| *** amandap has quit IRC | 00:00 | |
| *** saschpe has quit IRC | 00:00 | |
| *** amandap has joined #openstack-swift | 00:01 | |
| *** zaitcev has quit IRC | 00:01 | |
| *** saschpe has joined #openstack-swift | 00:06 | |
| *** d89 has joined #openstack-swift | 00:28 | |
| *** matsuhashi has joined #openstack-swift | 00:31 | |
| *** shri has quit IRC | 00:41 | |
| *** sunh has quit IRC | 00:46 | |
| *** _sEBAs_ has quit IRC | 00:59 | |
| *** judd7 has quit IRC | 01:03 | |
| *** RockKuo has joined #openstack-swift | 01:14 | |
| *** RockKuo_ has joined #openstack-swift | 01:16 | |
| *** RockKuo has quit IRC | 01:16 | |
| *** SpamapS has joined #openstack-swift | 01:19 | |
| SpamapS | FYI, http://git.openstack.org/cgit/openstack/swift/commit/?id=4b075c107201e3064418d3e5dc41f0ec5ca02e7d has broken TripleO .. | 01:19 |
|---|---|---|
| SpamapS | tempfile.NamedTemporaryFile creates files with a stricter umask | 01:19 |
| SpamapS | making the ring file 0600 instead of 0644 | 01:19 |
| SpamapS | devstack is not affected because of https://git.openstack.org/cgit/openstack-dev/devstack/tree/lib/swift#n285 .. which is a pretty big hammer. | 01:20 |
| SpamapS | We'll adapt, but just FYI.. users may experience issues. | 01:20 |
| *** piousbox has quit IRC | 01:29 | |
| *** nosnos has joined #openstack-swift | 01:37 | |
| *** csd_ has joined #openstack-swift | 01:38 | |
| gholt | pandemicsyn: ^ | 01:44 |
| *** d89 has quit IRC | 01:44 | |
| *** tdasilva has left #openstack-swift | 01:45 | |
| *** godb has joined #openstack-swift | 01:56 | |
| *** d89 has joined #openstack-swift | 02:02 | |
| *** h6w has quit IRC | 02:04 | |
| *** csd_ has quit IRC | 02:05 | |
| hugokuo | godb: HI | 02:16 |
| *** saschpe has quit IRC | 02:24 | |
| *** saschpe has joined #openstack-swift | 02:24 | |
| *** RockKuo_ has quit IRC | 02:38 | |
| *** matsuhashi has quit IRC | 03:04 | |
| *** occup4nt has quit IRC | 03:08 | |
| *** matsuhashi has joined #openstack-swift | 03:10 | |
| *** nosnos has quit IRC | 03:17 | |
| *** changbl has joined #openstack-swift | 03:19 | |
| *** csd has joined #openstack-swift | 03:21 | |
| *** occupant has joined #openstack-swift | 03:29 | |
| *** matsuhashi has quit IRC | 03:31 | |
| *** RockKuo has joined #openstack-swift | 03:41 | |
| openstackgerrit | Yuan Zhou proposed a change to openstack/swift: Update swift-object-info/swift-get-nodes to be storage policy aware https://review.openstack.org/82734 | 03:50 |
| *** matsuhashi has joined #openstack-swift | 03:53 | |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Mock out time and sleep to avoid races https://review.openstack.org/83277 | 03:57 |
| *** nosnos has joined #openstack-swift | 04:06 | |
| *** chandankumar_ has joined #openstack-swift | 04:23 | |
| *** jokke_ has quit IRC | 04:30 | |
| *** jokke_ has joined #openstack-swift | 04:30 | |
| *** ppai has joined #openstack-swift | 04:47 | |
| *** sungju has joined #openstack-swift | 04:57 | |
| *** sungju has quit IRC | 05:07 | |
| *** gyee has quit IRC | 05:33 | |
| *** sungju has joined #openstack-swift | 06:29 | |
| *** psharma has joined #openstack-swift | 06:30 | |
| ppai | Hi, does anyone know how to figure out which branches were not covered in nosetests when "BrMiss" is not zero. | 06:32 |
| *** saju_m has joined #openstack-swift | 06:38 | |
| *** csd has quit IRC | 06:52 | |
| *** bvandenh has joined #openstack-swift | 07:07 | |
| *** matsuhashi has quit IRC | 07:20 | |
| godb | hugokuo : hi :) | 07:22 |
| *** matsuhas_ has joined #openstack-swift | 07:24 | |
| godb | hugokuo : if you here. please call me. i have some question about swift handoff | 07:24 |
| hugokuo | godb: am back | 07:25 |
| godb | hi :) | 07:26 |
| hugokuo | hi | 07:26 |
| hugokuo | Had your .ts been removed ? | 07:26 |
| godb | today, i try to test about upload object / expired object / check .ts file and dir | 07:27 |
| godb | no, net yet :( | 07:27 |
| godb | my swift suffer from many .ts file.. it makes much disk inode | 07:27 |
| hugokuo | godb: it's normal. unless the partition content changed. | 07:28 |
| hugokuo | godb: If the partition been reused, then the .ts will be removed. | 07:28 |
| godb | how can i reuse partition?? | 07:29 |
| hugokuo | godb: The inode consumption question is worth to discuss here. | 07:29 |
| godb | hum.... | 07:30 |
| hugokuo | godb: once a new object been calculated to store on the same partition. That's swift's work... not yours | 07:30 |
| hugokuo | godb: But I recognize the inode consumption by .ts. I'd like to ask the logic with other. Maybe chmouel or notmyname. | 07:31 |
| godb | the partition have many objects but replicator never delete .ts file | 07:31 |
| hugokuo | godb: hmm... that's fun. give it a test to delete the hashed.pkl in a test partition directory. | 07:32 |
| godb | all replication files? | 07:32 |
| *** haomaiw__ has quit IRC | 07:33 | |
| hugokuo | After that, all replicas of the .ts will be flushed . | 07:33 |
| * hugokuo I tested it yesterday. | 07:33 | |
| godb | ok. i'll try it. | 07:33 |
| godb | that's good. but i will test on my hand :) | 07:34 |
| hugokuo | godb: so what can I help for the handoff question ? | 07:34 |
| godb | and .. i have some question about handoff | 07:34 |
| godb | yeah:) | 07:34 |
| godb | upload object -> expire object -> delete .data file and rename to .ts file | 07:35 |
| godb | and time is running out (reclaim_age) | 07:36 |
| godb | then, swift makes another directory | 07:37 |
| hugokuo | yes | 07:37 |
| godb | the dir consist .ts file ;( | 07:37 |
| godb | but .ts's timestamp is different | 07:38 |
| godb | i wonder why swift make new directory about expire object after reclaim_age | 07:38 |
| hugokuo | thinking ... never deep dive the behavior of object-expire. | 07:39 |
| godb | is it object handoff? i don't know why do that. for what? | 07:39 |
| godb | object-expirer demon delete .data file and change file name to timestamp.ts but i didnt know why make another dir and another .ts file | 07:40 |
| godb | do you know any doc about it ? | 07:41 |
| hugokuo | I guess the developer document or source code would be the best place. | 07:41 |
| hugokuo | For this question, I'm learning from code now. | 07:42 |
| godb | yes.. so i read source file a little.. replicator.py. diskfile.py. | 07:42 |
| godb | in the ../swift/obj/diskfile.py has logic about delete .ts file | 07:43 |
| godb | i will read consistently | 07:43 |
| godb | consistently | 07:45 |
| godb | and when expired object sqlite3 object table do change 'delete=1' | 07:46 |
| *** cheri has joined #openstack-swift | 07:47 | |
| godb | but in my think. it is needless row | 07:47 |
| godb | this row will be delete? when? who? | 07:47 |
| * hugokuo testing | 07:49 | |
| hugokuo | godb: I did not see "swift makes another directory" tho. Can your explain more ? | 07:59 |
| hugokuo | it looks normal for me. The 1395993134.85122.data turn into 1395993354.43130.ts , the timestamp of .ts is the timing when the object been deleted at. But still in the same directory. | 08:02 |
| hugokuo | There's no additional directory created here ... | 08:03 |
| godb | hugokuo : what is "reclaim_age" in your object-server.conf | 08:23 |
| godb | defalut is 1 week. | 08:23 |
| hugokuo | reclaim_age = 30 | 08:23 |
| godb | okey. after file rename to .ts | 08:24 |
| hugokuo | I test with expirer after 30secs. | 08:24 |
| hugokuo | k | 08:24 |
| godb | k | 08:24 |
| hugokuo | What's ur case | 08:24 |
| *** Manish_ has joined #openstack-swift | 08:24 | |
| godb | no more add dir?? | 08:24 |
| hugokuo | show me your's | 08:24 |
| Manish_ | hugokuo: Hi..there | 08:25 |
| hugokuo | prolly paste it on pastebin | 08:25 |
| hugokuo | Manish_: yes... but about to grab a coffee | 08:25 |
| Manish_ | ok..:) | 08:25 |
| openstackgerrit | Victor Stinner proposed a change to openstack/python-swiftclient: Python 3: Add six dependency https://review.openstack.org/83683 | 08:25 |
| Manish_ | Just one quick query.. | 08:25 |
| godb | yeah.. give me little time.. i will delete all partition and test it | 08:25 |
| Manish_ | if u have 2 minutes before coffee...:) | 08:26 |
| hugokuo | Manish_: k | 08:26 |
| Manish_ | In case of Object versioning..if i have a object by Name "Manish".... | 08:28 |
| godb | hugokuo : sorry, please give a pastebin url? | 08:28 |
| godb | here is not my desktop | 08:28 |
| hugokuo | godb: http://paste.openstack.org/ | 08:28 |
| Manish_ | if a new version comes, it will be ranemd to "006Manish/timestamp" and will be moved to Archive folder | 08:28 |
| godb | tk. | 08:28 |
| Manish_ | right? | 08:28 |
| Manish_ | Now can some one delete directly "006Manish/timestamp" file from archive folder? | 08:29 |
| hugokuo | Manish_: yes... the versioned object can be deleted directly by pointing the DELETE against it. | 08:30 |
| openstackgerrit | Clay Gerrard proposed a change to openstack/swift: Add object-reconciler daemon https://review.openstack.org/80730 | 08:30 |
| Manish_ | hugokuo: Thanks alot... | 08:31 |
| openstackgerrit | Prashanth Pai proposed a change to openstack/swift: Read-Only Middleware https://review.openstack.org/83687 | 08:34 |
| *** Longgeek_ has joined #openstack-swift | 08:38 | |
| *** haypo has joined #openstack-swift | 08:43 | |
| haypo | hi | 08:43 |
| haypo | i sent patches for swiftclient to port it to Python 3. chmouel just approved 4 patches. 2 remainings patches requires https://review.openstack.org/#/c/83683/ which adds six dependency | 08:47 |
| hugokuo | godb: I may understand what you mean now http://paste.openstack.org/show/74526/ | 08:48 |
| haypo | six is already used by almost all OpenStack projects and it is already required in test-requirements.txt of swiftclient | 08:48 |
| haypo | my goal is to fix all Python 3 issues in swiftclient and then make the py33 gate voting to not regress | 08:50 |
| *** matsuhas_ has quit IRC | 08:53 | |
| *** matsuhas_ has joined #openstack-swift | 08:57 | |
| *** psharma has quit IRC | 08:58 | |
| *** psharma has joined #openstack-swift | 08:58 | |
| *** saju_m has quit IRC | 09:05 | |
| *** mmcardle has joined #openstack-swift | 09:06 | |
| godb | hugokuo : yes. buf after reclaim age time | 09:11 |
| godb | swift make another dir | 09:11 |
| godb | i will check now | 09:12 |
| *** nacim has joined #openstack-swift | 09:12 | |
| hugokuo | hmm.. I tried the blueprint description of expirer for more details. But can't find it :( | 09:13 |
| godb | hugokuo : i heve same test result | 09:16 |
| godb | in my think.. | 09:16 |
| godb | nono,, | 09:16 |
| godb | i don'n understand why swift make .ts file and different "Ring Locations" | 09:17 |
| godb | anyway thx your help. i need more testing | 09:17 |
| godb | if i know this reason. tell you :) | 09:18 |
| hugokuo | godb: got it. | 09:19 |
| godb | hugokuo : good day. good week. i need some beer :( | 09:22 |
| hugokuo | godb: r u operator or developer ? | 09:24 |
| godb | i am developer | 09:24 |
| godb | live in south korea | 09:24 |
| *** saju_m has joined #openstack-swift | 09:25 | |
| godb | i develop c/c++ server side system mostly. | 09:25 |
| godb | do you? | 09:25 |
| godb | in my memory you live in singapore right? | 09:26 |
| hugokuo | cool.... Am much more works as an operator. Live in Taiwan but work in SwiftStack | 09:26 |
| hugokuo | That's nice to have asian people around this timezone XD | 09:26 |
| godb | oh. sorry. :) and that is very good luck to me | 09:27 |
| godb | because of talking with you. | 09:27 |
| godb | no one asked around me. | 09:28 |
| godb | and my english skills short ; | 09:28 |
| hugokuo | Well, it's good. I can have more experience with any kind of questions about Swift. Sometimes I know the behavior of a feature but not deep dive in the code. While someone has question about Swift, that means the question prolly asked by our customers in the future XD | 09:29 |
| hugokuo | no problem.... I think my english is suck... | 09:30 |
| hugokuo | haha | 09:30 |
| godb | thx kindly. | 09:30 |
| hugokuo | as this case, I learned much more of how expirer working in the underlying. So feel free to ping me on IRC. We can discuss thing here. Of course, I may not know everything in details. | 09:32 |
| godb | i considering make external script for delete .ts file and directory. but before that . i will more stduy and test about swift operation | 09:32 |
| *** psharma has quit IRC | 09:33 | |
| godb | sounds good .. i if anyone give question if i knows .. as possible as answer it. if he understand my engilist :( | 09:34 |
| godb | english | 09:34 |
| *** bada has joined #openstack-swift | 09:35 | |
| *** godb has quit IRC | 09:35 | |
| *** ahale has quit IRC | 09:39 | |
| hugokuo | I have same confusion about .ts occupied inodes forever if the partition hasn't been used by new object. | 09:41 |
| * hugokuo feel not make sense. Maybe there's some reason for such implementation. | 09:42 | |
| *** jamie_h_ has joined #openstack-swift | 09:47 | |
| *** Trixboxer has joined #openstack-swift | 09:53 | |
| *** Longgeek_ has quit IRC | 09:54 | |
| openstackgerrit | A change was merged to openstack/python-swiftclient: Replace dict.iteritems() with dict.items() https://review.openstack.org/82557 | 09:55 |
| *** Manish_ has quit IRC | 09:55 | |
| openstackgerrit | A change was merged to openstack/python-swiftclient: Python 3: Replace iter.next() with six.next(iter) https://review.openstack.org/82555 | 09:59 |
| *** matsuhas_ has quit IRC | 10:00 | |
| *** Midnightmyth has joined #openstack-swift | 10:03 | |
| *** Longgeek_ has joined #openstack-swift | 10:12 | |
| openstackgerrit | A change was merged to openstack/python-swiftclient: Python 3: Fix tests using temporary text files https://review.openstack.org/82554 | 10:14 |
| openstackgerrit | A change was merged to openstack/python-swiftclient: Python 3: cast map() result to list https://review.openstack.org/82553 | 10:14 |
| *** nosnos has quit IRC | 10:16 | |
| *** Longgeek_ has quit IRC | 10:46 | |
| *** Longgeek_ has joined #openstack-swift | 10:57 | |
| *** Longgeek_ has joined #openstack-swift | 10:58 | |
| *** d89 has quit IRC | 10:59 | |
| *** RockKuo has quit IRC | 11:05 | |
| *** RockKuo_iPad has joined #openstack-swift | 11:05 | |
| *** Longgeek_ has quit IRC | 11:07 | |
| *** lpabon has joined #openstack-swift | 11:07 | |
| *** Longgeek_ has joined #openstack-swift | 11:11 | |
| *** RockKuo_iPad has quit IRC | 11:16 | |
| *** RockKuo_iPad has joined #openstack-swift | 11:16 | |
| *** RockKuo_iPad has quit IRC | 11:17 | |
| *** RockKuo_iPad has joined #openstack-swift | 11:17 | |
| *** RockKuo_iPad has quit IRC | 11:17 | |
| *** RockKuo_iPad has joined #openstack-swift | 11:18 | |
| *** RockKuo_iPad has quit IRC | 11:24 | |
| *** Longgeek_ has quit IRC | 11:27 | |
| *** RockKuo_iPad has joined #openstack-swift | 11:41 | |
| *** RockKuo_iPad has quit IRC | 11:45 | |
| *** saju_m has quit IRC | 11:50 | |
| *** madhuri_ has joined #openstack-swift | 11:53 | |
| *** RockKuo_iPad has joined #openstack-swift | 12:01 | |
| *** tanee-away is now known as tanee | 12:07 | |
| *** mkollaro has joined #openstack-swift | 12:07 | |
| *** JuanManuelOlle has joined #openstack-swift | 12:09 | |
| *** JuanManuelOlle1 has joined #openstack-swift | 12:28 | |
| *** JuanManuelOlle has quit IRC | 12:30 | |
| *** Longgeek_ has joined #openstack-swift | 12:31 | |
| *** madhuri_ has quit IRC | 12:39 | |
| *** Longgeek_ has quit IRC | 12:40 | |
| *** tdasilva has joined #openstack-swift | 12:41 | |
| *** tanee is now known as tanee-away | 12:41 | |
| *** erlon has joined #openstack-swift | 12:45 | |
| *** Longgeek_ has joined #openstack-swift | 12:47 | |
| *** chandan_kumar has quit IRC | 12:51 | |
| *** tanee-away is now known as tanee | 12:59 | |
| *** Midnightmyth has quit IRC | 13:00 | |
| *** JuanManuelOlle1 has quit IRC | 13:10 | |
| *** cheri has quit IRC | 13:11 | |
| *** ppai has quit IRC | 13:11 | |
| *** tanee is now known as tanee-away | 13:11 | |
| *** RockKuo_iPad has quit IRC | 13:11 | |
| *** zigo has quit IRC | 13:13 | |
| *** zigo has joined #openstack-swift | 13:14 | |
| *** mmcardle has quit IRC | 13:15 | |
| *** saju_m has joined #openstack-swift | 13:15 | |
| *** mrsnivvel has quit IRC | 13:17 | |
| *** JuanManuelOlle has joined #openstack-swift | 13:19 | |
| *** mmcardle has joined #openstack-swift | 13:23 | |
| *** JuanManuelOlle has quit IRC | 13:38 | |
| *** JuanManuelOlle has joined #openstack-swift | 13:47 | |
| *** zigo has quit IRC | 13:47 | |
| *** zigo has joined #openstack-swift | 13:47 | |
| *** RockKuo_iPad has joined #openstack-swift | 13:48 | |
| *** zigo has quit IRC | 13:51 | |
| *** zigo has joined #openstack-swift | 13:55 | |
| portante | Regarding eventlet and Python3, from Sergey Shepelev: "Hopefully we can have Eventlet working on Python3 in next few months" | 13:56 |
| *** tanee-away is now known as tanee | 13:58 | |
| *** tanee is now known as tanee-away | 13:58 | |
| *** zigo has quit IRC | 13:59 | |
| *** RockKuo_iPad has quit IRC | 13:59 | |
| *** zigo has joined #openstack-swift | 13:59 | |
| *** nacim has quit IRC | 14:00 | |
| *** nacim has joined #openstack-swift | 14:01 | |
| *** Longgeek_ has quit IRC | 14:01 | |
| *** russellb is now known as rustlebee | 14:08 | |
| *** openstack has joined #openstack-swift | 14:21 | |
| *** Midnightmyth has joined #openstack-swift | 14:23 | |
| *** piyush1 has joined #openstack-swift | 14:27 | |
| *** Longgeek_ has joined #openstack-swift | 14:35 | |
| openstackgerrit | A change was merged to openstack/python-swiftclient: Python 3: Add six dependency https://review.openstack.org/83683 | 14:37 |
| *** csd has joined #openstack-swift | 14:52 | |
| creiht | portante: yeah I saw that | 14:54 |
| haypo | portante: nice | 14:57 |
| creiht | it is nice that they are making progress | 14:58 |
| openstackgerrit | Victor Stinner proposed a change to openstack/python-swiftclient: Python 3: Fix module names in import https://review.openstack.org/82556 | 15:04 |
| haypo | eventlet working on python3 will reduce the pressure to port openstack to python3 | 15:05 |
| chmouel | yes that would be very nice | 15:06 |
| chmouel | https://github.com/eventlet/eventlet/issues/6#issuecomment-38912904 | 15:07 |
| chmouel | @ewindisch please share it with OpenStack team that Eventlet has some momentum real progress into Python3 compatibility. | 15:07 |
| *** JuanManuelOlle has quit IRC | 15:08 | |
| portante | creiht: it might be worth our time to see the details of that progress to get a sense of their trajectory | 15:23 |
| *** saju_m has quit IRC | 15:35 | |
| openstackgerrit | Victor Stinner proposed a change to openstack/python-swiftclient: Python 3: Get compatible types from six https://review.openstack.org/82552 | 15:48 |
| creiht | for some measure of our :) | 15:50 |
| *** judd7 has joined #openstack-swift | 15:50 | |
| *** jogo is now known as flashgordon | 16:07 | |
| *** Longgeek_ has quit IRC | 16:08 | |
| *** changbl has quit IRC | 16:14 | |
| *** marcusvrn has joined #openstack-swift | 16:19 | |
| *** dencaval has joined #openstack-swift | 16:23 | |
| *** dencaval has quit IRC | 16:24 | |
| *** thebloggu has joined #openstack-swift | 16:33 | |
| *** gyee has joined #openstack-swift | 16:35 | |
| *** thebloggu has quit IRC | 16:45 | |
| *** thebloggu has joined #openstack-swift | 16:46 | |
| *** thebloggu has left #openstack-swift | 16:46 | |
| *** thebloggu has joined #openstack-swift | 16:46 | |
| thebloggu | can someone help me understand why my requests are really slow (between 10 to 20 s) altough not failing? how can i find the problem? i have 1 proxy node and 1 storage node in the same real machine and 4 virtual machines running storage nodes) and all of them seem to be operating fine | 16:58 |
| *** JuanManuelOlle has joined #openstack-swift | 17:02 | |
| *** therve has left #openstack-swift | 17:03 | |
| *** piousbox has joined #openstack-swift | 17:06 | |
| *** bvandenh has quit IRC | 17:13 | |
| *** saju_m has joined #openstack-swift | 17:19 | |
| *** zaitcev has joined #openstack-swift | 17:20 | |
| *** ChanServ sets mode: +v zaitcev | 17:20 | |
| *** changbl has joined #openstack-swift | 17:27 | |
| *** saju_m has quit IRC | 17:35 | |
| *** saju_m has joined #openstack-swift | 17:35 | |
| *** piyush1 has quit IRC | 17:36 | |
| openstackgerrit | paul luse proposed a change to openstack/swift: Fix issue with account broker and pre-SPI databases https://review.openstack.org/83840 | 17:44 |
| *** haypo has left #openstack-swift | 17:45 | |
| *** occupant has quit IRC | 17:54 | |
| *** occupant has joined #openstack-swift | 17:54 | |
| *** nacim has quit IRC | 17:54 | |
| *** krtaylor has quit IRC | 17:55 | |
| *** csd has quit IRC | 17:58 | |
| *** piyush1 has joined #openstack-swift | 18:01 | |
| *** mkerrin has quit IRC | 18:02 | |
| *** piyush2 has joined #openstack-swift | 18:03 | |
| *** piyush1 has quit IRC | 18:06 | |
| *** thebloggu has quit IRC | 18:12 | |
| *** thebloggu has joined #openstack-swift | 18:16 | |
| *** csd has joined #openstack-swift | 18:19 | |
| *** jamie_h_ has quit IRC | 18:35 | |
| *** openstackgerrit has quit IRC | 18:48 | |
| *** openstackgerrit has joined #openstack-swift | 18:48 | |
| piousbox | hi guys | 18:49 |
| piousbox | is there v1.1 of Swift API? | 18:49 |
| piousbox | Or v2? | 18:49 |
| notmyname | piousbox: no. just v1 | 18:56 |
| piousbox | thanx | 19:06 |
| *** jamie_h_ has joined #openstack-swift | 19:06 | |
| notmyname | piousbox: we've added things to the API, but if you have a client written 4 years ago, it should still work with current tip of master today | 19:07 |
| piousbox | right on. We don't expect there to be v2 right now, but I was asked to make sure that in fact there isn't one ;) | 19:08 |
| *** thebloggu has quit IRC | 19:10 | |
| *** mmcardle has quit IRC | 19:25 | |
| *** csd has quit IRC | 19:43 | |
| *** mkollaro has quit IRC | 19:49 | |
| *** lpabon has quit IRC | 19:53 | |
| *** krtaylor has joined #openstack-swift | 20:07 | |
| *** jamie_h_ has quit IRC | 20:09 | |
| *** csd has joined #openstack-swift | 20:17 | |
| *** dmsimard has joined #openstack-swift | 20:25 | |
| *** csd has quit IRC | 20:51 | |
| *** dmsimard1 has joined #openstack-swift | 20:53 | |
| *** dmsimard has quit IRC | 20:57 | |
| *** piyush2 has quit IRC | 21:06 | |
| *** jergerber has joined #openstack-swift | 21:08 | |
| *** mmcardle has joined #openstack-swift | 21:09 | |
| *** dmsimard1 has quit IRC | 21:10 | |
| *** JuanManuelOlle has quit IRC | 21:12 | |
| notmyname | really quiet in here today | 21:20 |
| *** shri has joined #openstack-swift | 21:23 | |
| *** marcusvrn has quit IRC | 21:38 | |
| * portante just hears the crickets and croaking frogs, hoping Spring will come to New England soon | 21:43 | |
| *** mmcardle has quit IRC | 21:53 | |
| *** judd7 has quit IRC | 21:58 | |
| openstackgerrit | Peter Portante proposed a change to openstack/swift: Speed up obj auditor sorting hsh dirs by inode # https://review.openstack.org/81590 | 22:05 |
| portante | notmyname: have two bug fix commits out and ready for review | 22:06 |
| portante | https://review.openstack.org/#/c/83276 | 22:07 |
| portante | https://review.openstack.org/#/c/83277 | 22:07 |
| notmyname | portante: ya, I was just going over what's outstanding when I got distracted by if or how much the proxy-logging bug is affecting our customers. doing some research on it | 22:07 |
| portante | They fix bug 1298154 | 22:08 |
| portante | notmyname: so we think that proxy-logging fix covers both SLO and DLO, right? | 22:08 |
| notmyname | yes I think so. but there is still the SLO range request thing | 22:09 |
| portante | I was not sure if there was a problem with DLO, had a hard time following the exchanges | 22:09 |
| portante | notmyname: yes, but that will take some consideration because we don't know to log that request as a SLO manifest until we read the object itself | 22:09 |
| notmyname | portante: basically it comes down to filtering the subrequests (swift.source set) or not | 22:10 |
| portante | we just don't know if an object is a SLO or DLO manifest until we get a hold of it | 22:10 |
| portante | really, until the headers come back | 22:10 |
| portante | so if SLO marked every GET that it thinks might be a SLO manifest, all GETs would end up getting logged twice at it stands today | 22:11 |
| portante | if SLO is in the pipeline | 22:12 |
| portante | well, that is obvious ... | 22:12 |
| portante | so the best that we can do is have SLO enter a fake log request when it knows it is a manifest that is going to be fetched, at least that is what I can see so far | 22:13 |
| portante | anyways, gotta go | 22:13 |
| portante | back on line later | 22:13 |
| notmyname | portante: running my last test now. I'll paste my summary | 22:14 |
| notmyname | portante: looks like everything is correct in both SLO and DLO in master | 22:16 |
| notmyname | portante: the segments are logged with swift.source set, and there is a final proxy log line without swift.source set that has the full number of bytes | 22:16 |
| notmyname | swifterdarrell: 1.11.0.1 looks fine for both SLO and DLO. one log line, no subrequests, b/w is correct. in 1.13.0.1, each chunk has the bytes but the log line without swift.source set only has the bytes in the manifest | 22:18 |
| *** Trixboxer has quit IRC | 22:19 | |
| openstackgerrit | John Dickinson proposed a change to openstack/python-swiftclient: set user-agent header https://review.openstack.org/83919 | 22:46 |
| *** JuanManuelOlle has joined #openstack-swift | 22:53 | |
| *** jergerber has quit IRC | 22:57 | |
| notmyname | there are some tempest patches for swift tests referenced at the bottom of https://wiki.openstack.org/wiki/Swift/PriorityReviews now | 23:08 |
| notmyname | the tempest team has asked that we take a look at them, please | 23:08 |
| *** erlon has quit IRC | 23:22 | |
| *** gyee has quit IRC | 23:31 | |
| *** mmcardle has joined #openstack-swift | 23:35 | |
| *** mmcardle has quit IRC | 23:37 | |
| *** JuanManuelOlle has quit IRC | 23:47 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!