| hosanai | pdardeau: hello :-) | 00:01 |
|---|---|---|
| NM | o/ | 00:05 |
| * NM is handling 8m obejects in a single container… | 00:06 | |
| *** chlong has joined #openstack-swift | 00:26 | |
| *** chlong has quit IRC | 00:30 | |
| *** NM has quit IRC | 00:36 | |
| *** zhengyin has quit IRC | 00:38 | |
| *** NM has joined #openstack-swift | 00:40 | |
| *** chlong has joined #openstack-swift | 00:41 | |
| *** tqtran has joined #openstack-swift | 00:46 | |
| *** tqtran has quit IRC | 00:51 | |
| *** mmcardle has joined #openstack-swift | 00:54 | |
| *** m_kazuhiro has joined #openstack-swift | 00:54 | |
| openstackgerrit | Matthew Oliver proposed openstack/swift: Always reset epoch when reseting last_part_moves https://review.openstack.org/346979 | 00:54 |
| mattoliverau | ^ just typo fixed in a commit message. | 01:03 |
| *** chlong has quit IRC | 01:10 | |
| *** chlong has joined #openstack-swift | 01:23 | |
| *** zhengyin has joined #openstack-swift | 01:27 | |
| *** chlong has quit IRC | 01:28 | |
| *** chlong has joined #openstack-swift | 01:30 | |
| *** mmcardle has quit IRC | 01:33 | |
| kota_ | good morning | 01:34 |
| *** NM has quit IRC | 01:44 | |
| mattoliverau | kota_: morning | 01:44 |
| kota_ | mattoliverau: o/ | 01:44 |
| *** trananhkma has joined #openstack-swift | 02:09 | |
| zhengyin | tburke:hello | 02:16 |
| *** vinsh has joined #openstack-swift | 02:18 | |
| *** vinsh has quit IRC | 02:20 | |
| openstackgerrit | zheng yin proposed openstack/python-swiftclient: return exit code https://review.openstack.org/349327 | 02:51 |
| *** jamiec has quit IRC | 03:20 | |
| *** jamiec has joined #openstack-swift | 03:21 | |
| *** amitkqed has quit IRC | 03:38 | |
| *** amitkqed has joined #openstack-swift | 03:38 | |
| *** tqtran has joined #openstack-swift | 03:47 | |
| *** tqtran has quit IRC | 03:52 | |
| *** takashi has joined #openstack-swift | 04:07 | |
| *** chlong has quit IRC | 04:15 | |
| *** klrmn has quit IRC | 04:18 | |
| *** jidar has quit IRC | 04:20 | |
| *** jidar has joined #openstack-swift | 04:24 | |
| *** jidar has left #openstack-swift | 04:30 | |
| *** chlong has joined #openstack-swift | 04:31 | |
| *** psachin has joined #openstack-swift | 04:48 | |
| *** KriskK has joined #openstack-swift | 04:48 | |
| *** psachin has quit IRC | 04:48 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Add Pros/Cons table for global cluster consideration https://review.openstack.org/304382 | 04:55 |
| *** Jeffrey4l has joined #openstack-swift | 04:57 | |
| *** KriskK has quit IRC | 05:00 | |
| *** flwang1 has quit IRC | 05:03 | |
| *** psachin has joined #openstack-swift | 05:13 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:13 | |
| *** ppai has joined #openstack-swift | 05:22 | |
| *** KriskK has joined #openstack-swift | 05:23 | |
| *** KriskK has quit IRC | 05:27 | |
| *** chsc has joined #openstack-swift | 05:33 | |
| *** ppai has quit IRC | 05:37 | |
| *** swat30 has quit IRC | 05:37 | |
| *** thurloat has quit IRC | 05:37 | |
| *** janonymous has joined #openstack-swift | 05:46 | |
| *** tqtran has joined #openstack-swift | 05:49 | |
| *** ppai has joined #openstack-swift | 05:51 | |
| *** tqtran has quit IRC | 05:53 | |
| *** chsc has quit IRC | 05:58 | |
| *** janonymous has quit IRC | 06:04 | |
| *** takashi has quit IRC | 06:04 | |
| *** janonymous has joined #openstack-swift | 06:12 | |
| *** tqtran has joined #openstack-swift | 06:20 | |
| *** KriskK has joined #openstack-swift | 06:22 | |
| *** tqtran has quit IRC | 06:24 | |
| *** manous has joined #openstack-swift | 06:31 | |
| *** geaaru has joined #openstack-swift | 06:33 | |
| *** manous has quit IRC | 06:35 | |
| *** takashi has joined #openstack-swift | 06:38 | |
| *** KriskK has quit IRC | 06:42 | |
| *** oshritf has joined #openstack-swift | 06:43 | |
| *** tesseract- has joined #openstack-swift | 06:44 | |
| *** manous has joined #openstack-swift | 06:48 | |
| *** manous has quit IRC | 06:52 | |
| *** daemontool_ has joined #openstack-swift | 06:53 | |
| *** chlong has quit IRC | 06:59 | |
| *** chlong has joined #openstack-swift | 07:11 | |
| *** m_kazuhiro has quit IRC | 07:11 | |
| *** zhengyin has quit IRC | 07:13 | |
| *** flwang1 has joined #openstack-swift | 07:13 | |
| *** KriskK has joined #openstack-swift | 07:17 | |
| *** rledisez has joined #openstack-swift | 07:20 | |
| *** KriskK has quit IRC | 07:23 | |
| *** chlong has quit IRC | 07:34 | |
| *** zhengyin has joined #openstack-swift | 07:44 | |
| *** admin6_ has joined #openstack-swift | 07:46 | |
| *** ppai has quit IRC | 07:47 | |
| *** ppai has joined #openstack-swift | 08:04 | |
| *** geaaru has quit IRC | 08:07 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/swift: [WIP] Testing latest u-c https://review.openstack.org/318441 | 08:10 |
| *** asettle has joined #openstack-swift | 08:18 | |
| *** geaaru has joined #openstack-swift | 08:19 | |
| *** zhengyin has quit IRC | 08:49 | |
| *** zhengyin has joined #openstack-swift | 08:50 | |
| *** acoles_ is now known as acoles | 09:19 | |
| *** acoles is now known as acoles_ | 09:25 | |
| *** asettle has quit IRC | 09:26 | |
| *** asettle has joined #openstack-swift | 09:27 | |
| *** acoles_ is now known as acoles | 09:36 | |
| *** admin6__ has joined #openstack-swift | 09:45 | |
| *** admin6_ has quit IRC | 09:47 | |
| *** acoles is now known as acoles_ | 09:50 | |
| *** acoles_ is now known as acoles | 09:55 | |
| *** hosanai has quit IRC | 10:15 | |
| *** manous has joined #openstack-swift | 10:21 | |
| *** janonymous has quit IRC | 10:37 | |
| *** trananhkma has quit IRC | 10:44 | |
| *** ppai has quit IRC | 11:32 | |
| *** SkyRocknRoll has quit IRC | 11:34 | |
| *** SkyRocknRoll has joined #openstack-swift | 11:41 | |
| *** ppai has joined #openstack-swift | 11:45 | |
| *** asettle has quit IRC | 11:56 | |
| *** SkyRocknRoll has quit IRC | 12:02 | |
| *** asettle has joined #openstack-swift | 12:02 | |
| *** asettle has quit IRC | 12:06 | |
| *** asettle has joined #openstack-swift | 12:07 | |
| *** SkyRocknRoll has joined #openstack-swift | 12:19 | |
| *** silor has joined #openstack-swift | 12:34 | |
| *** klamath has joined #openstack-swift | 12:41 | |
| *** NM has joined #openstack-swift | 12:44 | |
| *** sileht has quit IRC | 12:44 | |
| *** kei_yama has quit IRC | 12:44 | |
| *** ppai has quit IRC | 12:45 | |
| *** sileht has joined #openstack-swift | 12:46 | |
| *** ppai has joined #openstack-swift | 12:58 | |
| openstackgerrit | Wei Cao proposed openstack/swift: add setuptools Requirement https://review.openstack.org/349492 | 13:02 |
| *** takashi has quit IRC | 13:04 | |
| pdardeau | good morning | 13:05 |
| *** bill_az has joined #openstack-swift | 13:06 | |
| *** SkyRocknRoll has quit IRC | 13:07 | |
| *** _JZ_ has joined #openstack-swift | 13:08 | |
| acoles | pdardeau: good morning | 13:24 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add test that authorize callback cannot violate constraints https://review.openstack.org/349501 | 13:24 |
| pdardeau | acoles: good morning | 13:25 |
| openstackgerrit | oshritf proposed openstack/swift: Add process level concurrency to container sync https://review.openstack.org/210099 | 13:29 |
| *** cdelatte has joined #openstack-swift | 13:29 | |
| *** zul_ has quit IRC | 13:38 | |
| *** ametts has joined #openstack-swift | 13:41 | |
| openstackgerrit | oshritf proposed openstack/swift: Add process level concurrency to container sync https://review.openstack.org/210099 | 13:41 |
| *** zul has joined #openstack-swift | 13:43 | |
| *** thumpba has joined #openstack-swift | 13:50 | |
| tdasilva | good morning | 13:51 |
| *** ametts has quit IRC | 13:51 | |
| *** vinsh has joined #openstack-swift | 13:57 | |
| openstackgerrit | maoshuai proposed openstack/swift: corrected the link https://review.openstack.org/349518 | 13:59 |
| *** ametts has joined #openstack-swift | 14:00 | |
| *** KriskK has joined #openstack-swift | 14:02 | |
| *** nottrobin has quit IRC | 14:04 | |
| *** cppforlife_ has quit IRC | 14:04 | |
| *** amit213 has quit IRC | 14:04 | |
| *** DuncanT has quit IRC | 14:04 | |
| *** zhiyan has quit IRC | 14:04 | |
| *** briancurtin has quit IRC | 14:04 | |
| *** kozhukalov has quit IRC | 14:04 | |
| *** kong has quit IRC | 14:04 | |
| *** serverascode has quit IRC | 14:04 | |
| *** derekjhyang has quit IRC | 14:04 | |
| *** ellen has quit IRC | 14:04 | |
| *** nottrobin has joined #openstack-swift | 14:06 | |
| mmotiani | good morning | 14:07 |
| *** cebreidian has quit IRC | 14:08 | |
| admin6__ | Good morning all, I’d like to know what I can do to speed up the object reconstructor for my EC ring. A full reconstruction pass is currently taking more than 100 hours. and I can’t add new disks as fast as they are filled up… | 14:12 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Add test for copy using a Range header https://review.openstack.org/324607 | 14:12 |
| *** ellen has joined #openstack-swift | 14:14 | |
| admin6__ | I recently set up the object-reconstructor concurrency to 8 (which is the number of devices included in the EC ring per ACO) and it speeds up the reconstruction from about 1 partition per sec to 6-8 partitions per seconds. but now after two weeks a few rebuilds, it drops down to only 2 partitions per seconds. | 14:15 |
| admin6__ | object-reconstructor: 121205/216807 (55.90%) partitions of 2/8 (25.00%) devices reconstructed in 63043.39s (1.92/sec, 107h remaining) | 14:15 |
| *** cebreidian has joined #openstack-swift | 14:16 | |
| *** cppforlife_ has joined #openstack-swift | 14:16 | |
| *** DuncanT has joined #openstack-swift | 14:18 | |
| *** kong has joined #openstack-swift | 14:18 | |
| *** briancurtin has joined #openstack-swift | 14:18 | |
| *** kozhukalov has joined #openstack-swift | 14:18 | |
| *** amit213 has joined #openstack-swift | 14:18 | |
| *** zhiyan has joined #openstack-swift | 14:19 | |
| *** vint_bra has joined #openstack-swift | 14:19 | |
| *** serverascode has joined #openstack-swift | 14:19 | |
| *** derekjhyang has joined #openstack-swift | 14:21 | |
| oshritf | acoles: Thanks again for the container sync review, surfaced a bug which the current set resolves | 14:22 |
| *** chlong has joined #openstack-swift | 14:26 | |
| *** chlong is now known as chlong|gone | 14:27 | |
| *** KriskK has quit IRC | 14:28 | |
| *** chlong|gone has quit IRC | 14:32 | |
| *** chlong|gone has joined #openstack-swift | 14:32 | |
| *** chlong|gone is now known as chlong | 14:34 | |
| *** manous has quit IRC | 14:38 | |
| openstackgerrit | Merged openstack/swift: Check object metadata constraints after authorizing https://review.openstack.org/342295 | 14:38 |
| openstackgerrit | Merged openstack/swift: Fix intermittent SLO unit test failures https://review.openstack.org/335658 | 14:39 |
| *** thumpba has quit IRC | 14:39 | |
| *** diogogmt has quit IRC | 14:40 | |
| *** thumpba has joined #openstack-swift | 14:45 | |
| *** siva_krish has joined #openstack-swift | 14:47 | |
| *** KriskK has joined #openstack-swift | 14:53 | |
| *** manous has joined #openstack-swift | 14:54 | |
| *** vinsh has quit IRC | 14:54 | |
| *** vinsh has joined #openstack-swift | 14:54 | |
| *** vinsh has quit IRC | 14:55 | |
| *** zaitcev has joined #openstack-swift | 14:55 | |
| *** ChanServ sets mode: +v zaitcev | 14:55 | |
| *** vinsh has joined #openstack-swift | 14:55 | |
| *** silor1 has joined #openstack-swift | 14:55 | |
| *** silor has quit IRC | 14:57 | |
| *** silor1 is now known as silor | 14:57 | |
| *** manous has quit IRC | 14:58 | |
| *** jistr is now known as jistr|call | 15:00 | |
| *** mvk has quit IRC | 15:04 | |
| *** StraubTW has joined #openstack-swift | 15:05 | |
| *** nadeem has joined #openstack-swift | 15:07 | |
| *** admin6__ has quit IRC | 15:08 | |
| *** KriskK has quit IRC | 15:10 | |
| *** admin6__ has joined #openstack-swift | 15:10 | |
| *** manous has joined #openstack-swift | 15:11 | |
| *** jistr|call is now known as jistr | 15:12 | |
| *** pgbridge has joined #openstack-swift | 15:13 | |
| *** loki___ has joined #openstack-swift | 15:13 | |
| *** vint_bra has quit IRC | 15:14 | |
| *** manous has quit IRC | 15:16 | |
| *** ppai has quit IRC | 15:22 | |
| *** KriskK has joined #openstack-swift | 15:23 | |
| *** diogogmt has joined #openstack-swift | 15:23 | |
| *** manous has joined #openstack-swift | 15:29 | |
| notmyname | hello, world | 15:32 |
| notmyname | NM: 8 million in a container. how's that working for you? | 15:32 |
| *** psachin has quit IRC | 15:34 | |
| notmyname | clarkb: on https://review.openstack.org/#/c/349492/ should setuptools go in requirements.txt or other-requirements.txt? (assuming that setuptools *should* be in either) | 15:34 |
| patchbot | notmyname: patch 349492 - swift - add setuptools Requirement | 15:34 |
| *** manous has quit IRC | 15:35 | |
| *** nadeem has quit IRC | 15:36 | |
| loki___ | Hi, review pleasehttps://review.openstack.org/#/c/347864/ https://review.openstack.org/#/c/347799/ | 15:39 |
| patchbot | loki___: patch 347864 - swift - Py3:Fixes TypeError:NamedTemporaryFile() got an un... | 15:39 |
| patchbot | loki___: patch 347799 - swift - Py3: Fixes AttributeError: 'generator' object issue | 15:39 |
| notmyname | wait. where did the "edit commit message" link go in gerrit? did that go away in the new design and I'm just noticing it? | 15:39 |
| notmyname | FYI tomorrows's TC meeting (2000UTC) will include a discussion on the 2 alt.lang.* proposals | 15:41 |
| openstackgerrit | John Dickinson proposed openstack/swift: made link in README.rst more clear https://review.openstack.org/349518 | 15:43 |
| notmyname | so to edit (any file) in gerrit now, click the "edit" button that's right above the file list to enter edit mode. Then click the file you want to edit (including the commit message). Then typey typey. Then "Save", then "close", then "publish edits" where the normal review button is | 15:45 |
| *** tsg has joined #openstack-swift | 15:54 | |
| *** tsg has quit IRC | 15:55 | |
| *** oshritf has quit IRC | 15:57 | |
| *** loki___ has quit IRC | 15:59 | |
| *** vinsh has quit IRC | 16:00 | |
| *** jistr is now known as jistr|biab | 16:01 | |
| rledisez | notmyname: hello, i'm very (very) interested by https://wiki.openstack.org/wiki/Swift/ideas/small_files . do you know if somebody started to work on it? | 16:11 |
| rledisez | notmyname: after benching and testing others FS, XFS is still the best, but it's not enought ;) | 16:11 |
| *** qwang has left #openstack-swift | 16:11 | |
| *** tesseract- has quit IRC | 16:14 | |
| *** cdelatte has quit IRC | 16:17 | |
| *** siva_krish has left #openstack-swift | 16:17 | |
| *** KriskK has quit IRC | 16:18 | |
| *** cdelatte has joined #openstack-swift | 16:24 | |
| *** asettle has quit IRC | 16:32 | |
| *** rledisez has quit IRC | 16:33 | |
| *** CaioBrentano has joined #openstack-swift | 16:38 | |
| *** nadeem has joined #openstack-swift | 16:39 | |
| notmyname | bah, he left. I was on a call | 16:43 |
| *** tesseract- has joined #openstack-swift | 16:46 | |
| acoles | notmyname: good morning. FYI patch 348292 which will take advantage of patch 347101 when it lands | 16:46 |
| patchbot | acoles: https://review.openstack.org/#/c/348292/ - openstack-infra/project-config - Add swift job tox-func-in-process-encryption | 16:46 |
| patchbot | acoles: https://review.openstack.org/#/c/347101/ - swift - Enable in-process func tests to optionally use enc... | 16:46 |
| acoles | notmyname: non-voting to start with | 16:46 |
| *** dmorita has joined #openstack-swift | 16:47 | |
| *** daemontool__ has joined #openstack-swift | 16:48 | |
| *** jistr|biab is now known as jistr | 16:49 | |
| *** daemontool_ has quit IRC | 16:52 | |
| *** tesseract- has quit IRC | 16:54 | |
| *** rledisez has joined #openstack-swift | 16:57 | |
| rledisez | notmyname: sorry, had to move. i’m back :) | 16:58 |
| *** tesseract- has joined #openstack-swift | 16:58 | |
| *** tesseract- has quit IRC | 16:58 | |
| *** Suyi_ has joined #openstack-swift | 17:08 | |
| *** oshritf has joined #openstack-swift | 17:11 | |
| notmyname | rledisez: did you see my comment? | 17:11 |
| notmyname | oh, all I said was "he left" | 17:11 |
| rledisez | notmyname: yeah, i saw it | 17:12 |
| notmyname | rledisez: the summary is that nobody is working on small file optimization right now | 17:13 |
| notmyname | rledisez: but it's something I'm excited about | 17:13 |
| notmyname | basically, here's what I'm thinking for the next things in swift | 17:13 |
| notmyname | now: improve repl/EC and obj server with repconn and golang. also get container sharding done | 17:14 |
| notmyname | I hope these will be done by the end of the year | 17:14 |
| *** vint_bra has joined #openstack-swift | 17:14 | |
| notmyname | the golang obj server is basically helping with the scheduling of work on the storage nodes. next, then, is optimizing the on disk layout of data. this includes small-file optimizations | 17:15 |
| notmyname | I hope that will be done in 2017 | 17:15 |
| notmyname | and once we have a good work scheduler and a good on-disk format, then we can tackle proxy<->storage protocols | 17:15 |
| notmyname | there's of course a lot more than that going on, but wrt to the storage nodes, that's what I hope the next 18-24 months looks like in swift | 17:16 |
| *** klrmn has joined #openstack-swift | 17:16 | |
| rledisez | ok, thx for info. good to know | 17:17 |
| *** vinsh has joined #openstack-swift | 17:17 | |
| notmyname | rledisez: so to your specific question, nobody is working on the small-file optimizations yet. we need to get the golang stuff finished first, but after that, i really want that | 17:17 |
| rledisez | notmyname: should i understand that there will be no python version of the small-file optimisations? | 17:18 |
| rledisez | the plan wasn’t that clear in austin, i may have missed some news since | 17:18 |
| notmyname | at our hackathon a few weeks ago, we all agreed (again and rather strongly) that there will not be two different implementations of the object server that are upstream at the same time. IOW we will not have both a python and a golang object server | 17:19 |
| notmyname | so yes, for now I expect that improvements like small-file optimizations will happen in golang instead of python | 17:20 |
| rledisez | ok, that’s clear now :) | 17:21 |
| notmyname | acoles: ok, thanks. looking now | 17:21 |
| timburke | notmyname: https://bugs.launchpad.net/swift/+bug/1608513 is screwy -- the only projects i see with a setuptools>=11.3 req seem to have something to do with ansible? none of swift's recursive deps mention it | 17:27 |
| openstack | Launchpad bug 1608513 in OpenStack Object Storage (swift) "Requirement.parse('setuptools>=11.3" [Undecided,In progress] - Assigned to caowei (caowei-e) | 17:27 |
| notmyname | timburke: interesting. and yes, I was surprised to see that bug too. that is, my thought was "surely we'd have seen this by now" | 17:28 |
| timburke | i'd chalk it up to something like keystonemiddleware or some other middleware, but the bug seems to indicate it was the *object-server* that failed. makes no sense to me | 17:28 |
| timburke | fwiw, though, i've currently got swift services running out of a setuptools==5.8 virtualenv, so...yeah. | 17:29 |
| notmyname | andymccr: ^ any ideas? maybe something from ansible? | 17:29 |
| acoles | jrichli: I left a long comment on bug 1487791. Seems like the only thing that is certain is that we need some tests ! | 17:30 |
| openstack | bug 1487791 in OpenStack Object Storage (swift) "No tests exist for POST to DLO manifest file" [Undecided,In progress] https://launchpad.net/bugs/1487791 - Assigned to Janie Richling (jrichli) | 17:30 |
| *** solomon has joined #openstack-swift | 17:30 | |
| *** acoles is now known as acoles_ | 17:32 | |
| *** geaaru has quit IRC | 17:33 | |
| *** KriskK has joined #openstack-swift | 17:33 | |
| jrichli | acoles: thanks, i will take a look. | 17:42 |
| solomon | am trying to upload object using EC2 keys , however am getting 401 unauthorized errors, following is the error stack | 17:43 |
| solomon | curl -v -i http://fqdn.hostname:8080/swift/v1/rtves/abc -X PUT -H 'Authorization: AWS 7b323d44eb424536b801464397866592:kkPFRzdbQiqxlX+p+Fo97YwFQ7I=' -T abc.txt * About to connect() to fqdn.hostname port 8080 (#0) * Trying xx.xxx.xx.xxx... connected * Connected to fqdn.hostname (xx.xx.xx.xx) port 8080 (#0) > PUT /swift/v1/rtves/abc HTTP/1.1 > User-Agent: curl/7.19.7 (x86_64-redhat-linux-gnu) libcurl/7.19.7 NSS/3.18 Basic | 17:43 |
| solomon | < HTTP/1.1 401 Unauthorized HTTP/1.1 401 Unauthorized < Content-Length: 12 Content-Length: 12 < Accept-Ranges: bytes Accept-Ranges: bytes < Content-type: text/plain; charset=utf-8 Content-type: text/plain; charset=utf-8 | 17:44 |
| timburke | acoles_: i think (but haven't yet confirmed) that the SLO+DLO combination could make some sort of twisted sense. you could have the *manifest* of the SLO be a *DLO* -- that seemed to be what was attempted when i got the TypeError | 17:44 |
| notmyname | no. a DLO defining the SLO manifest is a stupid idea | 17:45 |
| notmyname | (however a *LO referencing other *LOs is ok) | 17:46 |
| notmyname | ;-) | 17:46 |
| notmyname | I am 100% ok with us intentionally breaking any support for weird hybrid chimera objects that are DLO and SLO at the same time | 17:47 |
| timburke | oh yes. decidedly. either max_manifest_size/segments are still in effect (so what was the point?) or you're slurping up a large amount of JSON and risking OOM on the proxy | 17:47 |
| timburke | i *said* it was twisted ;-) | 17:47 |
| notmyname | I'm disagreeing with the "makes sense" part :-) | 17:47 |
| notmyname | and I'd be happy with changing current "undefined" behavior to "totally not possible" | 17:48 |
| *** thumpba has quit IRC | 17:48 | |
| *** thumpba has joined #openstack-swift | 17:49 | |
| timburke | notmyname: SLO referencing DLO is still stupid. and DLO referencing SLO is...not great. i've got this feeling like i explored it and it doesn't work, but i don't remember what broke down where | 17:49 |
| notmyname | *LOs should be able to reference other *LOs. why not? I've got 2 DLOs each for the last 2 years of log files. then I make an SLO (or DLO) referencing those. why is that bad? | 17:51 |
| notmyname | bad from a use case perspective, not bad from a current implementation is confusing perspective | 17:52 |
| notmyname | if the problem is that behavior changes based on the pipeline ordering, then that's what should be fixed :-) | 17:52 |
| *** chsc has joined #openstack-swift | 17:52 | |
| timburke | solomon: looks like you're using swift3+keystone, correct? if you look at the proxy logs, there should be a record of the credential verification going out to keystone; it should say something like 'Connecting to Keystone sending this JSON: {"credentials": {"access": "7b323d44eb424536b801464397866592", "token": "<some base64>", "signature": "kkPFRzdbQiqxlX+p+Fo97YwFQ7I="}}' | 17:55 |
| timburke | solomon: you'll want to decode the base64-encoded stuff and compare against the canonical request you used to generate the signature | 17:56 |
| notmyname | jrichli: I'm looking at patch 347101. about to run it, but it looks good so far. but I have one question | 17:56 |
| patchbot | notmyname: https://review.openstack.org/#/c/347101/ - swift - Enable in-process func tests to optionally use enc... | 17:56 |
| *** KriskK has quit IRC | 17:58 | |
| timburke | notmyname: what happens when the DLO changes? since, y'know, it's *dynamic*. DLOs referencing other DLOs are problematic because DLOs use container listings for verification of sizes, and the sub-DLO will be zero bytes | 17:58 |
| notmyname | jrichli: when you load conf_loader_label, you check if it is None instead of not truthy. So what happens if the envvar is set to an empty string? | 17:58 |
| timburke | maybe DLO-referencing-SLO was ok; i may have been thinking of the DLO-DLO issue... | 17:58 |
| jrichli | notmyname: gotta meeting right now, but will look after that | 18:00 |
| notmyname | jrichli: mostly wondering if (1) it was intentional or (2) if you think it's a big deal. I'm not convinced it's a big deal yet | 18:01 |
| tdasilva | notmyname: hi, I was talking to ppai earlier today and he noticed that we don't have pypi packages for openstack-swift only for python-swiftclient. do you remember why we dont build for swift? | 18:16 |
| notmyname | tdasilva: there aren't any packages for any server/service project in openstack (AFAIK) | 18:17 |
| notmyname | I think that stopped back when -infra got out of the business of making debs and rpms (or maybe it never happened?) | 18:18 |
| *** solomon has quit IRC | 18:19 | |
| *** StraubTW has quit IRC | 18:19 | |
| tdasilva | notmyname: mmm..interesting, thanks | 18:22 |
| notmyname | tdasilva: well I don't know if "interesting" is the word I'd use. but the answer is that I don't really know. I'm just guessing | 18:23 |
| *** permalac has joined #openstack-swift | 18:23 | |
| torgomatic | notmyname: you can't nest large objects at all | 18:37 |
| torgomatic | see request_helpers.SegmentedIterable for details; basically all segment requests come with ?multipart-manifest=get tacked on | 18:38 |
| torgomatic | timburke: ^^ | 18:38 |
| *** StraubTW has joined #openstack-swift | 18:44 | |
| *** oshritf has quit IRC | 18:47 | |
| *** fifieldt has quit IRC | 19:07 | |
| *** admin6 has joined #openstack-swift | 19:12 | |
| *** fifieldt has joined #openstack-swift | 19:18 | |
| *** ametts has quit IRC | 19:30 | |
| *** ametts has joined #openstack-swift | 19:36 | |
| jrichli | notmyname: Honestly, I did a copy/paste/mod of the "if object_post_as_copy is not None:" line. With an empty value, there is an error - whereas there would be no error with suggested change. | 19:47 |
| jrichli | i sort of prefer the error if you specify that var and give a bogus value. the error says there is no mapped function | 19:48 |
| jrichli | (as opposed to ignoring the bad value, and running as if it wasn't there) | 19:49 |
| *** MooingLemur has quit IRC | 19:53 | |
| *** MooingLemur has joined #openstack-swift | 19:54 | |
| *** ametts has quit IRC | 19:55 | |
| notmyname | jrichli: yeah, that makes sense. eg if you give it foobar, you get an error. but if you give it '' you also get an error | 20:00 |
| notmyname | jrichli: ok, +A | 20:00 |
| *** ametts has joined #openstack-swift | 20:00 | |
| jrichli | notmyname: nice. thanks! | 20:07 |
| *** rledisez has left #openstack-swift | 20:08 | |
| *** vint_bra has quit IRC | 20:13 | |
| *** flwang1 has quit IRC | 20:18 | |
| openstackgerrit | Merged openstack/swift: made link in README.rst more clear https://review.openstack.org/349518 | 20:34 |
| *** Jeffrey4l has quit IRC | 20:35 | |
| *** Jeffrey4l has joined #openstack-swift | 20:35 | |
| *** ametts has quit IRC | 20:36 | |
| *** ametts has joined #openstack-swift | 20:38 | |
| *** vint_bra has joined #openstack-swift | 20:50 | |
| notmyname | FYI there's been comments on the TC proposal related to golang (and it's on the agenda for the TC meeting tomorrow) https://review.openstack.org/#/c/346243/ | 20:51 |
| patchbot | notmyname: patch 346243 - governance - Policy and process for alternate programming langu... | 20:51 |
| *** siva_krish has joined #openstack-swift | 20:57 | |
| *** mvk has joined #openstack-swift | 21:04 | |
| openstackgerrit | Merged openstack/swift: Enable in-process func tests to optionally use encryption https://review.openstack.org/347101 | 21:14 |
| *** solomon has joined #openstack-swift | 21:18 | |
| solomon | could somebody confirm swift s3 emulation layer authorization format | 21:19 |
| solomon | curl -v -i http://fqdn.hostname:8080/swift/v1/container/obj -X PUT -H 'Authorization: AWS key:base64sec' -T <filename> | 21:20 |
| *** silor has quit IRC | 21:28 | |
| *** siva_krish has left #openstack-swift | 21:30 | |
| *** solomon has quit IRC | 21:33 | |
| *** NM has quit IRC | 21:33 | |
| openstackgerrit | Merged openstack/swift: go: time out flock differently https://review.openstack.org/334646 | 21:35 |
| openstackgerrit | Merged openstack/swift: go: set nodelay on repconn https://review.openstack.org/341631 | 21:36 |
| *** oshritf has joined #openstack-swift | 21:39 | |
| openstackgerrit | Merged openstack/swift: go: direct client concurrency https://review.openstack.org/341632 | 21:41 |
| *** oshritf has quit IRC | 21:47 | |
| *** jamielennox|away is now known as jamielennox | 21:48 | |
| *** nadeem has quit IRC | 21:52 | |
| *** solomon has joined #openstack-swift | 21:53 | |
| *** diogogmt has quit IRC | 21:55 | |
| notmyname | solomon: hey, sorry I missed you. I saw you left without getting an answer | 21:55 |
| notmyname | solomon: unfortunately, I don't have an answer (without digging into the code). but timburke or kota_ could probably help | 21:56 |
| *** ametts has quit IRC | 21:56 | |
| *** flwang1 has joined #openstack-swift | 22:00 | |
| timburke | solomon: the authorization header looks right, but the url looks a little funky. i'd expect something like http://fqdn.hostname:8080/container/obj (so, no v1 for the swift api, and the account is determined by the key used in authorization) | 22:03 |
| timburke | i'm not entirely sure where the /swift/ came from -- is it something to do with SCRIPT_NAME / PATH_INFO? i've not really played around with those... | 22:03 |
| *** admin6 has quit IRC | 22:06 | |
| *** admin6__ is now known as admin6 | 22:06 | |
| solomon | i tried with out swift/v1, i got http 403 forbidden | 22:12 |
| solomon | this is what i see from swift stat ,[root@ swift stat -v StorageURL: http://host.domain.com:8080/swift/v1 Auth Token: 26535080fafb4a3783dad7a1c0a86250 Account: v1 Containers: 3 Objects: 4 Bytes: 538608 X-Account-Bytes-Used-Actual: 540672 Content-Type: text/plain; charset=utf-8 | 22:13 |
| openstackgerrit | Shashirekha Gundur proposed openstack/swift: update manpage for -P / --policy-name option https://review.openstack.org/349725 | 22:17 |
| openstackgerrit | Merged openstack/swift: go: make container timeout configurable https://review.openstack.org/343320 | 22:18 |
| *** cschwede has quit IRC | 22:26 | |
| mattoliverau | Morning | 22:33 |
| notmyname | hello mattoliverau | 22:35 |
| *** manous has joined #openstack-swift | 22:36 | |
| *** manous has quit IRC | 22:41 | |
| *** solomon has quit IRC | 22:46 | |
| *** vint_bra has quit IRC | 22:50 | |
| *** thumpba has quit IRC | 22:51 | |
| *** thumpba has joined #openstack-swift | 22:51 | |
| *** thumpba has quit IRC | 22:55 | |
| *** fbo has quit IRC | 23:03 | |
| *** asettle has joined #openstack-swift | 23:03 | |
| *** StraubTW has quit IRC | 23:08 | |
| *** asettle has quit IRC | 23:09 | |
| timur | notmyname: jrichli clayg added the first attempt at metadata write up on the wiki (https://wiki.openstack.org/wiki/Swift/ideas/metadata-sync) (and linked from the ideas page). I have been working on factoring out some of the code I wrote internally into a library to be used for the metadata indexing | 23:14 |
| notmyname | timur: cool, thanks | 23:16 |
| *** diogogmt has joined #openstack-swift | 23:17 | |
| *** kei_yama has joined #openstack-swift | 23:17 | |
| *** klamath has quit IRC | 23:18 | |
| *** thumpba has joined #openstack-swift | 23:25 | |
| *** NM has joined #openstack-swift | 23:41 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!