| *** irclogbot_3 has quit IRC | 01:44 | |
| openstackgerrit | JoeYang proposed openstack/swift master: Change partition name from string to integer https://review.openstack.org/649240 | 01:45 |
|---|---|---|
| *** tokuzfunpi has joined #openstack-swift | 01:49 | |
| hugokuo | tokuzfunpi: ping | 03:14 |
| *** psachin has joined #openstack-swift | 03:20 | |
| *** psachin has quit IRC | 03:51 | |
| kota_ | hello world | 04:34 |
| kota_ | oh, hugokuo | 04:34 |
| hugokuo | @kota_ been a long time! | 04:42 |
| hugokuo | kota_: | 04:42 |
| kota_ | ¥o/ | 04:42 |
| *** pcaruana has joined #openstack-swift | 04:49 | |
| *** pcaruana has quit IRC | 04:55 | |
| *** e0ne has joined #openstack-swift | 05:31 | |
| *** e0ne has quit IRC | 05:39 | |
| *** e0ne has joined #openstack-swift | 05:56 | |
| *** e0ne has quit IRC | 06:04 | |
| *** pcaruana has joined #openstack-swift | 06:30 | |
| *** rdejoux has joined #openstack-swift | 07:14 | |
| *** e0ne has joined #openstack-swift | 07:24 | |
| *** e0ne has quit IRC | 07:52 | |
| *** mikecmpbll has joined #openstack-swift | 07:52 | |
| *** mikecmpbll has quit IRC | 07:55 | |
| *** tkajinam has quit IRC | 08:11 | |
| *** rcernin has quit IRC | 08:19 | |
| *** ccamacho has joined #openstack-swift | 08:28 | |
| *** e0ne has joined #openstack-swift | 08:37 | |
| *** zigo_ has joined #openstack-swift | 08:48 | |
| *** zigo_ is now known as zigo | 08:50 | |
| *** e0ne has quit IRC | 09:27 | |
| *** e0ne has joined #openstack-swift | 09:34 | |
| *** e0ne has quit IRC | 10:06 | |
| *** tokuzfunpi has quit IRC | 10:13 | |
| *** e0ne has joined #openstack-swift | 10:13 | |
| *** tokuzfunpi has joined #openstack-swift | 10:14 | |
| *** tokuzfunpi has quit IRC | 10:20 | |
| *** tokuzfunpi has joined #openstack-swift | 10:36 | |
| *** tokuzfunpi has quit IRC | 10:39 | |
| *** e0ne has quit IRC | 11:02 | |
| *** e0ne has joined #openstack-swift | 11:09 | |
| *** mvkr has quit IRC | 11:34 | |
| *** e0ne has quit IRC | 11:46 | |
| *** pcaruana has quit IRC | 11:47 | |
| *** e0ne has joined #openstack-swift | 11:51 | |
| *** e0ne has quit IRC | 11:53 | |
| *** mvkr has joined #openstack-swift | 12:29 | |
| *** pcaruana has joined #openstack-swift | 12:38 | |
| *** mrjk has quit IRC | 12:41 | |
| *** zaitcev has joined #openstack-swift | 12:49 | |
| *** ChanServ sets mode: +v zaitcev | 12:49 | |
| *** mrjk has joined #openstack-swift | 13:28 | |
| *** pcaruana has quit IRC | 13:31 | |
| *** pcaruana has joined #openstack-swift | 13:36 | |
| *** spotz has joined #openstack-swift | 14:27 | |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Install QuadIron for unit tests https://review.openstack.org/650483 | 14:51 |
| *** nicodemus_ has joined #openstack-swift | 14:53 | |
| nicodemus_ | hello | 14:53 |
| nicodemus_ | quick question... I have a three node swift cluster, and need to shut down one of the nodes for two hours for maintenance | 14:53 |
| nicodemus_ | the node will come back online with the same disks and data, is it rather safe to do so? | 14:54 |
| *** e0ne has joined #openstack-swift | 15:14 | |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Install QuadIron for unit tests https://review.openstack.org/650483 | 15:16 |
| *** ybunker has joined #openstack-swift | 15:17 | |
| *** gyee has joined #openstack-swift | 15:20 | |
| *** itlinux_ has quit IRC | 15:33 | |
| *** e0ne has quit IRC | 16:02 | |
| *** ybunker has quit IRC | 16:02 | |
| *** e0ne has joined #openstack-swift | 16:05 | |
| *** ccamacho has quit IRC | 16:11 | |
| *** e0ne has quit IRC | 16:17 | |
| *** itlinux has joined #openstack-swift | 16:28 | |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Split test and valgrind-test into separate jobs https://review.openstack.org/650952 | 16:28 |
| timburke | nicodemus_, yeah, that all sounds fine. when the node comes back up, the background daemons should ensure that it gets the latest updates from the other two | 16:32 |
| nicodemus_ | Thanks timburke ! | 16:32 |
| timburke | it may serve some old data in the mean time, but since swift is eventually consistent, applications should already be built with that possibility in mind | 16:33 |
| timburke | if the node was going to be offline for a longer period of time (say, a week or more) you would probably want to worry more -- most of the reclaim ages default to a week, which means that when the node comes back, it might not hear about some of the deletes that happened while it was offline | 16:34 |
| nicodemus_ | Yeah, I was reading in the docs that for long periods I should just go ahead and remove it from the ring | 16:35 |
| timburke | ...which leads to what we call "dark data" -- data that doesn't appear in listings but continues to occupy disk space and get replicated across the cluster | 16:35 |
| timburke | yup. then format and bring it back in as a fresh node | 16:36 |
| nicodemus_ | we just need to boot it in single-user mode to recover the OS password, perhaps it's less than two hours | 16:37 |
| nicodemus_ | many thanks for the info ! | 16:37 |
| timburke | (alternatively, you could bump up the reclaim ages to something like a month, so those tombstone records will still be around when it finally comes back online) | 16:37 |
| timburke | nicodemus_, yeah, that should all be fine. good luck! | 16:37 |
| *** e0ne has joined #openstack-swift | 16:46 | |
| *** e0ne has quit IRC | 17:16 | |
| *** NM has joined #openstack-swift | 17:29 | |
| *** ybunker has joined #openstack-swift | 17:53 | |
| *** rdejoux has quit IRC | 18:08 | |
| *** itlinux has quit IRC | 18:19 | |
| *** clayg has quit IRC | 18:20 | |
| *** clayg has joined #openstack-swift | 18:20 | |
| *** ChanServ sets mode: +v clayg | 18:20 | |
| *** fyx has quit IRC | 18:20 | |
| *** fyx has joined #openstack-swift | 18:21 | |
| notmyname | good morning | 18:21 |
| *** e0ne has joined #openstack-swift | 18:33 | |
| *** mvkr has quit IRC | 18:51 | |
| *** e0ne has quit IRC | 18:55 | |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Use nasm instead of yasm https://review.openstack.org/650975 | 19:44 |
| *** ybunker has quit IRC | 19:53 | |
| timburke | tdasilva, i forget -- did you add the experimental centos 7 jobs for liberasure code? you at least did the change to install jerasure/isa-l right? | 19:59 |
| timburke | what do you think about the two options i just left on https://review.openstack.org/#/c/650975/? as it is, it seems our experimental job is busted there... | 20:00 |
| patchbot | patch 650975 - liberasurecode - Use nasm instead of yasm - 1 patch set | 20:00 |
| tdasilva | timburke: mm..that's weird. I think I would have tested that on a centos vm :( | 20:02 |
| tdasilva | pretty sure I added the jerasure/isa-l install and I'd guess I added the experimental centos job too, but that would have been done much earlier...jerasure/isa-l install is more recent IIRC... | 20:03 |
| timburke | fwiw, on my own centos box, i think i just installed nasm from source... | 20:03 |
| tdasilva | timburke: i think yasm in centos 7 is in epel | 20:04 |
| timburke | that'd be another way :-) | 20:05 |
| tdasilva | https://review.openstack.org/#/c/604391/ never ran experimental :( | 20:08 |
| patchbot | patch 604391 - liberasurecode - Install Jerasure and ISA-L libs (MERGED) - 5 patch sets | 20:08 |
| tdasilva | altought you can see in my notes that I was testing on a centos7 vm...most likely I had epel installed and never realized there was an issue there... | 20:09 |
| timburke | hmm... speaking of, i wonder at what point https://github.com/openstack/swift/blob/2.21.0/tools/test-setup.sh#L28 is going to rust out from under us... | 20:13 |
| tdasilva | oh yeah, we should update that... | 20:14 |
| tdasilva | unfortunately there's no 'openstack-latest' repo, so i'm not sure how we can do a better job of installing libec, unless we build it from source... | 20:15 |
| timburke | meh. as long as we do it reasonably often, i suppose... | 20:15 |
| tdasilva | yeah, and I think zaitcev might have not packaged 1.6.0 yet...maybe... :) | 20:16 |
| timburke | i guess from source might be good to, but i'm not *so* worried about it -- libec doesn't change that often, and when it does, i'm much more worried about breaking shss or phazr or something than the builtin liberasurecode_rs_vand implementation that we use in our tests | 20:17 |
| *** pcaruana has quit IRC | 20:25 | |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Install EPEL repo on CentOS https://review.openstack.org/650975 | 20:33 |
| timburke | we'll see how that does... | 20:33 |
| openstackgerrit | Thiago da Silva proposed openstack/swift master: Update openstack repo for libec install https://review.openstack.org/650987 | 20:50 |
| *** nicodemus_ has quit IRC | 20:54 | |
| timburke | i wonder if we can remove our hack from https://github.com/openstack/swift/commit/67de0c8#diff-67e43d32a534c1f6d2e1ca157dc8a1a0 -- looks like upstream fixed this in 2.7, as https://github.com/python/cpython/commit/8c6d9d7 and i guess we needed it to support 2.6? | 20:59 |
| *** NM has quit IRC | 21:30 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Remove our urlparse wrapper https://review.openstack.org/651004 | 21:53 |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Install EPEL repo on CentOS https://review.openstack.org/650975 | 22:07 |
| *** rcernin has joined #openstack-swift | 22:14 | |
| *** NM has joined #openstack-swift | 22:29 | |
| *** persia has joined #openstack-swift | 22:31 | |
| tdasilva | what?? http://logs.openstack.org/75/650975/3/check/liberasurecode-unittests-centos-7/7bbe06f/job-output.txt.gz#_2019-04-08_22_41_01_318568 | 23:00 |
| *** tkajinam has joined #openstack-swift | 23:00 | |
| timburke | tdasilva, i probably need to do a makecache | 23:04 |
| *** NM has quit IRC | 23:05 | |
| openstackgerrit | Tim Burke proposed openstack/liberasurecode master: Install EPEL repo on CentOS https://review.openstack.org/650975 | 23:06 |
| mattoliverau | morning | 23:50 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!