| *** saschpe has quit IRC | 00:00 | |
| anticw | notmyname: i should read scroll-back but i'm lazy ... | 00:06 |
|---|---|---|
| anticw | is 1.14 with feature/ec | 00:06 |
| anticw | likely to drop soon? | 00:06 |
| openstackgerrit | paul luse proposed a change to openstack/swift: Move Storage Policy validation from parser to class https://review.openstack.org/83252 | 00:08 |
| *** saschpe has joined #openstack-swift | 00:09 | |
| peluse | anticw: about to take off but the specific ver for mergeing policies (on the feature/ec branch now but doesn't include ec) isn't determined yet other than not Icehouse. Same for EC once we start posting those patches | 00:13 |
| *** matsuhashi has joined #openstack-swift | 00:19 | |
| *** Trozz has joined #openstack-swift | 00:19 | |
| notmyname | anticw: (sorry, had some very important car racing to be done with a 6 year-old) | 00:19 |
| notmyname | anticw: ya, looks like peluse captured it. let me find a link to the summary | 00:20 |
| *** Trozz has left #openstack-swift | 00:20 | |
| notmyname | anticw: http://lists.openstack.org/pipermail/openstack-dev/2014-March/030937.html | 00:20 |
| h6w | torgomatic: Thanks. That's very helpful. And we're going to change our design slightly to cope with it as a result. :-) | 00:34 |
| *** d89 has joined #openstack-swift | 00:35 | |
| *** shri has quit IRC | 00:40 | |
| *** csd has joined #openstack-swift | 00:40 | |
| *** sunh has joined #openstack-swift | 00:47 | |
| sunh | I am newbie to SWIFT .. I was curious .. what will happen in we have kept replication as 2 and both container servers get down.. how information is updated in the containerdb.. | 00:49 |
| *** acoles_ has quit IRC | 00:59 | |
| sunh | Also in case proxy server gets down after sending request to Object server.. Probably user will not be able to know if the the information is populated in SWIFT or not? Is there any mechanism | 01:03 |
| *** csd has quit IRC | 01:07 | |
| *** occupant has quit IRC | 01:12 | |
| *** d89 has quit IRC | 01:12 | |
| *** matsuhashi has quit IRC | 01:12 | |
| *** creiht has quit IRC | 01:12 | |
| *** zigo has quit IRC | 01:12 | |
| *** StevenK has quit IRC | 01:12 | |
| *** zul has quit IRC | 01:12 | |
| *** h6w has quit IRC | 01:12 | |
| *** CrackerJackMack has quit IRC | 01:12 | |
| *** wer has quit IRC | 01:12 | |
| *** sudorandom has quit IRC | 01:12 | |
| clarkb | hello swifters this test https://git.openstack.org/cgit/openstack/swift/tree/test/unit/common/test_utils.py#n1091 just failed in the gate causing a reset. I am looking into that test and notice incr_by should always be >=0 according to the docs for that util function but that test passes in -1 twice | 01:22 |
| clarkb | any idea of why that is? | 01:22 |
| clarkb | in the case it is less than zero the util function appears to simply return the running time as is | 01:23 |
| _sEBAs_ | hey there! somebody already had this problem? "Unable to get bucket brigade for request" | 01:23 |
| _sEBAs_ | in result of that "UnicodeDecodeError" in the havana dashboard | 01:25 |
| _sEBAs_ | some clue? | 01:25 |
| clarkb | I am pretty sure that test is failing because ratelimit_sleep may sleep longer than the supplied value | 01:26 |
| _sEBAs_ | find "IOError: failed to write data" in the logs, thats just really bad | 01:29 |
| portante | clarkb: looking | 01:32 |
| clarkb | portante: the previous test says it is asserting it is accurate to a 10th of a second, but I don't think that we can rely on that as sleeps may go longer | 01:33 |
| portante | hmm | 01:33 |
| *** erlon has joined #openstack-swift | 01:33 | |
| portante | so ratelimit_sleep() should return the first argument, running_time, if incr_by is <= 0 | 01:34 |
| clarkb | portante: ya so I don't think incr_by being less than 0 is related | 01:35 |
| clarkb | that was just the first thing I noticed as funny but it is essentially a nop | 01:35 |
| portante | but what is this thing doing ... | 01:35 |
| clarkb | portante: I think it is sleeping 500 milliseconds then double checking that the actual sleep time was within a 10th of a second of that number | 01:37 |
| * clarkb reads more | 01:37 | |
| * portante too | 01:37 | |
| portante | subtle is this function's middle name | 01:38 |
| *** nosnos has joined #openstack-swift | 01:45 | |
| *** occup4nt has joined #openstack-swift | 01:49 | |
| *** d89 has joined #openstack-swift | 01:49 | |
| *** matsuhashi has joined #openstack-swift | 01:49 | |
| *** creiht has joined #openstack-swift | 01:49 | |
| *** h6w has joined #openstack-swift | 01:49 | |
| *** zigo has joined #openstack-swift | 01:49 | |
| *** StevenK has joined #openstack-swift | 01:49 | |
| *** zul has joined #openstack-swift | 01:49 | |
| *** CrackerJackMack has joined #openstack-swift | 01:49 | |
| *** wer has joined #openstack-swift | 01:49 | |
| *** sudorandom has joined #openstack-swift | 01:49 | |
| *** dickson.freenode.net sets mode: +v creiht | 01:49 | |
| clarkb | portante: with that vals list sleep doesn't happen until the 4th entry in the list | 01:55 |
| portante | clarkb: hmm | 01:55 |
| portante | and is scales time.time() by 1000, which is basically converting it to nanoseconds | 01:56 |
| clarkb | at that point the ratelimiting kick in and sleep happens | 01:56 |
| portante | but I think it wants to scale it to milliseconds, dividing my 1000 | 01:57 |
| portante | then the now-running_time comparison works, all in milliseconds | 01:57 |
| portante | ah | 01:59 |
| portante | sorry | 01:59 |
| portante | time.time() returns seconds as a float | 01:59 |
| portante | okay, that makes sense | 02:00 |
| *** acoles_ has joined #openstack-swift | 02:00 | |
| clarkb | ya | 02:00 |
| *** d89 has quit IRC | 02:01 | |
| portante | so that code assumes that the running_time value passed in is in milliseconds | 02:01 |
| clarkb | portante: right and the test doesnt | 02:03 |
| clarkb | but the first iteration sets it to now which may cause enough skew the fail the check at the end | 02:03 |
| clarkb | but docstring says start at 0 is good | 02:04 |
| openstackgerrit | A change was merged to openstack/swift: Merge branch 'master' into feature/ec https://review.openstack.org/83234 | 02:05 |
| portante | hmm, want me to post a fix? | 02:05 |
| portante | or are you considering a fix? | 02:05 |
| portante | clarkb: ^^^ | 02:07 |
| clarkb | portante: you can | 02:07 |
| portante | k | 02:07 |
| openstackgerrit | A change was merged to openstack/swift: Fix broker.storage_policy_index after .set_storage_policy_index https://review.openstack.org/83228 | 02:08 |
| clarkb | portante: I am not entierly sure I grok the proper fix yet | 02:09 |
| portante | I don't either, but am working it | 02:10 |
| *** Trixboxer has quit IRC | 02:13 | |
| portante | clarkb: I think that test is a candidate for mocking out time.time() to verify the code works as expected | 02:14 |
| clarkb | portante: ++ | 02:15 |
| clarkb | portante: also I think the assert should multiple its values by 10 so that the scale is the same as the function it is testing | 02:15 |
| clarkb | portante: remapping the scales was a pita but made sense once I figured itout | 02:15 |
| clarkb | self.assertTrue(abs(500 - (time.time() - start) * 1000) < 100) | 02:16 |
| clarkb | would be much more readable | 02:16 |
| portante | right, it is multiplying by only 100 | 02:16 |
| *** d89 has joined #openstack-swift | 02:20 | |
| portante | clarkb: do we have a bug tracking this already? | 02:20 |
| clarkb | portante: no, I will file one now | 02:21 |
| portante | thanks | 02:21 |
| *** saschpe has quit IRC | 02:25 | |
| clarkb | portante: https://bugs.launchpad.net/swift/+bug/1298154 | 02:25 |
| portante | thx | 02:25 |
| clarkb | portante: I am about ready to afk for the evening. thank you for looking at this | 02:26 |
| portante | clarkb: welcome | 02:26 |
| portante | have a good evening | 02:26 |
| *** saschpe has joined #openstack-swift | 02:29 | |
| *** gyee has quit IRC | 02:37 | |
| *** matsuhashi has quit IRC | 02:37 | |
| *** matsuhashi has joined #openstack-swift | 02:38 | |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Move like unit tests together; add comments https://review.openstack.org/83276 | 02:41 |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Mock out time and sleep to avoid races https://review.openstack.org/83277 | 02:41 |
| portante | clarkb, notmyname: ^^^ should fix the gate failure that clarkb mentioned | 02:41 |
| portante | wicked ugly code copying goin' on there, gonna spend more time to reduce that | 02:42 |
| portante | open to other suggestions, too late to think straight | 02:42 |
| *** matsuhashi has quit IRC | 02:47 | |
| *** godb has joined #openstack-swift | 02:54 | |
| *** judd7_ has quit IRC | 03:03 | |
| *** nosnos has quit IRC | 03:04 | |
| sunh | I am curious how swift ensures that during redistribution of ring , which request should go to which server. .. there could be some wrong assignment during ring copy operation.. how it is handled.. | 03:04 |
| *** fifieldt_ has joined #openstack-swift | 03:09 | |
| *** RockKuo has joined #openstack-swift | 03:34 | |
| *** nosnos has joined #openstack-swift | 03:48 | |
| *** matsuhashi has joined #openstack-swift | 03:48 | |
| openstackgerrit | Zhang Hua proposed a change to openstack/swift: Add profiling middleware in Swift https://review.openstack.org/53270 | 03:50 |
| *** chandan_kumar has joined #openstack-swift | 03:52 | |
| *** otoolee has quit IRC | 03:58 | |
| *** erlon has quit IRC | 04:02 | |
| godb | clayg : hi, i try to test delete file and directory that expired | 04:03 |
| godb | object-expirer used "x-delete-after" | 04:04 |
| godb | time has passed, .data file has changed to the .ts file.. and delete flag has reversed in the sqltie3 object table | 04:06 |
| godb | but, i wonder how long it takes to replicator delete .ts file and directory | 04:07 |
| *** MooingLemur has quit IRC | 04:08 | |
| godb | in addition, useless row in database and .db file when manke cleared, is it replicator's job?? | 04:09 |
| *** MooingLemur has joined #openstack-swift | 04:10 | |
| *** haomaiw__ has quit IRC | 04:14 | |
| *** haomaiwang has joined #openstack-swift | 04:15 | |
| *** haomaiw__ has joined #openstack-swift | 04:19 | |
| *** haomaiwang has quit IRC | 04:23 | |
| *** chandan_kumar has quit IRC | 04:34 | |
| *** fifieldt_ has quit IRC | 04:38 | |
| *** saju_m has joined #openstack-swift | 04:38 | |
| *** matsuhashi has quit IRC | 04:45 | |
| *** matsuhas_ has joined #openstack-swift | 04:49 | |
| *** matsuhas_ has quit IRC | 05:12 | |
| *** Midnightmyth has joined #openstack-swift | 05:13 | |
| *** matsuhashi has joined #openstack-swift | 05:29 | |
| *** chandan_kumar has joined #openstack-swift | 05:30 | |
| *** mmcardle has joined #openstack-swift | 05:36 | |
| *** zaitcev has quit IRC | 05:43 | |
| *** mmcardle has quit IRC | 05:47 | |
| *** ppai has joined #openstack-swift | 05:54 | |
| *** saju_m has quit IRC | 06:07 | |
| _sEBAs_ | somebody is having an issue at the openstach dashboard swift uploader? like UnicodeDecodeError after submiting a file | 06:17 |
| *** psharma has joined #openstack-swift | 06:18 | |
| *** Midnightmyth has quit IRC | 06:26 | |
| *** russellb has quit IRC | 06:32 | |
| anticw | _sEBAs_: object path? | 06:40 |
| anticw | is it valid utf-8 ? | 06:40 |
| _sEBAs_ | anticw: must be? | 06:41 |
| _sEBAs_ | anticw: cant just be any kind of file? | 06:41 |
| _sEBAs_ | i tried through command line "swift upload ...." and it worked ! but through the dashboard doesn't :( | 06:42 |
| _sEBAs_ | btw thanks for answer ;) | 06:42 |
| _sEBAs_ | anticw: Im re installing the openstack | 06:45 |
| _sEBAs_ | using the last branch of devstack "master" | 06:45 |
| *** saju_m has joined #openstack-swift | 06:50 | |
| *** nshaikh has joined #openstack-swift | 06:54 | |
| *** mrsnivvel has joined #openstack-swift | 07:04 | |
| *** cheri has joined #openstack-swift | 07:32 | |
| cheri | HI All, | 07:43 |
| cheri | https://review.openstack.org/#/c/47713/12 | 07:43 |
| *** russellb has joined #openstack-swift | 07:47 | |
| openstackgerrit | Zhang Hua proposed a change to openstack/swift: Add profiling middleware in Swift https://review.openstack.org/53270 | 08:01 |
| *** ppai has quit IRC | 08:19 | |
| *** matsuhashi has quit IRC | 08:33 | |
| *** manish_ has joined #openstack-swift | 08:40 | |
| manish_ | Hi anyone there? | 08:41 |
| *** saju_m has quit IRC | 08:43 | |
| *** nosnos has quit IRC | 08:43 | |
| manish_ | clayg: Hi, there!! Can I trouble you with some queries?? | 08:44 |
| hugokuo | manish_: may I ? | 08:47 |
| manish_ | hugokuo: sure | 08:51 |
| *** nacim has joined #openstack-swift | 08:52 | |
| manish_ | What happens when object server crashes before renaming the temp file to data file | 08:52 |
| manish_ | Who actually removes the temp file? | 08:52 |
| *** mmcardle has joined #openstack-swift | 08:53 | |
| godb | hello everybody. | 08:55 |
| *** saju_m has joined #openstack-swift | 08:55 | |
| manish_ | hugokuo: ^^^ | 09:00 |
| *** omame has quit IRC | 09:03 | |
| *** omame has joined #openstack-swift | 09:05 | |
| manish_ | hugokuo: there? | 09:16 |
| *** saju_m has quit IRC | 09:16 | |
| hugokuo | manish_: yup... am reading the code for your question who removes the temp file | 09:16 |
| manish_ | oh..thanks :) | 09:16 |
| hugokuo | it looks a bit different than before(1.4.x). | 09:17 |
| manish_ | hmmm..ok | 09:25 |
| *** saju_m has joined #openstack-swift | 09:27 | |
| *** tanee-away is now known as tanee | 09:35 | |
| *** tburnes has quit IRC | 09:36 | |
| *** bvandenh has joined #openstack-swift | 09:37 | |
| *** tburnes has joined #openstack-swift | 09:37 | |
| hugokuo | manish_: ok... replicator | 09:39 |
| hugokuo | I mean object-replicator | 09:39 |
| hugokuo | The object-replicator will call unlink file path if the tmp data resided over the reclaom_age. Then just remove it. | 09:41 |
| *** mlipchuk has joined #openstack-swift | 09:43 | |
| *** matsuhashi has joined #openstack-swift | 09:45 | |
| saurabh_ | Hi there, what is the maximum memory requirement of proxy-server when handling 1024 request with 0 worker process | 09:50 |
| saurabh_ | hugokuo: ^^^^ | 09:50 |
| hugokuo | saurabh_: Don't get it. Do you mean the max memory usage ? | 09:52 |
| saurabh_ | hugokuo: yes for a worker process when handling 1024 request in parallel | 09:53 |
| hugokuo | for 1024 concurrent requests ? | 09:53 |
| saurabh_ | yes | 09:53 |
| *** d89 has quit IRC | 09:54 | |
| hugokuo | Well, I never care about that tho. Proxy-server not memory bound component. The Memory is for memcached and caching rings. | 09:54 |
| hugokuo | In most case, proxy-server is CPU bounded ... | 09:56 |
| openstackgerrit | Amala Basha proposed a change to openstack/python-swiftclient: swiftclient should check response before upload https://review.openstack.org/83330 | 10:03 |
| saurabh_ | hugokuo: ok thnx | 10:08 |
| hugokuo | you are welcome ... | 10:09 |
| manish_ | Thanks Hugokuo.... | 10:13 |
| manish_ | One more query : If data file has been created properly in target path but object server crashes even beore sending PUT request to container server? | 10:14 |
| manish_ | what will happen in above scenario? Some process deletes thhis data file or some process updated the container DB? | 10:15 |
| hugokuo | manish_: not a problem tho... b/c the rest two replica will update the container DB | 10:15 |
| hugokuo | And all(3replicas) container DB will consistency eventually :) | 10:16 |
| manish_ | ok...if one replica has only passed and two had failed...so in this case who will delete the data file placed by successful object server? | 10:16 |
| hugokuo | manish_: very good question..... Let me find a discussion in mailing list which answered by torgomatic ... That's brilliant. | 10:17 |
| hugokuo | manish_: actually.... The only copy will not be deleted ....... But user will get a 503. But 503 doesn't mean the object PUT was a "failed" operation. | 10:19 |
| * hugokuo looking for the thread | 10:19 | |
| godb | upoos | 10:24 |
| godb | asdf | 10:24 |
| *** saju_m has quit IRC | 10:24 | |
| godb | hu... why don't delete .ts file!! :( | 10:24 |
| godb | hi hugokuo | 10:25 |
| hugokuo | godb: b/c it's not over the reclaim_age .... | 10:25 |
| godb | i setting recalim_age 120 second | 10:25 |
| godb | and swfit version upgrede to 1.13 ver | 10:25 |
| hugokuo | godb: is replicator running ? | 10:25 |
| godb | of courcs | 10:26 |
| hugokuo | godb: hmm.... give it a quick test in my end. 5 mins | 10:26 |
| hugokuo | manish_: sorry ... I can't find the thread from mailing archive. | 10:26 |
| manish_ | thanks hugokuo...but is it like that in this scenario, that object will forever be kept on disk? | 10:26 |
| manish_ | its ok... | 10:26 |
| godb | do you settinn recalim_age 5min for test? | 10:27 |
| hugokuo | manish_: or you can say, it was uploaded successfully...... | 10:28 |
| hugokuo | godb: nope.. I mean I need 5mins to test it on Swift 1.13 | 10:28 |
| *** EmilienM has quit IRC | 10:28 | |
| godb | thx.. i wait more. | 10:28 |
| manish_ | ok..but in this case since quorum was not acheived..so container DB would not be updated..right? | 10:28 |
| *** EmilienM has joined #openstack-swift | 10:28 | |
| godb | well.. it need object metadata?? to delete tombfile ? | 10:28 |
| godb | like .meta | 10:29 |
| hugokuo | manish_: I'd like to say "depends" | 10:30 |
| * hugokuo I do really want to find the thread for you. It's pretty nice to explain the philosophy.... | 10:31 | |
| godb | i used to upload "x-delete-after: 120" for the test | 10:31 |
| godb | and waiting and exploing | 10:32 |
| godb | the object file change to ".ts" | 10:33 |
| godb | but never delete .ts | 10:33 |
| godb | :( | 10:33 |
| hugokuo | godb: ..... | 10:34 |
| godb | T_T | 10:34 |
| hugokuo | godb: That's for object-expire feature right ? | 10:35 |
| hugokuo | godb: Let me clarify it.... the x-delete-after is != reclaim_age.... | 10:35 |
| hugokuo | The object were deleted and set as .ts for ensuring all replicas are removed ... | 10:36 |
| godb | that's right | 10:36 |
| godb | i know that too. | 10:36 |
| godb | after the expiration of the file (.ts file) | 10:38 |
| godb | i waiting to about reclaim_age .. but never delete .ts file | 10:38 |
| *** bvandenh has quit IRC | 10:40 | |
| godb | and.. how often do object-replicator check .ts file?? whenever each run?? or put event?? | 10:41 |
| *** nosnos has joined #openstack-swift | 10:44 | |
| hugokuo | godb: hmm.. same behavior here.... I need to have deeper look after dinner :) | 10:48 |
| *** bvandenh has joined #openstack-swift | 10:53 | |
| *** mkollaro has joined #openstack-swift | 10:55 | |
| godb | hugokuo : have good dinner :) i going home | 10:55 |
| *** godb has quit IRC | 10:55 | |
| *** shashi has joined #openstack-swift | 11:07 | |
| *** matsuhashi has quit IRC | 11:07 | |
| shashi | Hi, I have a qeury, if I have one worker process on proxy server so by default it can handle 1024 concurrent request, so how many request it can put on wait before starting dropping them?? | 11:08 |
| shashi | when 1024 request currently proccessing. | 11:09 |
| *** saju_m has joined #openstack-swift | 11:14 | |
| openstackgerrit | Victor Stinner proposed a change to openstack/python-swiftclient: Python 3: Replace iter.next() with six.next(iter) https://review.openstack.org/82555 | 11:16 |
| openstackgerrit | Victor Stinner proposed a change to openstack/python-swiftclient: Replace dict.iteritems() with dict.items() https://review.openstack.org/82557 | 11:21 |
| openstackgerrit | Victor Stinner proposed a change to openstack/python-swiftclient: Replace dict.iteritems() with dict.items() https://review.openstack.org/82557 | 11:21 |
| *** RockKuo has quit IRC | 11:25 | |
| *** matsuhashi has joined #openstack-swift | 11:32 | |
| *** saju_m has quit IRC | 11:33 | |
| *** gvernik has joined #openstack-swift | 11:34 | |
| *** saju_m has joined #openstack-swift | 11:35 | |
| *** matsuhashi has quit IRC | 11:45 | |
| *** shashi has quit IRC | 11:55 | |
| *** ppai has joined #openstack-swift | 11:59 | |
| *** mmcardle has quit IRC | 12:01 | |
| *** tanee is now known as tanee-away | 12:02 | |
| *** tanee-away is now known as tanee | 12:02 | |
| *** Trixboxer has joined #openstack-swift | 12:03 | |
| *** cheri has quit IRC | 12:03 | |
| *** ppai has quit IRC | 12:04 | |
| *** matsuhashi has joined #openstack-swift | 12:05 | |
| openstackgerrit | paul luse proposed a change to openstack/swift: Move Storage Policy validation from parser to class https://review.openstack.org/83252 | 12:07 |
| *** tanee is now known as tanee-away | 12:08 | |
| *** tanee-away is now known as tanee | 12:09 | |
| *** tanee is now known as tanee-away | 12:10 | |
| *** JuanManuelOlle has joined #openstack-swift | 12:12 | |
| *** tanee-away is now known as tanee | 12:12 | |
| *** mmcardle has joined #openstack-swift | 12:14 | |
| *** cheri has joined #openstack-swift | 12:21 | |
| *** ppai has joined #openstack-swift | 12:21 | |
| *** tanee is now known as tanee-away | 12:22 | |
| *** tanee-away is now known as tanee | 12:22 | |
| *** mkollaro has quit IRC | 12:29 | |
| *** mkollaro1 has joined #openstack-swift | 12:29 | |
| *** tanee is now known as tanee-away | 12:37 | |
| *** gordc has joined #openstack-swift | 12:40 | |
| *** matsuhashi has quit IRC | 12:45 | |
| *** matsuhashi has joined #openstack-swift | 12:45 | |
| *** tanee-away is now known as tanee | 12:48 | |
| *** tanee is now known as tanee-away | 12:48 | |
| *** matsuhashi has quit IRC | 12:50 | |
| chmouel | torgomatic: https://bugs.launchpad.net/swift/+bug/1298351 | 12:53 |
| *** tanee-away is now known as tanee | 12:53 | |
| chmouel | have you seen that bug before? ^ | 12:53 |
| chmouel | i think this is related to the change of having the ring stored as json | 12:53 |
| *** tanee is now known as tanee-away | 12:53 | |
| chmouel | (and simplejson woes) | 12:53 |
| *** boblebauce has joined #openstack-swift | 12:54 | |
| *** boblebauce has quit IRC | 12:54 | |
| *** boblebauce has joined #openstack-swift | 12:54 | |
| *** judd7 has joined #openstack-swift | 12:56 | |
| *** psharma has quit IRC | 12:56 | |
| *** jamie_h_ has joined #openstack-swift | 12:57 | |
| *** ppai has quit IRC | 13:02 | |
| *** tanee-away is now known as tanee | 13:06 | |
| *** tanee is now known as tanee-away | 13:06 | |
| *** cheri has quit IRC | 13:07 | |
| *** tanee-away is now known as tanee | 13:09 | |
| *** ppai has joined #openstack-swift | 13:15 | |
| *** changbl has quit IRC | 13:17 | |
| *** matsuhashi has joined #openstack-swift | 13:18 | |
| *** cheri has joined #openstack-swift | 13:20 | |
| *** nshaikh has quit IRC | 13:24 | |
| *** matsuhashi has quit IRC | 13:25 | |
| *** matsuhashi has joined #openstack-swift | 13:26 | |
| *** matsuhashi has quit IRC | 13:26 | |
| *** matsuhashi has joined #openstack-swift | 13:26 | |
| *** mkollaro has joined #openstack-swift | 13:26 | |
| *** mkollaro1 has quit IRC | 13:28 | |
| *** chandankumar_ has joined #openstack-swift | 13:35 | |
| *** ppai has quit IRC | 13:35 | |
| *** krtaylor has quit IRC | 13:35 | |
| *** Stevo__ has joined #openstack-swift | 13:47 | |
| *** Stevo__ has left #openstack-swift | 13:49 | |
| openstackgerrit | Constantine Peresypkin proposed a change to openstack/swift: account to account copy implementation https://review.openstack.org/72157 | 13:57 |
| *** zigo has quit IRC | 14:01 | |
| *** zigo has joined #openstack-swift | 14:03 | |
| *** krtaylor has joined #openstack-swift | 14:06 | |
| *** mordred has quit IRC | 14:08 | |
| *** dmsimard has joined #openstack-swift | 14:08 | |
| *** boblebauce has quit IRC | 14:13 | |
| *** boblebauce has joined #openstack-swift | 14:36 | |
| *** gvernik has quit IRC | 14:36 | |
| *** chandankumar_ has quit IRC | 14:46 | |
| *** cheri has quit IRC | 14:47 | |
| *** piyush1 has joined #openstack-swift | 14:50 | |
| *** chandankumar_ has joined #openstack-swift | 14:53 | |
| *** jergerber has joined #openstack-swift | 14:55 | |
| *** mmcardle has quit IRC | 14:55 | |
| *** mmcardle has joined #openstack-swift | 15:04 | |
| torgomatic | chmouel: I have not seen that before, but I'd certainly believe it. This just reinforces my view that simplejson is awful. | 15:13 |
| chmouel | torgomatic: indeed, will get the bug reporter to clean the patch and add test to submit it | 15:14 |
| *** erlon has joined #openstack-swift | 15:15 | |
| *** matsuhashi has quit IRC | 15:18 | |
| *** nosnos has quit IRC | 15:19 | |
| *** chandankumar_ has quit IRC | 15:24 | |
| openstackgerrit | Greg Lange proposed a change to openstack/swift: Unify backend logging https://review.openstack.org/83237 | 15:25 |
| *** mmcardle has quit IRC | 15:30 | |
| *** mmcardle has joined #openstack-swift | 15:36 | |
| *** hurricanerix has joined #openstack-swift | 15:39 | |
| openstackgerrit | A change was merged to openstack/swift: Invert which proxy logging middleware instance logs https://review.openstack.org/82909 | 15:40 |
| *** gyee has joined #openstack-swift | 15:41 | |
| *** manish_ has quit IRC | 15:41 | |
| *** saju_m has quit IRC | 15:44 | |
| *** cheri has joined #openstack-swift | 15:56 | |
| *** csd has joined #openstack-swift | 15:58 | |
| *** changbl has joined #openstack-swift | 16:00 | |
| *** cheri has quit IRC | 16:01 | |
| *** zanc has joined #openstack-swift | 16:07 | |
| notmyname | good morning | 16:13 |
| *** catintheroof has joined #openstack-swift | 16:18 | |
| *** catintheroof has left #openstack-swift | 16:18 | |
| *** mmcardle has quit IRC | 16:20 | |
| *** zigo has quit IRC | 16:41 | |
| *** mmcardle has joined #openstack-swift | 16:44 | |
| *** nacim has quit IRC | 16:48 | |
| *** nacim has joined #openstack-swift | 16:48 | |
| *** mlipchuk has quit IRC | 16:56 | |
| portante | mornin' | 16:59 |
| luisbg | portante, morning :) | 17:03 |
| portante | luisbg: top-o'-the-mornin'-to-ya | 17:06 |
| luisbg | portante, very well sir | 17:06 |
| *** piyush1 has quit IRC | 17:08 | |
| *** bvandenh has quit IRC | 17:20 | |
| *** Midnightmyth has joined #openstack-swift | 17:20 | |
| peluse | good morning! | 17:21 |
| *** tanee is now known as tanee-away | 17:24 | |
| *** piousbox has joined #openstack-swift | 17:24 | |
| *** zaitcev has joined #openstack-swift | 17:25 | |
| *** ChanServ sets mode: +v zaitcev | 17:25 | |
| openstackgerrit | Greg Lange proposed a change to openstack/swift: Unify backend logging https://review.openstack.org/83237 | 17:26 |
| *** shri has joined #openstack-swift | 17:28 | |
| *** bada has quit IRC | 17:28 | |
| *** lpabon has joined #openstack-swift | 17:29 | |
| *** boblebauce has left #openstack-swift | 17:30 | |
| anticw | _sEBAs_: the container/object names should be valid utf-8 as there are places which have inplicit assumptions along those lines | 17:34 |
| anticw | it could be relaxed with some effort but i think it's quite reasonable as-is | 17:35 |
| anticw | peluse: the EC patches ... in whatever state they are in ... are they in the public eye anywhere i can browse? | 17:35 |
| peluse | yes, there are only a few right now. One sec | 17:37 |
| _sEBAs_ | anticw: yeah, actually there is a patch already to treat that, but, I think that error is caused by an untreated throw exception | 17:37 |
| peluse | anticw: https://review.openstack.org/#/c/66482/ https://review.openstack.org/#/c/66906/ https://review.openstack.org/#/c/66905/ | 17:39 |
| *** zigo has joined #openstack-swift | 17:39 | |
| peluse | anticw: note that there is a new requirement not yet approved that prevents these from passing on Jenkins so many will show abandoned because of that but they're not. Also, not guarantee that any have been recently rebased. Once SP goes through they'll be cleaned up and made more consumable | 17:39 |
| anticw | understood, i'm looking over the second one right now | 17:40 |
| anticw | peluse: the EC fragments are stored across multiple disks on the same object server still i take it? | 17:41 |
| anticw | s/disks/devices/ | 17:41 |
| peluse | no, fragments are scaterred over multiple storage nodes | 17:41 |
| anticw | oh, i see, you're fragmentin the the proxy with a minimally bufferred stream | 17:42 |
| anticw | (that came out quite garbled sorry but i think it makes sense) | 17:43 |
| peluse | yes, we'll buffer up to N size of an object (1MB right now I think) and then EC that, spread it out over the nodes, then do the next portion of the object, etc., etc. The fragments are not stored as separate objects, they are appended (so we aren't doing a whole ton of 1MB files per object) | 17:43 |
| *** tanee-away is now known as tanee | 17:43 | |
| anticw | right, i actually have a use case for 1MB objects and having tons of them :) | 17:44 |
| peluse | clarrification: fragments are actuially objects, I meant the 1MB sections of fragments. | 17:44 |
| peluse | excellent! | 17:44 |
| peluse | if you obj is 1MB in size... very good fit | 17:44 |
| peluse | but its all tunable and we haven't characterized much of anything yet... | 17:45 |
| anticw | peluse: no, my storage target potentally wants <= 1MB chunks | 17:45 |
| peluse | cool, like I said, lots of knobs still to mess with for various use cases... | 17:46 |
| *** nacim has quit IRC | 17:48 | |
| *** tanee is now known as tanee-away | 18:10 | |
| *** tanee-away is now known as tanee | 18:12 | |
| *** piyush1 has joined #openstack-swift | 18:13 | |
| *** pandemicsyn has quit IRC | 18:17 | |
| *** pandemicsyn has joined #openstack-swift | 18:18 | |
| *** lpabon has quit IRC | 18:19 | |
| *** pandemicsyn has quit IRC | 18:20 | |
| *** pandemicsyn has joined #openstack-swift | 18:21 | |
| *** jamie_h_ has quit IRC | 18:22 | |
| *** gholt has quit IRC | 18:25 | |
| *** gholt has joined #openstack-swift | 18:26 | |
| *** ChanServ sets mode: +v gholt | 18:26 | |
| *** mkollaro has quit IRC | 18:28 | |
| *** mmcardle has quit IRC | 18:31 | |
| *** csd_ has joined #openstack-swift | 18:32 | |
| *** gholt has quit IRC | 18:34 | |
| *** csd has quit IRC | 18:35 | |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Only run flake8 on Swift code https://review.openstack.org/83177 | 18:44 |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Add object-reconciler daemon https://review.openstack.org/80730 | 18:47 |
| *** ChanServ sets mode: +v acoles_ | 18:49 | |
| *** creiht has quit IRC | 18:57 | |
| *** creiht has joined #openstack-swift | 18:57 | |
| *** ChanServ sets mode: +v creiht | 18:57 | |
| *** MooingLemur has quit IRC | 19:12 | |
| *** MooingLemur has joined #openstack-swift | 19:12 | |
| *** tanee is now known as tanee-away | 19:16 | |
| *** csd_ has quit IRC | 19:17 | |
| *** csd_ has joined #openstack-swift | 19:21 | |
| gordc | creiht: ping | 19:25 |
| *** gholt has joined #openstack-swift | 19:34 | |
| *** ChanServ sets mode: +v gholt | 19:37 | |
| *** gholt has quit IRC | 19:47 | |
| *** gholt has joined #openstack-swift | 19:48 | |
| *** ChanServ sets mode: +v gholt | 19:48 | |
| *** piyush has joined #openstack-swift | 19:51 | |
| *** piyush1 has quit IRC | 19:53 | |
| *** csd_ has quit IRC | 19:58 | |
| *** cschwede has quit IRC | 20:05 | |
| openstackgerrit | Pete Zaitcev proposed a change to openstack/swift: Pluggable Back-ends for account and container servers https://review.openstack.org/47713 | 20:06 |
| *** cschwede_ has joined #openstack-swift | 20:09 | |
| *** cschwede_ has quit IRC | 20:10 | |
| *** cschwede has joined #openstack-swift | 20:10 | |
| *** cschwede has quit IRC | 20:10 | |
| *** cschwede has joined #openstack-swift | 20:11 | |
| *** csd_ has joined #openstack-swift | 20:15 | |
| *** ChanServ sets mode: +v cschwede | 20:20 | |
| *** mkollaro has joined #openstack-swift | 20:22 | |
| notmyname | I just sent an email to the -dev mailing list: acoles_ and cschwede are now part of swift core | 20:35 |
| notmyname | (so now you can bug them for reviews too ;-)) | 20:35 |
| *** fifieldt has quit IRC | 20:37 | |
| *** Trixboxer has quit IRC | 20:39 | |
| *** EmilienM has quit IRC | 20:42 | |
| *** EmilienM has joined #openstack-swift | 20:43 | |
| creiht | gordc: howdy | 20:43 |
| *** jamie_h_ has joined #openstack-swift | 20:44 | |
| gordc | creiht: hey | 20:48 |
| gordc | creiht: i was hoping, when you have time, could you take a look at https://review.openstack.org/#/c/82131/ to verify we're using middleware logging appropriately. | 20:48 |
| *** bobf has joined #openstack-swift | 20:48 | |
| *** fifieldt has joined #openstack-swift | 20:49 | |
| creiht | gordc: it should be WARNING instead of WARN | 20:50 |
| creiht | otherwise that *should* work | 20:50 |
| *** bobf1 has joined #openstack-swift | 20:50 | |
| gordc | creiht: cool. i'll let Nadya know. it seems to have cleared logs with WARN but i guess it's just filtering everything. | 20:52 |
| creiht | heh | 20:52 |
| creiht | there *may* be an alias for that | 20:52 |
| creiht | but the python docs indicate WARNING | 20:52 |
| creiht | for loging | 20:52 |
| gordc | creiht: i see. cool cool. thanks for taking a look. | 20:53 |
| creiht | np | 20:53 |
| creiht | glad it is working out | 20:53 |
| portante | acoles_, cschwede welcome! | 20:56 |
| creiht | indeed | 21:00 |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Mock out time and sleep to avoid races https://review.openstack.org/83277 | 21:04 |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Move like unit tests together; add comments https://review.openstack.org/83276 | 21:04 |
| portante | clarkb: ^^^ | 21:05 |
| portante | gotta head out to the '99s for a drink after three days of all day meetings, back online a bit later | 21:05 |
| * portante out | 21:05 | |
| creiht | gordc: just tested, and it seems WARN works as well | 21:06 |
| cschwede | portante: thank you very much - i'm still a little bit speechless :) | 21:06 |
| creiht | cschwede: just wait for the coronation ceremony :) | 21:07 |
| gordc | creiht: cool. think we need to change it in patch still? | 21:07 |
| creiht | nah | 21:07 |
| creiht | unless you want to for some consistency reason | 21:07 |
| cschwede | creiht: ha, now you make me even more curious :) | 21:08 |
| gordc | creiht: lol consistency isn't in my repetoire... i'll leave it for now. | 21:08 |
| creiht | cschwede: hehe | 21:09 |
| creiht | just kidding :) | 21:09 |
| *** JuanManuelOlle has quit IRC | 21:11 | |
| clarkb | portante: thanks again | 21:19 |
| openstackgerrit | A change was merged to openstack/swift: Write out ring.gz's in a safer fashion https://review.openstack.org/83209 | 21:23 |
| *** jamie_h_ has quit IRC | 21:47 | |
| *** gordc has left #openstack-swift | 21:51 | |
| *** piyush has quit IRC | 21:58 | |
| *** sungju has joined #openstack-swift | 22:04 | |
| *** changbl has quit IRC | 22:07 | |
| *** csd_ has quit IRC | 22:22 | |
| *** sungju has quit IRC | 22:24 | |
| *** bobf1 has quit IRC | 22:29 | |
| Alex_Gaynor | There's a handful of small patches for siwftclient on py3k that shoudl be really easy to get merged, they're the set I just +1'd :-) | 22:29 |
| *** bobf has quit IRC | 22:30 | |
| zaitcev | peluse: Please have a look at https://gist.github.com/zaitcev/9699561 (use "raw", it needs a 142 column screen) | 22:45 |
| *** jergerber has quit IRC | 22:47 | |
| *** dmsimard has quit IRC | 22:49 | |
| *** mkollaro has quit IRC | 22:54 | |
| rpedde | Qiot | 23:20 |
| rpedde | wow. | 23:20 |
| *** tdasilva has joined #openstack-swift | 23:20 | |
| peluse | zaitcev: will do | 23:32 |
| *** erlon has quit IRC | 23:32 | |
| zaitcev | peluse: Thanks. I meant to say "look while reading the PBE review" | 23:33 |
| peluse | zaitcev: cool | 23:36 |
| *** Midnightmyth has quit IRC | 23:42 | |
| *** early has quit IRC | 23:49 | |
| *** early has joined #openstack-swift | 23:53 | |
| *** rturk-away has quit IRC | 23:54 | |
| *** rturk-away has joined #openstack-swift | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!