |
| timburke | yeah, i'd expect so. but it's enough of a pain to package a version of swift *once* :P | 15:56 |
| timburke | either way, this removes a foot-gun | 15:56 |
| timburke | (and a subtle one at that) | 15:56 |
| timburke | one more reason to get people on board with a history-based view of versioning! | 15:57 |
| tdasilva | timburke: heh! nice | 15:58 |
| * vinsh_ appreciates the operator mindset used in the developer conversation above :) | 15:59 | |
| *** ametts has quit IRC | 16:01 | |
| timburke | every now and then i remember that people actually have to *deploy* this stuff. it's not *just* here for me to do silly things like https://review.openstack.org/#/c/300577/ | 16:02 |
| patchbot | timburke: patch 300577 - swift - Colorize Account/Container listings (ABANDONED) | 16:02 |
| vinsh_ | and deploy at scale and across all sorts of crazy configs | 16:03 |
| *** vinsh_ is now known as Vinsh | 16:04 | |
| notmyname | good morning | 16:17 |
| *** ametts has joined #openstack-swift | 16:17 | |
| *** silor has quit IRC | 16:34 | |
| *** nadeem has joined #openstack-swift | 16:36 | |
| *** dmorita has joined #openstack-swift | 16:36 | |
| *** nadeem has quit IRC | 16:36 | |
| *** david-lyle has quit IRC | 16:36 | |
| *** lyrrad has joined #openstack-swift | 16:36 | |
| *** nadeem has joined #openstack-swift | 16:37 | |
| *** dmorita has quit IRC | 16:40 | |
| *** mingyu has quit IRC | 16:43 | |
| *** mingyu has joined #openstack-swift | 16:43 | |
| *** daemontool has quit IRC | 16:44 | |
| *** nadeem has quit IRC | 16:47 | |
| *** mingyu has quit IRC | 16:48 | |
| *** rcernin has quit IRC | 16:52 | |
| *** tmoreira has quit IRC | 16:53 | |
| *** zaitcev has joined #openstack-swift | 16:55 | |
| *** ChanServ sets mode: +v zaitcev | 16:55 | |
| Vinsh | notmyname: Good morning. | 16:57 |
| *** rledisez has quit IRC | 17:00 | |
| *** jistr has quit IRC | 17:01 | |
| *** gsaps has quit IRC | 17:02 | |
| *** esker has joined #openstack-swift | 17:03 | |
| *** david-lyle has joined #openstack-swift | 17:06 | |
| *** ametts has quit IRC | 17:07 | |
| *** pcaruana has joined #openstack-swift | 17:08 | |
| *** _JZ_ has joined #openstack-swift | 17:09 | |
| *** david-lyle has quit IRC | 17:11 | |
| *** mwheckmann has quit IRC | 17:12 | |
| *** mwheckmann has joined #openstack-swift | 17:17 | |
| *** david-lyle has joined #openstack-swift | 17:17 | |
| *** klrmn has joined #openstack-swift | 17:18 | |
| *** briancline has quit IRC | 17:18 | |
| *** CaioBrentano has joined #openstack-swift | 17:19 | |
| *** nadeem has joined #openstack-swift | 17:20 | |
| *** ametts has joined #openstack-swift | 17:20 | |
| *** briancline has joined #openstack-swift | 17:20 | |
| *** nadeem has quit IRC | 17:21 | |
| *** cschwede has quit IRC | 17:23 | |
| *** tmoreira has joined #openstack-swift | 17:23 | |
| *** david-lyle has quit IRC | 17:23 | |
| *** nadeem has joined #openstack-swift | 17:26 | |
| *** mingyu has joined #openstack-swift | 17:27 | |
| *** rcernin has joined #openstack-swift | 17:33 | |
| *** tqtran has joined #openstack-swift | 17:33 | |
| *** mmcardle has joined #openstack-swift | 17:41 | |
| *** mvk has quit IRC | 17:45 | |
| *** jordanP has quit IRC | 17:46 | |
| *** esker has quit IRC | 17:52 | |
| *** david-lyle has joined #openstack-swift | 17:58 | |
| *** dmorita has joined #openstack-swift | 18:05 | |
| *** mmcardle has quit IRC | 18:06 | |
| *** mmcardle has joined #openstack-swift | 18:11 | |
| *** joeljwright has quit IRC | 18:11 | |
| *** mingyu has quit IRC | 18:12 | |
| *** mmcardle has quit IRC | 18:15 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Refactor server side copy as middleware https://review.openstack.org/156923 | 18:25 |
| acoles | tdasilva: ^^ I'm not sure that the copy+versioning unit tests are needed. I moved them to their own class, lets see what reviewers think. | 18:27 |
| tdasilva | acoles: i'll check it out in bit. I'm sure it's great! :D | 18:28 |
| acoles | jrichli: tdasilva fyi, i pushed this follow up for footer earlier but I don't think the bot announced it in channel patch 302868 | 18:29 |
| patchbot | acoles: https://review.openstack.org/#/c/302868/ - swift (feature/crypto) - Wrap and call existing update_footers callbacks | 18:29 |
| tdasilva | acoles: oh nice, i had not seen that | 18:30 |
| Vinsh | acoles: Love the work on that middleware. I need to extend it to take an argument that includes a list of MANY accounts. | 18:31 |
| jrichli | acoles: i noticed the email. only read the description so far - but will review soon. thanks! | 18:31 |
| *** esker has joined #openstack-swift | 18:32 | |
| acoles | Vinsh: you mean you want a 'copy to many destinations'? | 18:32 |
| Vinsh | Tell the proxy "take this object and copy it to accounts 1 to X" | 18:32 |
| Vinsh | for storing legally required unique copies of the same object in many accounts. | 18:33 |
| openstackgerrit | Janie Richling proposed openstack/swift: Change fetch keys to use callback namespace. https://review.openstack.org/303027 | 18:33 |
| openstackgerrit | Daisuke Morita proposed openstack/swift: Metadata update in local replicator site https://review.openstack.org/302494 | 18:33 |
| Vinsh | acoles: so yeah.. a copy to many. | 18:33 |
| acoles | Vinsh: oic. i was about to ask if you would consider copying an SLO manifest to many accounts pointing to one copy | 18:34 |
| Vinsh | I wish.. but copyright laws make that a non starter :( | 18:34 |
| tdasilva | Vinsh: I wonder if that might be a feature for swift client??? maybe?? | 18:34 |
| acoles | Vinsh: its good for disk manufacturers :) | 18:35 |
| tdasilva | https://review.openstack.org/#/c/280200/ | 18:35 |
| patchbot | tdasilva: patch 280200 - python-swiftclient - Add copy object method | 18:35 |
| Vinsh | Maybe? the idea would be to save traffic into and out of the cluster.. move that duplication to the cluster internal network. | 18:35 |
| Vinsh | I like the direct write to other disk ability of the kinetic drives.. | 18:35 |
| Vinsh | or "third party copy" | 18:35 |
| acoles | Vinsh: so does that also prohibit ever using de-dup technologies? does every owner need to have a 1:1 mapping to their own bytes on disk? | 18:36 |
| Vinsh | you got it. 1-1 mapping. | 18:36 |
| Vinsh | Laws may change someday to allow de-dupe.. but are not there yet for this. | 18:36 |
| acoles | Vinsh: you going to be in Austin for the summit? | 18:37 |
| Vinsh | heck yes! | 18:37 |
| acoles | great | 18:37 |
| clarkb | acoles: I believe filesystem level dedup is fine but not userland dedup | 18:37 |
| Vinsh | looming in the back of the swift meetings breathing heavy. | 18:37 |
| Vinsh | :-0 | 18:37 |
| tdasilva | Vinsh, acoles: I heard that "DVR" laws in the use require media companies to keep a copy of whatever is recorded for every household | 18:38 |
| acoles | clarkb: right, interesting. | 18:38 |
| tdasilva | sounded crazy to me | 18:38 |
| Vinsh | tdasilva: NAILED IT | 18:38 |
| Vinsh | we have to start there with that requirement.. if it changes in the future.. then great. we have more capacity. | 18:38 |
| acoles | tdasilva: Vinsh is a media company :) | 18:39 |
| tdasilva | Vinsh: so in that patch in the client, there seems to be a capability of specifying multiple destinations: https://review.openstack.org/#/c/280200/7/swiftclient/service.py | 18:39 |
| patchbot | tdasilva: patch 280200 - python-swiftclient - Add copy object method | 18:39 |
| tdasilva | acoles: oh | 18:39 |
| *** ChubYann has joined #openstack-swift | 18:39 | |
| acoles | well he and his colleagues collectively are | 18:39 |
| Vinsh | tdasilva: That is a good reference... I think though that it will be get/put calls directly from a recorder to swift | 18:40 |
| Vinsh | so some extension of what acoles has in that middleware sounds close | 18:41 |
| Vinsh | tdasilva: This was the deal that started it all: https://en.wikipedia.org/wiki/Cartoon_Network,_LP_v._CSC_Holdings,_Inc. | 18:42 |
| Vinsh | Companies in the EU are not as restricted I hear | 18:42 |
| *** esker has quit IRC | 18:46 | |
| *** delattec has quit IRC | 18:49 | |
| Vinsh | acoles: are there any immediate "gotchas" you think I would hit with a one to many copy middleware based on yours? | 18:50 |
| acoles | Vinsh: figuring out how to handle some copies failing - do you punt to a repeat of the same request or somehow indicate to client which destinations succeeded? | 18:55 |
| acoles | at least if the request is repeated the data won't get moved where it succeeded before because there would be a 409 conflict due to timestamp being the same | 18:56 |
| acoles | so just a flurry of unnecessary requests | 18:57 |
| Vinsh | Ah-hah.. the repeat seems a better fit | 18:57 |
| acoles | Vinsh: how many destination accounts are you thinking, typically, in a single request? | 18:57 |
| Vinsh | 10,000 | 18:57 |
| acoles | OIC | 18:58 |
| Vinsh | Yeah, on an appropriately scaled cluster. Consider each object is 1MB. Thats a sudden blip of 10G spread out across the network.. which isn't really that bad. | 18:59 |
| Vinsh | Using 40G bonds etc. | 18:59 |
| acoles | you're going to make one proxy pretty busy for a period of time, all that data is going to move through whichever proxy the request hits | 18:59 |
| Vinsh | This is true. Wish there was a way to farm that out to object nodes | 19:00 |
| Vinsh | Maybe we take the 10K and split it across 4 proxies. so each writes 2.5K of that object.. still less data to send into the cluster from the client side. | 19:01 |
| acoles | yeah this is where backend nodes directly copying would be nice, and in your case some kind of gossiping to spread the load off the source node. but we don;t have that yet... | 19:01 |
| Vinsh | acoles: It seems like swift with kinetic drives has something close.. with "third party copy" or direct disk - disk copy? | 19:01 |
| Vinsh | http://www.networkcomputing.com/storage/seagate-boosts-disk-drive-intelligence/399783141 | 19:03 |
| Vinsh | 3rd paragraph from the end. | 19:03 |
| *** cdelatte has joined #openstack-swift | 19:03 | |
| acoles | Vinsh so why not have the client upload once then send many single-copy requests which would get distributed across proxies? And once some of those client requests have completed, start to use the new copies as sources for the next batches of copies so that object server reads get spread too. | 19:03 |
| Vinsh | Heck yes | 19:04 |
| *** esker has joined #openstack-swift | 19:05 | |
| acoles | so you get the effect of "gossiping" but managed by your client | 19:05 |
| *** esker has quit IRC | 19:05 | |
| *** esker has joined #openstack-swift | 19:06 | |
| Vinsh | I suppose it would matter how quickly the entire operation can complete. So each user has access to that object quickly. Maybe the fan out from the proxy does that better.. or maybe the write-read -write copy method balances better. | 19:07 |
| sgundur | hi, created this blog entry on work-around tips for SAIO http://tipsonopenstackswift.blogspot.com/2016/04/openstack-swift-tips-to-get-through.html | 19:07 |
| acoles | Vinsh: I have to leave, sorry, chat again/see you in Austin | 19:08 |
| Vinsh | acoles: Thanks for the brain-share. Be well :) | 19:08 |
| notmyname | sgundur: that is really great (and terrible, IMO, that there are so many gotchas) | 19:09 |
| *** acoles is now known as acoles_ | 19:09 | |
| *** dmorita has quit IRC | 19:09 | |
| sgundur | notmyname: thanks :) | 19:10 |
| *** ankur-gupta-f has joined #openstack-swift | 19:12 | |
| *** ankur-gupta-f has left #openstack-swift | 19:12 | |
| *** dmorita has joined #openstack-swift | 19:12 | |
| *** mingyu has joined #openstack-swift | 19:12 | |
| *** david-lyle_ has joined #openstack-swift | 19:13 | |
| tdasilva | Vinsh: that many single-copy request was kind of what I was pointing to with that copy patch in the client | 19:13 |
| *** david-lyle has quit IRC | 19:15 | |
| *** bkeller` has joined #openstack-swift | 19:16 | |
| *** mingyu has quit IRC | 19:17 | |
| *** david-lyle_ has quit IRC | 19:19 | |
| *** dmorita has quit IRC | 19:22 | |
| *** dmorita has joined #openstack-swift | 19:24 | |
| *** briancline has quit IRC | 19:24 | |
| *** briancline has joined #openstack-swift | 19:25 | |
| *** dmorita has quit IRC | 19:28 | |
| *** david-lyle has joined #openstack-swift | 19:29 | |
| *** bkeller` has quit IRC | 19:30 | |
| *** david-lyle has quit IRC | 19:31 | |
| *** david-lyle has joined #openstack-swift | 19:31 | |
| *** cdelatte has quit IRC | 19:31 | |
| *** Jeffrey4l_ has joined #openstack-swift | 19:32 | |
| *** mvk has joined #openstack-swift | 19:32 | |
| *** Jeffrey4l has quit IRC | 19:34 | |
| *** mvk_ has joined #openstack-swift | 19:35 | |
| *** david-lyle has quit IRC | 19:37 | |
| *** mvk has quit IRC | 19:39 | |
| *** diogogmt has quit IRC | 19:52 | |
| *** diogogmt has joined #openstack-swift | 19:52 | |
| *** McMurlock1 has joined #openstack-swift | 19:57 | |
| *** ametts has quit IRC | 20:01 | |
| *** DCWilliams_VA has joined #openstack-swift | 20:02 | |
| *** mkrcmari__ has joined #openstack-swift | 20:07 | |
| *** mvk_ has quit IRC | 20:10 | |
| *** dmorita has joined #openstack-swift | 20:16 | |
| *** ametts has joined #openstack-swift | 20:17 | |
| openstackgerrit | Denis V. Meltsaykin proposed openstack/swift: Change pids' type to set https://review.openstack.org/303056 | 20:17 |
| *** DCWilliams_VA has quit IRC | 20:22 | |
| *** tongli has joined #openstack-swift | 20:29 | |
| *** ankur-gupta-f has joined #openstack-swift | 20:36 | |
| *** ametts has quit IRC | 20:41 | |
| *** cdelatte has joined #openstack-swift | 20:41 | |
| *** mvk_ has joined #openstack-swift | 20:46 | |
| *** mkrcmari__ has quit IRC | 20:50 | |
| *** mkrcmari__ has joined #openstack-swift | 20:50 | |
| *** mvk_ has quit IRC | 20:54 | |
| *** mvk_ has joined #openstack-swift | 20:55 | |
| *** david-lyle has joined #openstack-swift | 20:56 | |
| *** mkrcmari__ has quit IRC | 20:59 | |
| *** nadeem has quit IRC | 20:59 | |
| *** pcaruana has quit IRC | 21:01 | |
| *** mingyu has joined #openstack-swift | 21:02 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Simplify and DRY out LoggerFileObject recursion guard https://review.openstack.org/303072 | 21:03 |
| *** mingyu has quit IRC | 21:07 | |
| *** tongli has quit IRC | 21:17 | |
| *** shakamunyi has joined #openstack-swift | 21:17 | |
| *** McMurlock1 has quit IRC | 21:19 | |
| *** shakamunyi has quit IRC | 21:24 | |
| *** CaioBrentano has quit IRC | 21:28 | |
| *** NM has quit IRC | 21:37 | |
| *** _JZ_ has quit IRC | 21:44 | |
| *** pauloewerton has quit IRC | 21:54 | |
| *** nadeem has joined #openstack-swift | 22:03 | |
| *** garthb has joined #openstack-swift | 22:05 | |
| openstackgerrit | David Goetz proposed openstack/swift: go: hummingbird rescueparts https://review.openstack.org/303095 | 22:08 |
| openstackgerrit | Merged openstack/swift: Fix infinite recursion during logging when syslog is down https://review.openstack.org/302406 | 22:13 |
| *** briancline has quit IRC | 22:14 | |
| *** mmotiani_ has joined #openstack-swift | 22:15 | |
| *** ntata- has joined #openstack-swift | 22:17 | |
| *** briancline has joined #openstack-swift | 22:17 | |
| *** shakamunyi has joined #openstack-swift | 22:18 | |
| *** barra204 has joined #openstack-swift | 22:20 | |
| *** mwheckmann has quit IRC | 22:20 | |
| *** mingdang1 has joined #openstack-swift | 22:23 | |
| *** shakamunyi has quit IRC | 22:24 | |
| *** barra204 has quit IRC | 22:25 | |
| *** briancline has quit IRC | 22:25 | |
| *** briancline has joined #openstack-swift | 22:27 | |
| *** rcernin has quit IRC | 22:29 | |
| *** nadeem has quit IRC | 22:31 | |
| *** arch-nemesis has quit IRC | 22:33 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Add "history" mode to versioned_writes middleware https://review.openstack.org/214922 | 22:34 |
| *** mmotiani_ has quit IRC | 22:35 | |
| *** ntata- has quit IRC | 22:35 | |
| notmyname | clarkb: you said the summit website schedule thing works for you, right? | 22:37 |
| clarkb | notmyname: I did | 22:37 |
| clayg | notmyname: well it has a lot to do with "get a ubuntu 14.04 vm" is not a consistent starting state in all cases | 22:39 |
| notmyname | clayg: ? | 22:39 |
| clayg | notmyname: oh sorry - that was in reply to "and terrible, IMO, that there are so many gotchas" | 22:39 |
| clayg | which was like 3 hours ago? :P | 22:39 |
| notmyname | oh, right | 22:40 |
| notmyname | clarkb: firfox on linux, right? | 22:40 |
| notmyname | or firefox. whatever ;-) | 22:40 |
| clarkb | yes firefox on ubuntu trusty | 22:40 |
| *** mmotiani_ has joined #openstack-swift | 22:43 | |
| *** ntata- has joined #openstack-swift | 22:43 | |
| notmyname | clarkb: with some js console testing, I'm getting a 404 from the call to https://www.openstack.org/api/v1/summits/6/schedule/8928 (but hitting that via curl gives a 401) | 22:46 |
| *** briancline has quit IRC | 22:46 | |
| *** mingdang1 has quit IRC | 22:47 | |
| notmyname | clarkb: yes, I know you don't actually run or maintain that site. you're just the only person i've talked to so far who's seen it work | 22:47 |
| clarkb | "You don't have access to this item through the API." if I hit that directly | 22:47 |
| notmyname | me too | 22:48 |
| notmyname | with a401 status code | 22:48 |
| clarkb | yup 401 here too | 22:48 |
| notmyname | but addToMySchedule(6,8928); in a console on the page gives a 404 | 22:48 |
| *** daemontool has joined #openstack-swift | 22:48 | |
| notmyname | (which is the onClick action for the + My schedule for the first keynote) | 22:48 |
| clarkb | let me confirm it is still workin for me by adding something to my schedule | 22:49 |
| clarkb | oh hey, it may not work for me anymore | 22:49 |
| clarkb | cool (by cool I mean :( ) | 22:49 |
| clarkb | notmyname: I get a 412 trying to PUT https://www.openstack.org/api/v1/summits/6/schedule/9450 | 22:52 |
| clarkb | I have poked the people that dev the schedule site | 22:52 |
| *** briancline has joined #openstack-swift | 22:52 | |
| notmyname | 401 again for me. did you send other headers or a body? | 22:53 |
| notmyname | and thanks! I hope it's fixed soon | 22:53 |
| clarkb | notmyname: that was clicking on the + button to exercise the application itself | 22:53 |
| clarkb | the 412 that is | 22:53 |
| *** km has joined #openstack-swift | 22:59 | |
| *** briancline has quit IRC | 23:02 | |
| *** briancline has joined #openstack-swift | 23:07 | |
| *** mmotiani_ has quit IRC | 23:22 | |
| *** ntata- has quit IRC | 23:22 | |
| *** briancline has quit IRC | 23:22 | |
| *** briancline has joined #openstack-swift | 23:25 | |
| *** mmotiani_ has joined #openstack-swift | 23:27 | |
| *** tqtran has quit IRC | 23:27 | |
| *** mmotiani_ has quit IRC | 23:27 | |
| notmyname | mattoliverau: around? | 23:28 |
| *** nadeem has joined #openstack-swift | 23:29 | |
| clayg | torgomatic: might have to agree to disagree! Best to let the small stuff slide I think? https://review.openstack.org/#/c/303072/1 :) | 23:29 |
| * clayg hugs everyone | 23:29 | |
| patchbot | clayg: patch 303072 - swift - Simplify and DRY out LoggerFileObject recursion guard | 23:29 |
| *** sivasath has joined #openstack-swift | 23:33 | |
| *** briancline has quit IRC | 23:34 | |
| *** jamielennox|away is now known as jamielennox | 23:36 | |
| *** briancline has joined #openstack-swift | 23:38 | |
| *** kei_yama has joined #openstack-swift | 23:39 | |
| *** Jeffrey4l_ has quit IRC | 23:42 | |
| *** mingyu has joined #openstack-swift | 23:51 | |
| *** mingyu has quit IRC | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!