| *** corvus has quit IRC | 01:48 | |
| *** corvus has joined #openstack-swift | 01:49 | |
| openstackgerrit | Merged openstack/swift master: s3api: Propagate backend PATH_INFO in environ for other middleware https://review.opendev.org/718086 | 02:16 |
|---|---|---|
| viks____ | clayg: Thanks for the inputs... here is my iostat o/p -> http://paste.openstack.org/show/791836/ | 02:33 |
| seongsoocho | Hi, I found that swift can upload object with name that start with '/' . Is it normal operation ? or bug? | 03:01 |
| seongsoocho | for example, "a.com/v1/auth_aaa/con// " will be stored as a object name '/' | 03:02 |
| timburke | seongsoocho, that's expected -- you can even have things like /v1/AUTH_acct/cont/foo////bar or /v1/AUTH_acct/cont/foo/baz/../bar which is distinct from /v1/AUTH_acct/cont/foo/bar | 03:39 |
| timburke | basically, an object name is any valid UTF-8 string with no null bytes (except for the new internal namespace) and no surrogate pairs (so you don't have two different ways to write some codepoints, though you can still run into some trouble with combining characters) | 03:39 |
| timburke | that said, if you don't want such names *in your cluster*, you might look at the name_check middleware -- it should allow you to define a regex of disallowed paths | 03:41 |
| seongsoocho | timburke: I thought it's weird that the file name usually doesn't start with / . thank you !! | 04:21 |
| *** evrardjp has quit IRC | 04:37 | |
| *** evrardjp has joined #openstack-swift | 04:37 | |
| openstackgerrit | Merged openstack/swift stable/train: py3: stop barfing on message/rfc822 Content-Types https://review.opendev.org/718312 | 06:15 |
| *** rpittau|afk is now known as rpittau | 07:18 | |
| *** psachin has joined #openstack-swift | 07:37 | |
| *** threestrands has quit IRC | 08:00 | |
| *** tkajinam has quit IRC | 09:03 | |
| openstackgerrit | Alexandre Lécuyer proposed openstack/swift feature/losf: Use hashes.pkl files for LOSF and remove list_partition_recursive() https://review.opendev.org/718445 | 09:24 |
| *** rpittau is now known as rpittau|bbl | 10:13 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Imported Translations from Zanata https://review.opendev.org/718385 | 10:16 |
| openstackgerrit | Merged openstack/python-swiftclient master: Blacklist stestr 3.0.0 https://review.opendev.org/718309 | 10:56 |
| openstackgerrit | Merged openstack/python-swiftclient master: object versioning features https://review.opendev.org/691877 | 10:56 |
| mattoliverau | clayg: in the meeting today you mentioned you might take a look at that test intermittant failure that rledisez ran into. I ran a script in the background on an SAIO I had lying around today. And am seeing 1 test failure every now and again. | 11:38 |
| mattoliverau | Not sure its the same thing, and isn't as common as I'd expect.. although, maybe the gate resources make it more problematic | 11:38 |
| mattoliverau | dumped here: https://etherpad.openstack.org/p/swift-unit-test-obj-intermittant-2020-04 | 11:38 |
| clayg | Even w/o rledisez change? Or your saying the test failure is hard to repro? | 11:39 |
| mattoliverau | Haven't had time to look and tweak the test yet. Just ran some long running tests to see if there was more then 1 test intermittantly failing. | 11:39 |
| clayg | Yeah that read timeout wasn’t the one he pointed at wa ms | 11:40 |
| mattoliverau | kk | 11:40 |
| clayg | Was it! | 11:40 |
| mattoliverau | probably not | 11:40 |
| mattoliverau | no idea. I was just runnign a script to track down intermittent failures in the area he pointed out. | 11:40 |
| clayg | Well, link the script in the ether pad tho. | 11:41 |
| mattoliverau | kk | 11:41 |
| clayg | Thanks mattoliverau !! | 11:41 |
| mattoliverau | nps. sorry I didn't have a chance to really look closely or even closely at his patch. Just with all the rechecks lately I thought I'd run something in the background while I worked for the man :P | 11:42 |
| mattoliverau | Anyway, forgot I left it running and wanted to post the results. Going back to my evening, happy Easter all! | 11:47 |
| mattoliverau | Will run it on the entire rest of the unit tests to see what else smokes out. Maybe finally give it an upgrade/rewrite to continue running and log errors so it doensn't need to be rerun, so it can be hands off, let's see how this whole locked down easter goes :P | 11:49 |
| *** rpittau|bbl is now known as rpittau | 11:57 | |
| *** psachin has quit IRC | 12:26 | |
| *** rcernin has quit IRC | 13:54 | |
| *** tkajinam has joined #openstack-swift | 14:13 | |
| clayg | so the double und disables the test I guess | 14:14 |
| *** Jeffrey4l has quit IRC | 14:22 | |
| *** Jeffrey4l has joined #openstack-swift | 14:23 | |
| clayg | rledisez: I've been running this on my dev environment like while true; do nosetests swift/test/unit/proxy/controllers/test_obj.py:TestECObjController.test_GET_with_frags_swapped_around; if [ $? -ne 0 ]; then break; fi; done and it's not breaking | 14:41 |
| clayg | Have you figured out where the trace is yet so I can add some sleep to widen it up? | 14:42 |
| rledisez | clayg: you have to run nosetests test/unit/proxy | 14:45 |
| rledisez | it only fails when all run together (the fail is in test/unit/proxy/test_server.py- | 14:45 |
| rledisez | If they are run on their own, all tests are passing | 14:45 |
| rledisez | and to be fair, it may be only on py2. I didn't confirm on py3 | 14:48 |
| *** gyee has joined #openstack-swift | 15:05 | |
| openstackgerrit | Merged openstack/swift master: Update hacking for Python3 https://review.opendev.org/717271 | 15:05 |
| *** tkajinam has quit IRC | 15:49 | |
| *** diablo_rojo has joined #openstack-swift | 15:57 | |
| *** rpittau is now known as rpittau|afk | 15:59 | |
| *** evrardjp has quit IRC | 16:37 | |
| *** evrardjp has joined #openstack-swift | 16:37 | |
| openstackgerrit | Merged openstack/swift feature/losf: Use hashes.pkl files for LOSF and remove list_partition_recursive() https://review.opendev.org/718445 | 18:13 |
| *** spsurya_ has quit IRC | 19:08 | |
| *** diablo_rojo has quit IRC | 19:16 | |
| *** diablo_rojo has joined #openstack-swift | 19:17 | |
| *** NM has joined #openstack-swift | 20:22 | |
| *** NM has quit IRC | 21:21 | |
| *** NM has joined #openstack-swift | 21:46 | |
| *** jv has quit IRC | 21:55 | |
| *** NM has quit IRC | 22:16 | |
| *** tkajinam has joined #openstack-swift | 22:53 | |
| *** diablo_rojo has quit IRC | 23:11 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!