| *** charlesw has quit IRC | 00:03 | |
| *** charlesw has joined #openstack-swift | 00:30 | |
| *** charlesw has quit IRC | 00:30 | |
| *** B4rker has joined #openstack-swift | 00:46 | |
| *** B4rker has quit IRC | 00:51 | |
| *** B4rker has joined #openstack-swift | 00:51 | |
| *** B4rker has quit IRC | 00:56 | |
| *** aix has quit IRC | 01:12 | |
| *** kutija_ has joined #openstack-swift | 01:27 | |
| *** kutija has quit IRC | 01:27 | |
| *** ptb has joined #openstack-swift | 02:13 | |
| *** fanyaohong has joined #openstack-swift | 02:59 | |
| *** janonymous_ has joined #openstack-swift | 03:18 | |
| *** haomaiwang has quit IRC | 03:33 | |
| *** dimasot has joined #openstack-swift | 03:48 | |
| *** ptb has quit IRC | 04:08 | |
| *** haomaiwang has joined #openstack-swift | 04:25 | |
| *** chmouel has quit IRC | 04:32 | |
| *** fanyaohong has quit IRC | 05:07 | |
| *** chmouel has joined #openstack-swift | 05:07 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:35 | |
| *** ozialien has joined #openstack-swift | 05:42 | |
| openstackgerrit | John Dickinson proposed openstack/swift: do container listing updates in another (green)thread https://review.openstack.org/189080 | 05:49 |
|---|---|---|
| notmyname | I need some help with that one ^ | 05:49 |
| notmyname | ...leaving comments in gerrit now | 05:49 |
| openstackgerrit | John Dickinson proposed openstack/swift: do container listing updates in another (green)thread https://review.openstack.org/189080 | 05:53 |
| *** proteusguy has joined #openstack-swift | 06:24 | |
| *** proteusguy has quit IRC | 06:38 | |
| *** proteusguy has joined #openstack-swift | 06:38 | |
| *** ozialien has quit IRC | 06:40 | |
| *** proteusguy has quit IRC | 06:59 | |
| openstackgerrit | janonymous proposed openstack/swift: Object are sorted and concatenated by swift in cardinal order. https://review.openstack.org/189015 | 07:23 |
| *** proteusguy has joined #openstack-swift | 07:23 | |
| *** proteusguy has quit IRC | 07:27 | |
| *** proteusguy has joined #openstack-swift | 07:29 | |
| *** proteusguy has quit IRC | 07:42 | |
| *** gfarmerfr has joined #openstack-swift | 07:47 | |
| *** gfarmerfr has quit IRC | 07:49 | |
| janonymous_ | Hi Guys , After setting up my SAIO env i did a account head and found meta data written in other language, is it a valid behaviour . < X-Account-Meta-Temp-Url-Key-2: 鐋᠄嘈㜃鐌ः댍㜆댍嘇㜅嘈嘅鐋툍 < X-Timestamp: 1433531455.36234 < X-Account-Storage-Policy-Silver-Bytes-Used: 0 < X-Account-Meta-Temp-Url-Key: 㜃툎㜃㜅蔈ऀ툎㜄嘈ं᠃嘆ं | 07:49 |
| *** gfarmerfr has joined #openstack-swift | 07:50 | |
| *** janonymous_ has quit IRC | 08:22 | |
| *** [1]dimasot has joined #openstack-swift | 09:09 | |
| *** dimasot has quit IRC | 09:09 | |
| *** [1]dimasot is now known as dimasot | 09:09 | |
| *** janonymous_ has joined #openstack-swift | 09:19 | |
| *** dimasot has quit IRC | 09:23 | |
| *** dimasot has joined #openstack-swift | 09:24 | |
| *** janonymous has quit IRC | 09:40 | |
| *** proteusguy has joined #openstack-swift | 09:59 | |
| *** proteusguy has quit IRC | 10:01 | |
| *** proteusguy has joined #openstack-swift | 10:20 | |
| *** proteusguy has quit IRC | 10:26 | |
| *** janonymous_ has quit IRC | 10:26 | |
| *** proteusguy has joined #openstack-swift | 10:28 | |
| *** proteusguy has quit IRC | 10:33 | |
| *** proteusguy has joined #openstack-swift | 10:46 | |
| *** proteusguy has quit IRC | 11:11 | |
| *** proteusguy has joined #openstack-swift | 11:12 | |
| *** proteusguy has quit IRC | 11:23 | |
| *** proteusguy has joined #openstack-swift | 11:36 | |
| *** dimasot has quit IRC | 11:52 | |
| *** dimasot has joined #openstack-swift | 11:58 | |
| *** proteusguy has quit IRC | 12:01 | |
| *** proteusguy has joined #openstack-swift | 12:01 | |
| *** proteusguy has joined #openstack-swift | 12:03 | |
| *** [1]dimasot has joined #openstack-swift | 12:07 | |
| *** dimasot has quit IRC | 12:08 | |
| *** [1]dimasot is now known as dimasot | 12:08 | |
| *** dimasot has quit IRC | 12:19 | |
| *** dimasot has joined #openstack-swift | 12:19 | |
| *** dimasot has quit IRC | 12:21 | |
| *** dimasot has joined #openstack-swift | 12:22 | |
| *** dimasot has quit IRC | 12:28 | |
| *** dimasot has joined #openstack-swift | 12:40 | |
| *** silor has joined #openstack-swift | 13:08 | |
| *** jodah has quit IRC | 13:16 | |
| *** proteusguy has quit IRC | 13:16 | |
| *** dimasot has quit IRC | 13:24 | |
| *** silor has quit IRC | 14:02 | |
| *** marzif_ has quit IRC | 14:06 | |
| *** jkugel has quit IRC | 14:33 | |
| *** janonymous has joined #openstack-swift | 14:40 | |
| *** marzif_ has joined #openstack-swift | 14:40 | |
| *** haomaiwang has quit IRC | 15:08 | |
| *** haomaiwa_ has joined #openstack-swift | 15:09 | |
| *** thurloat is now known as thurloat_isgone | 15:18 | |
| *** proteusguy has joined #openstack-swift | 15:29 | |
| *** mordred has quit IRC | 15:36 | |
| notmyname | janonymous: yes, that looks like some of the output of test runs (we test non-ascii characters, too) | 15:36 |
| janonymous | did i stopped before cleanup ? | 15:47 |
| *** janonymous has quit IRC | 15:59 | |
| *** proteusguy has quit IRC | 16:07 | |
| *** proteusguy has joined #openstack-swift | 16:08 | |
| *** mmcardle has joined #openstack-swift | 16:09 | |
| *** proteusguy has quit IRC | 16:21 | |
| *** mmcardle has quit IRC | 16:21 | |
| *** mordred has joined #openstack-swift | 16:26 | |
| *** mordred has quit IRC | 16:27 | |
| *** mordred has joined #openstack-swift | 16:27 | |
| *** jkugel has joined #openstack-swift | 16:43 | |
| *** proteusguy has joined #openstack-swift | 17:09 | |
| openstackgerrit | Charles Hsu proposed openstack/python-swiftclient: Add ability to download objects to particular folder. https://review.openstack.org/160283 | 17:18 |
| *** dimasot has joined #openstack-swift | 17:26 | |
| *** mtaylor has joined #openstack-swift | 17:32 | |
| *** mtaylor has left #openstack-swift | 17:37 | |
| *** SkyRocknRoll has quit IRC | 17:37 | |
| *** mordred has quit IRC | 17:42 | |
| *** proteusguy_ has joined #openstack-swift | 17:45 | |
| *** proteusguy_ has quit IRC | 17:46 | |
| *** proteusguy_ has joined #openstack-swift | 17:47 | |
| *** proteusguy has quit IRC | 17:48 | |
| *** proteusguy_ has quit IRC | 17:48 | |
| *** proteusguy has joined #openstack-swift | 17:49 | |
| *** SkyRocknRoll has joined #openstack-swift | 17:51 | |
| *** ozialien has joined #openstack-swift | 17:52 | |
| *** ozialien has quit IRC | 18:15 | |
| *** jrichli has joined #openstack-swift | 18:17 | |
| *** dimasot has quit IRC | 19:19 | |
| *** marzif_ has quit IRC | 19:25 | |
| *** proteusguy has quit IRC | 19:27 | |
| *** proteusguy has joined #openstack-swift | 19:30 | |
| *** esmute has quit IRC | 19:32 | |
| *** proteusguy has quit IRC | 19:36 | |
| *** proteusguy has joined #openstack-swift | 19:52 | |
| *** SkyRocknRoll has quit IRC | 20:33 | |
| *** jistr has joined #openstack-swift | 21:18 | |
| jrichli | notmyname: for an idea of a test that breaks without the changes for patch 189080: | 21:22 |
| patchbot | jrichli: https://review.openstack.org/#/c/189080/ | 21:22 |
| jrichli | What was the motivation to do container listing updates in another (green)thread? What was the problem that the test should reveal, in other words? | 21:22 |
| notmyname | if you remember dima's results that he presented in vancouver, in an object PUT, the container update is actually the slowest part | 21:22 |
| notmyname | and more than that, the result of the container update does not affect the result of the PUT in any way | 21:23 |
| *** flaper87 has quit IRC | 21:23 | |
| notmyname | also, we know that object PUTs get slower as containers get large because the update gets slower | 21:23 |
| notmyname | so the reason to do it in a greenthread is to move it out of the write path that the client sees | 21:23 |
| notmyname | thus speeding up object PUTs (especially when the container device is busy) and removing the performance penalty associated with object writes in large containers | 21:24 |
| notmyname | so, yeah, one question is "what is sufficient testing" since the functional differences are zero. perhaps the updated tests that I did are good enough? (I suspect not). | 21:25 |
| notmyname | but mostly my main problem is the failing async pending write | 21:25 |
| jrichli | yes, that error you spoke of ... this happens in which test exactly? | 21:26 |
| jrichli | and thank you for the explanation. I understand now. | 21:27 |
| notmyname | IIRC I saw it 3 times | 21:27 |
| jrichli | so it is intermittent? | 21:27 |
| *** jistr has quit IRC | 21:27 | |
| notmyname | no | 21:27 |
| notmyname | let me fire up a vm and run tests again | 21:28 |
| notmyname | ah, I remember | 21:28 |
| notmyname | that's one of the issues. I know it happens in tests that were using 1.2.3.4 as an IP address. but I actually don't know which test. there are 2 it could have been | 21:28 |
| notmyname | it's in the object server unit tests | 21:29 |
| notmyname | https://gist.github.com/notmyname/aa4648dc2b427a1c2107 | 21:29 |
| notmyname | jrichli: there ^ | 21:29 |
| jrichli | oh, ok. I was confused because at first glance, it seemed strange that there is an unexpected error, but the tests pass. | 21:31 |
| jrichli | Is there a reason that is ok here? | 21:31 |
| notmyname | do you see the error too? | 21:31 |
| jrichli | i ran it, but haven't gone through all the output. one min | 21:31 |
| notmyname | no, I don't think it's ok, even though tests pass :-) | 21:31 |
| jrichli | ah, ok. I see it there in my output as well. | 21:33 |
| jrichli | so wouldn't another desire be to fix the tests so they fail for this case? | 21:33 |
| notmyname | I /think/ the reason it happens is because http_connect isn't patched during some call to async_update. therefore it's actually trying to talk to 1.2.3.4 instead of running a stub and therefore it's trying to write the async_pending. at which point it gets the OSError | 21:33 |
| jrichli | right. i will poke around and see what I find | 21:35 |
| notmyname | thanks | 21:35 |
| jrichli | first of all, it happens from running check_all_api_methods. a lot is happening there tho, so still tracking it down to which action | 21:44 |
| jrichli | I dont see where the connection is mocked there | 21:47 |
| jrichli | i am looking at the tests that call check_all_api_methods to see if they all mock | 21:49 |
| *** proteusguy has quit IRC | 21:59 | |
| *** aix has joined #openstack-swift | 22:36 | |
| *** km has joined #openstack-swift | 22:54 | |
| jrichli | I think the errors are happening in the teardown, and that is why the tests still pass. It looks like there are a few issues. Will get into more details later. | 23:10 |
| *** jrichli has quit IRC | 23:10 | |
| *** kei_yama has joined #openstack-swift | 23:20 | |
| *** Trixboxer has quit IRC | 23:33 | |
| *** Trixboxer has joined #openstack-swift | 23:35 | |
| *** ho has joined #openstack-swift | 23:48 | |
| ho | Good morning! | 23:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!