| *** rcernin has joined #openstack-swift | 00:08 | |
| *** jvisser has quit IRC | 00:25 | |
| *** UnfairFunction has quit IRC | 00:30 | |
| *** UnfairFunction has joined #openstack-swift | 00:38 | |
| zaitcev | tdasilva: could you look at https://review.opendev.org/694211 real quick? What do you think is better, "swift symlink [--target=cont2] cont1 obj link" or "swift symlink cont1 obj cont2 link"? | 00:40 |
|---|---|---|
| patchbot | patch 694211 - python-swiftclient - Support uploading Swift symlinks without content. - 8 patch sets | 00:40 |
| *** UnfairFunction has quit IRC | 00:42 | |
| *** rcernin has quit IRC | 00:55 | |
| *** rcernin has joined #openstack-swift | 00:55 | |
| timburke | zaitcev, clayg was reminding me that we have some prior art from `swift copy` for something like `swift symlink --target=/cont2[/link] cont1 [obj] [...]` | 01:36 |
| timburke | sorrison, got some more of the stack trace on the attribute error? that seems *very* suspicious... | 01:36 |
| timburke | i'm also curious about what happened proxy-side, but that may have just been dumping out the 500 responses from the object servers | 01:37 |
| sorrison | timburke: http://paste.openstack.org/show/789446/ | 01:38 |
| sorrison | this is the log we see on the proxy | 01:38 |
| sorrison | sorry for the no multiline mess | 01:39 |
| timburke | no worries; i'm used to it ;-) | 01:40 |
| timburke | (fwiw, i often add a `| sed -e 's/#012/\n/g'` on the ends of my log greps, which helps) | 01:40 |
| sorrison | So far it looks like only 1 container is affected, it has 28404 objects in it | 01:42 |
| timburke | which version of py3 is this? | 01:42 |
| sorrison | 3.6.7-1~18.04 | 01:42 |
| timburke | ...and eventlet... still getting the inheritance getting loaded in my head again | 01:43 |
| sorrison | 0.25.0 | 01:43 |
| sorrison | We are using the packages in the ubuntu train cloud archive | 01:43 |
| timburke | 👍 | 01:44 |
| *** UnfairFunction has joined #openstack-swift | 01:45 | |
| sorrison | that stack is when the user does a GET on the object | 01:46 |
| sorrison | get a different error if they try and PUT http://paste.openstack.org/show/789447/ | 01:47 |
| *** UnfairFunction has quit IRC | 01:50 | |
| timburke | comments like https://github.com/python/cpython/blob/v3.6.7/Lib/email/message.py#L233-L246 always tend to scare me... why do i need to get this must context loaded into my head to understand what your return type is going to be!? | 01:53 |
| timburke | on the PUT, it's probably worth looking for tracebacks in the object-server logs, too | 01:54 |
| timburke | ...were they requesting multiple ranges, maybe? | 01:56 |
| sorrison | For the PUT the traceback on the object server is the `AttributeError: 'list' object has no attribute 'rstrip'` error | 01:57 |
| sorrison | I can see they are using python-swiftclient-3.8.1 | 01:58 |
| timburke | interesting... do you know if there's any non-ascii metadata on the objects? either the ones already stored or the new ones your users are trying to write? | 01:58 |
| timburke | that's what that `if not six.PY2:` method definition is all about; working around https://bugs.python.org/issue37093 | 01:59 |
| kota_ | good morning | 02:00 |
| sorrison | OK I can see the object, it's a raw email, I can do a curl on it from the proxy ok, just looking for dodgy characters | 02:04 |
| timburke | what's the content-type? | 02:04 |
| sorrison | Content-Type: message/rfc822 | 02:05 |
| timburke | :-( | 02:07 |
| timburke | got a repro | 02:07 |
| timburke | it's all about the content-type | 02:07 |
| sorrison | bloody rfc822 | 02:08 |
| timburke | i blame https://github.com/eventlet/eventlet/blob/v0.25.0/eventlet/green/http/client.py#L218 (though in eventlet's defense, they just cribbed it from https://github.com/python/cpython/blob/v3.6.7/Lib/http/client.py#L164) | 02:09 |
| sorrison | this has taken me places I did not expect | 02:11 |
| timburke | meanwhile, i know i've been here before... | 02:16 |
| timburke | ok, tomorrow, i'm going to investigate https://github.com/python/cpython/blob/v3.6.7/Lib/email/message.py#L110-L112 and where that logic's implemented | 02:17 |
| timburke | or maybe just go looking at what the payloads are like when it has that c-t and try to make it work... | 02:17 |
| timburke | but i've gotta sign off | 02:17 |
| sorrison | ok thanks for you help, looks like only one user affected here so we can handle that fine | 02:18 |
| *** jvisser has joined #openstack-swift | 02:48 | |
| *** rcernin has quit IRC | 03:06 | |
| *** psachin has joined #openstack-swift | 03:22 | |
| *** rcernin has joined #openstack-swift | 03:23 | |
| *** zaitcev has quit IRC | 03:39 | |
| *** gyee has quit IRC | 04:08 | |
| *** rcernin has quit IRC | 04:16 | |
| *** rcernin has joined #openstack-swift | 04:16 | |
| *** psachin has quit IRC | 04:33 | |
| *** UnfairFunction has joined #openstack-swift | 04:35 | |
| *** jvisser has quit IRC | 04:37 | |
| *** UnfairFunction has quit IRC | 04:39 | |
| *** UnfairFunction has joined #openstack-swift | 04:45 | |
| *** UnfairFunction has quit IRC | 04:47 | |
| *** rcernin is now known as rcernin|lunch | 05:01 | |
| *** evrardjp has quit IRC | 05:34 | |
| *** evrardjp has joined #openstack-swift | 05:34 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: py3: stop barfing on message/rfc822 Content-Types https://review.opendev.org/707312 | 05:37 |
| timburke | sorrison, i think ^^^ should square you. needs tests and a bug tho -- definitely should backport to train | 05:38 |
| *** rpittau|afk is now known as rpittau | 07:21 | |
| *** ianychoi has joined #openstack-swift | 07:46 | |
| *** rdejoux has joined #openstack-swift | 08:08 | |
| *** tkajinam has quit IRC | 08:10 | |
| *** ccamacho has joined #openstack-swift | 08:27 | |
| *** mikecmpbll has joined #openstack-swift | 09:05 | |
| *** xiaolin has quit IRC | 09:34 | |
| *** xiaolin has joined #openstack-swift | 09:43 | |
| *** xiaolin has quit IRC | 10:03 | |
| *** rcernin|lunch has quit IRC | 10:22 | |
| *** rcernin|lunch has joined #openstack-swift | 10:36 | |
| *** rpittau is now known as rpittau|bbl | 11:33 | |
| *** jvisser has joined #openstack-swift | 12:18 | |
| *** ianychoi has quit IRC | 12:25 | |
| *** rpittau|bbl is now known as rpittau | 13:05 | |
| *** jvisser has quit IRC | 13:10 | |
| *** manuvakery has quit IRC | 13:13 | |
| *** rcernin|lunch has quit IRC | 13:50 | |
| *** jvisser has joined #openstack-swift | 14:06 | |
| *** tesseract has quit IRC | 14:20 | |
| openstackgerrit | Charles Hsu proposed openstack/python-swiftclient master: WIP: Preload and dump configurable variables https://review.opendev.org/707409 | 15:28 |
| *** rpittau is now known as rpittau|afk | 15:31 | |
| *** tesseract has joined #openstack-swift | 16:02 | |
| *** gyee has joined #openstack-swift | 16:52 | |
| *** rdejoux has quit IRC | 17:10 | |
| openstackgerrit | Charles Hsu proposed openstack/python-swiftclient master: WIP: Preload and dump configurable variables https://review.opendev.org/707409 | 17:27 |
| *** tesseract has quit IRC | 17:32 | |
| *** evrardjp has quit IRC | 17:34 | |
| *** evrardjp has joined #openstack-swift | 17:34 | |
| *** mikecmpbll has quit IRC | 17:39 | |
| *** rdejoux has joined #openstack-swift | 18:31 | |
| *** rdejoux has quit IRC | 19:00 | |
| *** manuvakery has joined #openstack-swift | 19:05 | |
| *** zaitcev has joined #openstack-swift | 19:11 | |
| *** ChanServ sets mode: +v zaitcev | 19:11 | |
| *** rdejoux has joined #openstack-swift | 20:48 | |
| *** ChanServ has quit IRC | 20:50 | |
| timburke | reminder: no meeting this week | 20:53 |
| *** jvisser has quit IRC | 20:53 | |
| *** rdejoux has quit IRC | 20:54 | |
| tdasilva | timburke: thanks! | 20:54 |
| kota_ | wow | 20:58 |
| kota_ | does something happen for no meeting? | 20:58 |
| kota_ | ok. I found timburke's proposal to skip the meeting in the log. make sense. | 21:06 |
| kota_ | let's go back to my bed... | 21:07 |
| *** ChanServ has joined #openstack-swift | 21:13 | |
| *** orwell.freenode.net sets mode: +o ChanServ | 21:13 | |
| *** rcernin has joined #openstack-swift | 21:18 | |
| *** jvisser has joined #openstack-swift | 21:29 | |
| timburke | sorry kota_, i should have called it out again yesterday once you were online | 21:46 |
| *** manuvakery has quit IRC | 21:55 | |
| sorrison | timburke: Thanks! I'll take a look | 21:55 |
| *** jvisser has quit IRC | 22:19 | |
| mattoliverau | morning | 22:25 |
| mattoliverau | thanks for the slower morning :) | 22:26 |
| *** tkajinam has joined #openstack-swift | 22:54 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!