| *** itlinux has joined #openstack-swift | 00:16 | |
| *** links has joined #openstack-swift | 00:40 | |
| openstackgerrit | Gerard Gine proposed openstack/swift master: Improved usage of args in .functests https://review.openstack.org/544142 | 01:03 |
|---|---|---|
| *** itlinux has quit IRC | 01:19 | |
| *** gyee has quit IRC | 01:25 | |
| *** m_kazuhiro has joined #openstack-swift | 01:37 | |
| m_kazuhiro | good morning | 01:38 |
| m_kazuhiro | mattoliverau: I listed related patches for expirer general task queue in https://etherpad.openstack.org/p/swift_general_task_queue . I want to merge patch #1 ~ #7 BEFORE PTG. | 01:43 |
| patchbot | https://review.openstack.org/#/c/1/ - openstack-infra/system-config - Add puppet module for ssh that installs an sshd_co... (MERGED) | 01:43 |
| *** itlinux has joined #openstack-swift | 01:44 | |
| mattoliverau | m_kazuhiro: morning, cool, I'll try and to some reviews today. Sorry work has had me working hard on some other things :( | 01:44 |
| m_kazuhiro | mattoliverau: ok! I'm looking forward to your reviews :) | 01:46 |
| kota_ | good morning | 01:58 |
| kota_ | hmm... the keystonemiddleware patch seems from the conversation in IRC (while I was asleep) | 02:02 |
| *** itlinux has quit IRC | 02:10 | |
| mattoliverau | kota_: morning | 02:27 |
| *** itlinux has joined #openstack-swift | 02:32 | |
| kota_ | mattoliverau: o/ | 02:33 |
| kota_ | mattoliverau: did you attend the voice conf on container sharding? | 02:33 |
| mattoliverau | kota_: yup :) | 02:34 |
| kota_ | mattoliverau: how's going? I missed acoles's reminder :/ | 02:35 |
| mattoliverau | it was a good chat, talked for about and hour about shard, sharding and unsharded replication with each other and from handoffs -> primary nodes. | 02:35 |
| mattoliverau | and the annoying edge cases. I think there are some notes on the etherpad about the discussions. I'll try and update it some more if it isn't entirely up to date :) | 02:36 |
| kota_ | sounds cool. Please let me know the pad. | 02:37 |
| mattoliverau | It ended up being, acoles, timburke, torgomatic and myself. | 02:37 |
| mattoliverau | kota_: https://etherpad.openstack.org/p/deep-containers-replication | 02:37 |
| *** notmyname has quit IRC | 02:38 | |
| kota_ | what I remember is we had discussed about around replication, async sharding work in the Denver PTG... | 02:38 |
| kota_ | mattoliverau: thx! | 02:38 |
| mattoliverau | looks like the 'proposal' is the part we ended up with. There is probably more to think about the area, but I think it was great progress :) | 02:38 |
| kota_ | hehe, "To rsync or not to rsync?" sounds nice discussion always for us :P | 02:39 |
| mattoliverau | yeah. and only 1.5 weeks until we can discuss it all again in person :) | 02:39 |
| mattoliverau | lol, yup | 02:39 |
| kota_ | too close, I cannot wait the PTG! | 02:39 |
| *** notmyname has joined #openstack-swift | 02:41 | |
| *** ChanServ sets mode: +v notmyname | 02:41 | |
| kota_ | it looks like i have to learn about the term of "usync" first. It seems a new component on the container sharding? | 02:42 |
| kota_ | oh | 02:44 |
| kota_ | no, that seems just I forgot the term. when i grepped the current master, usync seems to exist | 02:44 |
| *** itlinux has quit IRC | 02:44 | |
| kota_ | i see it's "Sync a db by sending all records since the last sync." | 02:45 |
| kota_ | we have 3 different sync routine in the db_replicator, just rsync, rsync_then_merge, and usync that sends the diff records. Make sense. | 02:46 |
| kota_ | routines | 02:46 |
| *** d0ugal has quit IRC | 02:49 | |
| mattoliverau | kota_: yeah, welcome to the world on swift db replication (or the container/account recplication) | 02:51 |
| mattoliverau | kota_: it's differernt to object replication | 02:51 |
| mattoliverau | usync is syncing chucks of whats changed. Then rsync (rsync or rsync then merge) is used if there isn't a file (rebalance) or if the differences are too great | 02:53 |
| kota_ | mattoliverau: i know! it's complicated but absolutely awesome works from the predecessors | 02:53 |
| kota_ | yup | 02:53 |
| kota_ | thanks mattoliverau I'll catch up them | 02:53 |
| * kota_ is going to lunch | 02:53 | |
| mattoliverau | some of it has been tweaked and changed in the feature/deep already (I needed to hack some for the sharding POC). So look at both :) | 02:54 |
| *** d0ugal has joined #openstack-swift | 02:59 | |
| kota_ | mattoliverau: ;) | 03:24 |
| *** gkadam has joined #openstack-swift | 03:24 | |
| openstackgerrit | Matthew Oliver proposed openstack/swift master: Add a basic partition diffing tool https://review.openstack.org/539466 | 03:27 |
| mattoliverau | m_kazuhiro: ^^ thanks so much, again for your review! | 03:27 |
| *** vinsh has joined #openstack-swift | 03:31 | |
| *** kei_yama has quit IRC | 03:39 | |
| m_kazuhiro | mattoliverau: great! I will check it today again! | 03:39 |
| *** links has quit IRC | 03:50 | |
| *** d0ugal has quit IRC | 03:55 | |
| *** vinsh has quit IRC | 03:59 | |
| *** d0ugal has joined #openstack-swift | 03:59 | |
| *** vinsh has joined #openstack-swift | 04:00 | |
| *** yashmurty has joined #openstack-swift | 04:02 | |
| *** kei_yama has joined #openstack-swift | 04:03 | |
| *** links has joined #openstack-swift | 04:07 | |
| *** ukaynar has quit IRC | 04:10 | |
| *** yashmurty has quit IRC | 04:10 | |
| *** ukaynar has joined #openstack-swift | 04:10 | |
| *** yashmurty has joined #openstack-swift | 04:12 | |
| *** yashmurty has quit IRC | 04:12 | |
| *** yashmurty has joined #openstack-swift | 04:12 | |
| *** gkadam_ has joined #openstack-swift | 04:13 | |
| *** gkadam_ has quit IRC | 04:14 | |
| *** gkadam_ has joined #openstack-swift | 04:14 | |
| *** ukaynar has quit IRC | 04:15 | |
| *** gkadam has quit IRC | 04:15 | |
| notmyname | kota_: the keystonemiddleware version i used is pulled from global requirements | 04:18 |
| *** vinsh has quit IRC | 04:18 | |
| *** cshastri has joined #openstack-swift | 04:20 | |
| *** m_kazuhiro_ has joined #openstack-swift | 04:37 | |
| *** m_kazuhiro has quit IRC | 04:39 | |
| *** yashmurty has quit IRC | 04:46 | |
| *** yashmurty has joined #openstack-swift | 04:50 | |
| *** yashmurty has quit IRC | 04:54 | |
| *** SkyRocknRoll has joined #openstack-swift | 04:59 | |
| *** psachin` has joined #openstack-swift | 05:04 | |
| *** psachin has joined #openstack-swift | 05:04 | |
| *** SkyRocknRoll has quit IRC | 05:11 | |
| *** yashmurty has joined #openstack-swift | 05:20 | |
| *** m_kazuhiro_ has quit IRC | 05:21 | |
| *** m_kazuhiro has joined #openstack-swift | 05:21 | |
| *** yashmurty has left #openstack-swift | 05:21 | |
| *** yashmurty has joined #openstack-swift | 05:22 | |
| *** psachin has quit IRC | 05:45 | |
| *** psachin` has quit IRC | 05:45 | |
| *** psachin has joined #openstack-swift | 05:51 | |
| *** psachin has quit IRC | 06:20 | |
| kota_ | notmyname: yeah, I confirmed that from timburke's comment and then, added my +2 +A | 06:28 |
| *** psachin has joined #openstack-swift | 06:28 | |
| notmyname | thanks :-) | 06:28 |
| *** yashmurty has quit IRC | 06:30 | |
| *** yashmurty has joined #openstack-swift | 06:32 | |
| *** geaaru has quit IRC | 07:03 | |
| *** ianychoi_ is now known as ianychoi | 07:08 | |
| openstackgerrit | Merged openstack/swift master: add keystonemiddleware to extras area https://review.openstack.org/544121 | 07:31 |
| *** armaan has joined #openstack-swift | 07:37 | |
| *** m_kazuhiro_ has joined #openstack-swift | 07:37 | |
| *** m_kazuhiro has quit IRC | 07:39 | |
| *** hseipp has joined #openstack-swift | 07:43 | |
| *** silor has joined #openstack-swift | 07:49 | |
| *** hseipp has quit IRC | 07:55 | |
| *** pcaruana has joined #openstack-swift | 07:56 | |
| *** armaan has quit IRC | 08:13 | |
| *** silor has quit IRC | 08:27 | |
| *** tesseract has joined #openstack-swift | 08:28 | |
| *** rcernin has quit IRC | 08:33 | |
| *** kei_yama has quit IRC | 08:34 | |
| *** geaaru has joined #openstack-swift | 08:39 | |
| *** gkadam__ has joined #openstack-swift | 08:50 | |
| *** gkadam_ has quit IRC | 08:53 | |
| *** links has quit IRC | 08:53 | |
| *** links has joined #openstack-swift | 09:06 | |
| *** cbartz has joined #openstack-swift | 09:06 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 09:30 |
| acoles | good morning | 09:47 |
| m_kazuhiro_ | acoles: good morning | 09:57 |
| acoles | m_kazuhiro_: o/ | 09:58 |
| *** yashmurty has quit IRC | 10:00 | |
| *** pcaruana has quit IRC | 10:05 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: Refactor expirer's task round robin implementation https://review.openstack.org/538171 | 10:09 |
| m_kazuhiro_ | mattoliverau: kota_ : I updated patch 538171. Please check it :) | 10:11 |
| patchbot | https://review.openstack.org/#/c/538171/ - swift - Refactor expirer's task round robin implementation | 10:11 |
| *** ^andrea^ has joined #openstack-swift | 10:15 | |
| acoles | mattoliverau: timburke torgomatic I added a few notes from our discussion to etherpad https://etherpad.openstack.org/p/deep-containers-replication under "Notes from call 13th Feb 2018 2230 UTC" | 10:17 |
| acoles | kota_: cc ^ | 10:17 |
| *** pcaruana has joined #openstack-swift | 10:20 | |
| acoles | mattoliverau: on master > https://review.openstack.org/542513 | 10:21 |
| patchbot | patch 542513 - swift - kill orphans during probe test setup | 10:21 |
| acoles | sorry I should have noted that on the feature/deep review | 10:21 |
| mattoliverau | acoles: oh cool! I'll take a look on the morning :) | 10:28 |
| mattoliverau | ^ both the probe test patch and the etherpad :) | 10:28 |
| mattoliverau | kota_: no worries about patch 538171, if you can make it better and easier to read that's awesome! That's why more eyes make us a better team :) | 10:30 |
| patchbot | https://review.openstack.org/#/c/538171/ - swift - Refactor expirer's task round robin implementation | 10:30 |
| *** armaan has joined #openstack-swift | 10:34 | |
| *** m_kazuhiro has joined #openstack-swift | 10:37 | |
| m_kazuhiro | acoles: I'm making and merging preparation patches for expirer's general task queue patch. There are seven preparation patches and they are listed in https://etherpad.openstack.org/p/swift_general_task_queue . Two of them are already merged. I want to merge remaining patches BEFORE Dublin PTG. If you have time, please review them. | 10:41 |
| *** m_kazuhiro_ has quit IRC | 10:41 | |
| acoles | m_kazuhiro: ok, I'll try to look at them | 10:41 |
| m_kazuhiro | acoles: Thank you! | 10:42 |
| *** m_kazuhiro has quit IRC | 10:49 | |
| *** yashmurty has joined #openstack-swift | 11:07 | |
| *** yashmurty has quit IRC | 11:12 | |
| *** pcaruana has quit IRC | 11:30 | |
| *** pcaruana has joined #openstack-swift | 11:42 | |
| *** armaan has quit IRC | 12:05 | |
| *** armaan has joined #openstack-swift | 12:05 | |
| *** cshastri_ has joined #openstack-swift | 12:08 | |
| *** cshastri has quit IRC | 12:11 | |
| *** gkadam has joined #openstack-swift | 12:17 | |
| *** gkadam__ has quit IRC | 12:19 | |
| *** armaan has quit IRC | 12:27 | |
| *** gkadam has quit IRC | 12:38 | |
| *** joeljwright has joined #openstack-swift | 12:54 | |
| *** ChanServ sets mode: +v joeljwright | 12:54 | |
| *** mvk_ has quit IRC | 12:58 | |
| openstackgerrit | Kazuhiro MIYAHARA proposed openstack/swift master: Refactor expirer's task round robin implementation https://review.openstack.org/538171 | 12:58 |
| *** armaan has joined #openstack-swift | 13:00 | |
| *** ianychoi has quit IRC | 13:00 | |
| *** ^andrea^ has quit IRC | 13:01 | |
| *** yashmurty has joined #openstack-swift | 13:01 | |
| *** ianychoi has joined #openstack-swift | 13:04 | |
| *** yashmurty has quit IRC | 13:05 | |
| *** ianychoi has quit IRC | 13:09 | |
| *** mvk_ has joined #openstack-swift | 13:24 | |
| *** cshastri_ has quit IRC | 13:53 | |
| *** armaan has quit IRC | 13:57 | |
| *** armaan has joined #openstack-swift | 13:58 | |
| *** armaan has quit IRC | 14:05 | |
| *** cshastri has joined #openstack-swift | 14:14 | |
| *** armaan has joined #openstack-swift | 14:15 | |
| *** links has quit IRC | 14:33 | |
| *** ianychoi has joined #openstack-swift | 14:47 | |
| *** DHE has joined #openstack-swift | 14:53 | |
| -openstackstatus- NOTICE: Due to a race in stable/queens branch creation and some job removals, Zuul has reported syntax errors for the past hour; if you saw a syntax error reported for "Job tripleo-ci-centos-7-ovb-containers-oooq not defined" you can safely recheck now | 15:05 | |
| *** armaan has quit IRC | 15:09 | |
| *** gkadam has joined #openstack-swift | 15:12 | |
| *** cshastri has quit IRC | 15:20 | |
| bob_cheesey | Hi all, i'm looking to implement rate limiting on my cluster as it's been getting quite a hammering lately - i just want to limit all operations to x requests per second. How can I go about this? set just container_ratelimit_0 = x? | 15:39 |
| *** links has joined #openstack-swift | 15:40 | |
| notmyname | bob_cheesey: yes | 15:41 |
| notmyname | bob_cheesey: but note that will only limit *write* requests (PUT, POST, DELETE) | 15:41 |
| notmyname | read requests will not be limited | 15:41 |
| bob_cheesey | notmyname: it's the PUTting of tiny files which is the issue, so that is exactly what I'm after | 15:42 |
| bob_cheesey | thanks! | 15:42 |
| notmyname | oh. yikes. yeah writing a lot of tiny files is always a challenging issue | 15:42 |
| notmyname | lots of socket management, lots of internal overhead | 15:43 |
| bob_cheesey | yeah, it's not quite caused an outage yet but it's hit our main cluster rather hard | 15:43 |
| bob_cheesey | the joys of running a public cloud storage offering | 15:43 |
| *** itlinux has joined #openstack-swift | 15:50 | |
| *** gkadam_ has joined #openstack-swift | 15:52 | |
| mrjk_ | Hi notmyname, I disabled the concurrency for the container-updater, and things seems to work well now. So I guess I should report a bug ? | 15:56 |
| notmyname | bob_cheesey: where are you seeing the impact? async pendings piling up? proxy cpu busy with network management? | 15:56 |
| notmyname | mrjk_: you had a hung process right? | 15:57 |
| mrjk_ | Yep | 15:57 |
| mrjk_ | I've looked up into the bug tracker, I can't find anything related | 15:57 |
| notmyname | bob_cheesey: container DBs using up all your drive space? | 15:57 |
| notmyname | mrjk_: sure. if you file a bug, we'll take a look and see if we can hunt down the source. it's really helpful if you can provide steps to reproduce it | 15:58 |
| bob_cheesey | notmyname: async pendings mostly, slow transactions in general | 16:02 |
| *** cbartz has quit IRC | 16:03 | |
| notmyname | bob_cheesey: ok. that's what I'd expect (and have seen myself) as the first issues. | 16:04 |
| *** gkadam has quit IRC | 16:04 | |
| notmyname | bob_cheesey: if you're running >=2.4.0 you can set container_update_timeout on the object server to something lower than the default value of 1 | 16:05 |
| notmyname | lowering the value will allow the object serve to return to the client more quickly, but it may cause more async pendings | 16:05 |
| notmyname | it lowers the impact to clients by queueing more internal cluster work for later. clients get better, ops gets worse | 16:07 |
| notmyname | (and if you're not on at least 2.4.0 yet then upgrade! ;-) ) | 16:07 |
| bob_cheesey | notmyname: slightly older version - upgrade is on the cards at some point though | 16:08 |
| bob_cheesey | made a note of that var so i can set it twhen nwe get there | 16:08 |
| bob_cheesey | *when we | 16:08 |
| notmyname | yeah, with that setting, the client impact of big containers and lots of writes has been removed (or hidden) | 16:09 |
| notmyname | 2.4.0 is 2.5 years old now. it was released september 1, 2015 | 16:10 |
| notmyname | there is a *lot* of good stuff you'll get by upgrading to the latest (2.17.0) | 16:10 |
| notmyname | and remember you can upgrade directly without any client downtime! do it today! ;-) | 16:11 |
| *** pcaruana has quit IRC | 16:13 | |
| bob_cheesey | yeah i took the management of this cluster over - it's been in need of some loving! | 16:14 |
| notmyname | today is valentine's day! no better day to show your swift cluster some love. give it a present by upgrading it today! | 16:15 |
| tdasilva | lol | 16:15 |
| bob_cheesey | it was hometime 15 minutes ago so that's a job for tomorrow | 16:15 |
| bob_cheesey | :p | 16:15 |
| bob_cheesey | my 'better' half might not like me staying late tonight h;) | 16:15 |
| notmyname | :-) | 16:15 |
| notmyname | bob_cheesey: please drop in when you start that. we'd love to hear your play-by-play as you upgrade | 16:16 |
| notmyname | bob_cheesey: also, if you join us in dublin for the PTG in a couple of weeks, that would be awesome. we love to have swift admins around | 16:17 |
| bob_cheesey | notmyname: i wasn't aware it was dublin this time around, i'll have to see if i can swing it with management | 16:19 |
| bob_cheesey | cheaper than going to vancouver in may :D | 16:19 |
| notmyname | https://www.openstack.org/ptg | 16:19 |
| notmyname | as a sysadmin or dev, IMO the PTG is vastly better than the summit | 16:19 |
| notmyname | I'd recommend going to the PTG over the summit | 16:19 |
| bob_cheesey | yeah i've done a couple of summits, they were useful at the outset but less so now | 16:20 |
| bob_cheesey | still, if the company wants to send me who am i to complain..? | 16:21 |
| notmyname | heh | 16:21 |
| notmyname | morning commute time for me (and a typical wednesday full of meetings). I'll be back online later | 16:21 |
| bob_cheesey | enjoy! | 16:21 |
| bob_cheesey | thanks for the help, much appreciated | 16:21 |
| mrjk_ | I hope I did well :) https://bugs.launchpad.net/swift/+bug/1568650 | 16:22 |
| openstack | Launchpad bug 1568650 in OpenStack Object Storage (swift) "Connection between client and proxy service does not closes" [High,Confirmed] - Assigned to drax (devesh-gupta) | 16:22 |
| *** armaan has joined #openstack-swift | 16:23 | |
| *** psachin has quit IRC | 16:29 | |
| *** hseipp has joined #openstack-swift | 16:33 | |
| *** itlinux has quit IRC | 16:41 | |
| *** itlinux has joined #openstack-swift | 16:51 | |
| *** yashmurty has joined #openstack-swift | 17:01 | |
| *** _alastor__ has joined #openstack-swift | 17:06 | |
| *** yashmurty has quit IRC | 17:06 | |
| *** geaaru has quit IRC | 17:11 | |
| *** gyee has joined #openstack-swift | 17:11 | |
| *** pcaruana has joined #openstack-swift | 17:17 | |
| *** links has quit IRC | 17:21 | |
| *** hseipp has quit IRC | 17:28 | |
| *** hoonetorg has quit IRC | 18:05 | |
| *** tosky has joined #openstack-swift | 18:07 | |
| tosky | hi! After some discussion about the optional dependencies of Swift, yesterday notmyname kindly proposed https://review.openstack.org/#/c/544121/ (now merged) | 18:08 |
| patchbot | patch 544121 - swift - add keystonemiddleware to extras area (MERGED) | 18:08 |
| *** mvk_ has quit IRC | 18:08 | |
| notmyname | tdasilva: cschwede: mattoliverau: kota_: ^ you may have answers to something tosky is about to ask | 18:09 |
| notmyname | tosky: note, most of those people are asleep right now ;-) | 18:09 |
| tosky | my use case is run as many services as possible with python3, with the exception of swift, and that's where the optional dependencies are important | 18:09 |
| tosky | uh, I will leave it here anyway :) | 18:09 |
| notmyname | yeah, they've all got irc bouncers | 18:09 |
| tosky | so, I tried to use the extra dependencies (https://review.openstack.org/#/c/544383/) and now I get a different error in my experimental job | 18:10 |
| patchbot | patch 544383 - openstack-dev/devstack - WIP Install extra dependencies for Swift | 18:10 |
| tosky | well, tdasilva is not sleeping for sure :) | 18:10 |
| *** hseipp has joined #openstack-swift | 18:10 | |
| tosky | previously it was failing on a missing keystonemiddleware, which is now installed thanks to the extra used by my patch above | 18:10 |
| tosky | but what I get now is: ImportError: No module named ceilometermiddleware.swift | 18:11 |
| tosky | see the logs on http://logs.openstack.org/19/304019/25/experimental/sahara-tests-scenario-py3/b186232/controller/logs/screen-s-proxy.txt.gz#_Feb_14_16_45_24_210470 | 18:11 |
| tosky | the question are: should ceilimetermiddleware be added to the extra as well? To another extra? Extension of the same extra? Other ideas? | 18:12 |
| * tosky breaths | 18:12 | |
| tosky | breathes, even | 18:12 |
| tosky | (or should I just disable ceilometer? But still...) | 18:18 |
| *** hoonetorg has joined #openstack-swift | 18:22 | |
| *** hseipp has quit IRC | 18:22 | |
| notmyname | tosky: that's a question I have. why do you need ceilometer? | 18:23 |
| timburke | how is ceilometermiddleware getting into the pipeline? shouldn't whatever's doing that also install ceilometermiddleware? | 18:24 |
| notmyname | yeah. timburke has a much better question | 18:25 |
| *** psachin has joined #openstack-swift | 18:26 | |
| tosky | notmyname: it's in the default set of services | 18:26 |
| timburke | i guess... it's probably installing it in the wrong venv? can we just have it install in both? | 18:26 |
| tosky | it's lib/swift that tries to configure ceilometermiddleware when ceilometer is enabled | 18:27 |
| tosky | namely here: http://git.openstack.org/cgit/openstack-dev/devstack/tree/lib/swift#n411 | 18:27 |
| tosky | so, I could disable ceilometer in my specific job; I just wondering if there is a case where someone else may be interested in having it running in my same scenario (almost all services running with python3) | 18:29 |
| timburke | can we have that block do the install? | 18:29 |
| tosky | uh, I don't know, I mean, technically it's code; it's a question for the devstack maintainers | 18:30 |
| tosky | I can at least test it extending my patch | 18:30 |
| tosky | but uhm, not sure if I would need to do some cloning too, not just installation | 18:31 |
| timburke | or maybe we could add a `if is_service_enabled ceilometer` block to the install_swift function | 18:31 |
| timburke | mmm... true... | 18:31 |
| timburke | hey, there's even a install_ceilometermiddleware function... | 18:32 |
| timburke | ...and nothing in that file calls it :-/ | 18:33 |
| *** openstackgerrit has quit IRC | 18:33 | |
| tosky | timburke: so maybe calling that function inside the if could be enough? I can try if I think it's worth to check | 18:40 |
| tosky | even if that function is called already, uhm | 18:41 |
| tosky | inside stack.sh - but maybe it installs the py3 version? | 18:41 |
| DHE | is python 2.6 supported? asking because I'm trying on a spare centos6 system and requirements building is failing miserably | 18:49 |
| tosky | confirmed: the call to install_ceilometermiddleware in stack.sh uses python 3 | 18:51 |
| tdasilva | tosky: hello, goood morning, i just woke up ;) | 18:53 |
| tosky | :D | 18:53 |
| *** tesseract has quit IRC | 18:54 | |
| tosky | so probably the cleanest way is to use extras, so that the dependency is installed using the same pip call and the correct python version | 18:55 |
| tosky | but that would not gate against the version from git | 18:55 |
| tosky | argh | 18:55 |
| tdasilva | i guess the challenge is: everytime there's a new swift 3rd party middleware that uses some depedency, should it be added to [extras] in swift? | 18:55 |
| *** armaan has quit IRC | 18:57 | |
| *** armaan has joined #openstack-swift | 18:58 | |
| notmyname | DHE: I think we broke py26 compat a while back | 19:01 |
| notmyname | tdasilva: that doesn't seem quite right | 19:01 |
| tdasilva | y | 19:02 |
| notmyname | because swift shouldn't be responsible for defining 3rd party middleware dependencies | 19:02 |
| tosky | and it may break testing chained patches (if keystonemidleware is installed through pip, a switch change can't be gated against a keystonemiddleware; but I may misunderstood how it works) | 19:03 |
| tdasilva | heh, that 'y' was for a 'dnf install' but works i guess... | 19:03 |
| notmyname | lol | 19:03 |
| notmyname | didn't know if that was "why" or "yes" | 19:03 |
| tdasilva | notmyname: yeah, i agree | 19:06 |
| notmyname | however, "other openstack projects" may be a special case | 19:06 |
| notmyname | for example, keystone | 19:06 |
| notmyname | FYI I just came across https://github.com/osallou/goswift. looks like another swift web client. I havne't run it yet | 19:10 |
| *** geaaru has joined #openstack-swift | 19:25 | |
| *** zaitcev has joined #openstack-swift | 19:26 | |
| *** ChanServ sets mode: +v zaitcev | 19:26 | |
| *** hseipp has joined #openstack-swift | 19:27 | |
| notmyname | rledisez: onovy: martin is adding some stuff to the PTG etherpad that are very ops/support/admin focused | 19:46 |
| *** ChanServ sets mode: +o notmyname | 19:47 | |
| *** notmyname changes topic to "https://bugs.not.mn/project/Swift | Logs: http://eavesdrop.openstack.org/irclogs/%23openstack-swift/ | Meetings: https://wiki.openstack.org/wiki/Meetings/Swift | Review Dashboard: http://not.mn/reviews.html | https://etherpad.openstack.org/p/Dublin_PTG_Swift" | 19:47 | |
| *** notmyname sets mode: -o notmyname | 19:48 | |
| *** hseipp has quit IRC | 19:50 | |
| *** openstackgerrit has joined #openstack-swift | 20:32 | |
| openstackgerrit | Gerard Gine proposed openstack/swift master: Improved usage of args in .functests https://review.openstack.org/544142 | 20:32 |
| *** ukaynar has joined #openstack-swift | 20:46 | |
| DHE | so crazy question. how large do the account and container ring/storage get? would it be sane to store them on smaller but faster systems than the object storage systems? eg: on a system with dedicated proxies, store them on the proxies themselves? | 20:51 |
| DHE | I'm preparing to build a lab, don't have anything working yet so I have nothing to go by. | 20:52 |
| clayg | yeah PAC+O is a great configuration - SSD's for A/C disks is recommend | 20:54 |
| *** m_kazuhiro has joined #openstack-swift | 20:57 | |
| kota_ | good morning | 20:57 |
| mattoliverau | Morning | 20:57 |
| m_kazuhiro | kota_: mattoliverau: morning | 20:57 |
| kota_ | m_kazuhiro: o/ | 20:58 |
| *** joeljwright_ has joined #openstack-swift | 21:00 | |
| *** ChanServ sets mode: +v joeljwright_ | 21:00 | |
| kota_ | notmyname: meeting time? | 21:01 |
| notmyname | yes. | 21:01 |
| notmyname | sorry.w as just on the phone | 21:01 |
| notmyname | meeting time! | 21:01 |
| *** yashmurty has joined #openstack-swift | 21:02 | |
| *** yashmurty has quit IRC | 21:07 | |
| *** psachin has quit IRC | 21:12 | |
| notmyname | m_kazuhiro: kota_: mattoliverau: now go get your early coffee/breakfast :-) | 21:19 |
| clayg | oops - missed the meeting | 21:20 |
| clayg | was it GREAT?! | 21:20 |
| *** joeljwright_ has quit IRC | 21:20 | |
| mattoliverau | \o/ early mark! | 21:20 |
| acoles | clayg: only one thing could have made it better | 21:20 |
| mattoliverau | Pizza? | 21:20 |
| m_kazuhiro | notmyname: ok! :) | 21:20 |
| acoles | mattoliverau: clayg !! | 21:20 |
| acoles | jk. pizza was the better answer | 21:21 |
| mattoliverau | Oh to have clayg there.. of course. That would have made it infinitely better! :) | 21:21 |
| clayg | +2 🍕 | 21:21 |
| *** m_kazuhiro has quit IRC | 21:25 | |
| *** pcaruana has quit IRC | 21:28 | |
| *** geaaru has quit IRC | 21:32 | |
| *** mvk_ has joined #openstack-swift | 21:38 | |
| *** rcernin has joined #openstack-swift | 22:00 | |
| *** threestrands has joined #openstack-swift | 22:31 | |
| *** itlinux has quit IRC | 22:53 | |
| *** ukaynar has quit IRC | 22:55 | |
| *** ukaynar has joined #openstack-swift | 22:55 | |
| *** ukaynar has quit IRC | 23:00 | |
| *** kei_yama has joined #openstack-swift | 23:32 | |
| timburke | mattoliverau: you might like https://review.openstack.org/#/c/542514/ too -- or i could just +A since there's already 2 +2s :-) | 23:44 |
| patchbot | patch 542514 - swift - probetests: Close our handle to subprocess' stdouts | 23:44 |
| mattoliverau | ahh nice, I've seen myself run out of file handles while playing with probe tests.Sweet. | 23:50 |
| openstackgerrit | Matthew Oliver proposed openstack/swift master: Add a basic partition diffing tool https://review.openstack.org/539466 | 23:52 |
| timburke | we still seem to leak some, but it's *so much better* | 23:52 |
| openstackgerrit | Matthew Oliver proposed openstack/swift master: Add a basic partition diffing tool https://review.openstack.org/539466 | 23:57 |
| mattoliverau | ^ that last one was a commit message update now that the the cli output chaged a little, I manually added and made a typo.. opps. | 23:58 |
| mattoliverau | timburke: just waiting on the probetests to pass locally :) | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!