Wednesday, 2020-10-07

*** hemna has quit IRC00:08
*** hemna has joined #openstack-meeting00:09
*** rh-jlabarre has quit IRC00:21
*** hemna_ has joined #openstack-meeting00:30
*** hemna has quit IRC00:30
*** moguimar has quit IRC01:04
*** moguimar has joined #openstack-meeting01:06
*** armax has quit IRC01:25
*** moguimar has quit IRC01:28
*** moguimar has joined #openstack-meeting01:29
*** hemna_ has quit IRC01:36
*** hemna has joined #openstack-meeting01:37
*** hemna has quit IRC02:11
*** armax has joined #openstack-meeting02:13
*** hemna has joined #openstack-meeting02:18
*** macz_ has joined #openstack-meeting02:34
*** macz_ has quit IRC02:38
*** rcernin has quit IRC02:45
*** rcernin has joined #openstack-meeting02:48
*** gyee has quit IRC02:56
*** rcernin has quit IRC03:01
*** rcernin has joined #openstack-meeting03:17
*** rcernin has quit IRC03:22
*** rcernin has joined #openstack-meeting03:23
*** psachin has joined #openstack-meeting03:41
*** vishalmanchanda has joined #openstack-meeting03:43
*** armax has quit IRC04:01
*** macz_ has joined #openstack-meeting04:22
*** yasufum has joined #openstack-meeting04:23
*** macz_ has quit IRC04:26
*** evrardjp has quit IRC04:33
*** evrardjp has joined #openstack-meeting04:33
*** ricolin_ has joined #openstack-meeting05:15
*** ricolin_ has quit IRC05:26
*** yasufum has quit IRC06:31
*** icey has quit IRC06:39
*** icey has joined #openstack-meeting06:41
*** icey has quit IRC06:45
*** icey has joined #openstack-meeting06:48
*** ralonsoh has joined #openstack-meeting06:54
*** vishalmanchanda has quit IRC07:02
*** icey has quit IRC07:04
*** slaweq has joined #openstack-meeting07:05
*** e0ne has joined #openstack-meeting07:07
*** slaweq has quit IRC07:09
*** haleyb has quit IRC07:10
*** haleyb has joined #openstack-meeting07:11
*** vishalmanchanda has joined #openstack-meeting07:11
*** slaweq has joined #openstack-meeting07:14
*** icey has joined #openstack-meeting07:18
*** gagehugo has quit IRC07:20
*** gagehugo has joined #openstack-meeting07:20
*** njohnston has quit IRC07:33
*** rcernin has quit IRC07:39
*** tosky has joined #openstack-meeting07:41
*** rpittau|afk is now known as rpittau07:41
*** ykatabam has quit IRC07:49
*** icey has quit IRC07:57
*** icey has joined #openstack-meeting08:00
*** tosky_ has joined #openstack-meeting08:12
*** tosky is now known as Guest9816008:13
*** tosky_ is now known as tosky08:13
*** Guest98160 has quit IRC08:15
*** ykatabam has joined #openstack-meeting08:30
*** hyunsikyang has joined #openstack-meeting08:31
*** hyunsikyang__ has quit IRC08:34
*** hyunsikyang__ has joined #openstack-meeting08:35
*** hyunsikyang has quit IRC08:37
*** hyunsikyang has joined #openstack-meeting08:42
*** Lucas_Gray has joined #openstack-meeting08:43
*** hyunsikyang__ has quit IRC08:45
*** Lucas_Gray has quit IRC08:49
*** Lucas_Gray has joined #openstack-meeting08:51
*** manpreet has joined #openstack-meeting08:55
*** Lucas_Gray has quit IRC09:06
*** Lucas_Gray has joined #openstack-meeting09:07
*** lpetrut has joined #openstack-meeting09:13
*** ykatabam has quit IRC09:13
*** hemna has quit IRC09:19
*** hemna has joined #openstack-meeting09:21
*** ociuhandu has joined #openstack-meeting09:24
*** ociuhandu has quit IRC09:29
*** Lucas_Gray has quit IRC09:43
*** yasufum has joined #openstack-meeting09:43
*** ociuhandu has joined #openstack-meeting09:56
*** Lucas_Gray has joined #openstack-meeting09:56
*** ykatabam has joined #openstack-meeting10:19
*** rcernin has joined #openstack-meeting10:21
*** yasufum has quit IRC10:24
*** Lucas_Gray has quit IRC10:34
*** Lucas_Gray has joined #openstack-meeting10:37
*** ykatabam has quit IRC10:37
*** rcernin has quit IRC10:37
*** oneswig has joined #openstack-meeting10:51
*** oneswig has quit IRC10:59
*** oneswig_ has joined #openstack-meeting10:59
oneswig_#startmeeting scientific-sig11:00
openstackMeeting started Wed Oct  7 11:00:21 2020 UTC and is due to finish in 60 minutes.  The chair is oneswig_. Information about MeetBot at http://wiki.debian.org/MeetBot.11:00
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.11:00
*** openstack changes topic to " (Meeting topic: scientific-sig)"11:00
openstackThe meeting name has been set to 'scientific_sig'11:00
oneswig_hello o/11:01
*** oneswig_ has quit IRC11:02
*** raildo has joined #openstack-meeting11:04
*** Lucas_Gray has quit IRC11:07
*** zerozephyrum has joined #openstack-meeting11:09
*** Lucas_Gray has joined #openstack-meeting11:12
*** zerozephyrum has quit IRC11:18
*** belmoreira has joined #openstack-meeting11:22
*** njohnston_ has joined #openstack-meeting11:25
*** Wryhder has joined #openstack-meeting11:46
*** Lucas_Gray has quit IRC11:46
*** Wryhder is now known as Lucas_Gray11:47
*** masazumi-ota has quit IRC11:51
*** psahoo has joined #openstack-meeting11:52
*** rfolco has joined #openstack-meeting12:11
*** eharney_ is now known as eharney12:27
*** rf0lc0 has joined #openstack-meeting13:03
*** raildo_ has joined #openstack-meeting13:03
*** rh-jlabarre has joined #openstack-meeting13:03
*** rfolco has quit IRC13:03
*** raildo has quit IRC13:03
*** manpreet has quit IRC13:03
*** vishalmanchanda has quit IRC13:03
*** evrardjp has quit IRC13:04
*** moguimar has quit IRC13:04
*** jmasud has quit IRC13:04
*** jamesdenton has quit IRC13:04
*** ttx has quit IRC13:04
*** patrickeast has quit IRC13:04
*** mattoliverau has quit IRC13:04
*** freefood has quit IRC13:04
*** Lucas_Gray has quit IRC13:04
*** manpreet has joined #openstack-meeting13:04
*** vishalmanchanda has joined #openstack-meeting13:04
*** evrardjp has joined #openstack-meeting13:04
*** jmasud has joined #openstack-meeting13:07
*** moguimar has joined #openstack-meeting13:09
*** jamesdenton has joined #openstack-meeting13:09
*** patrickeast has joined #openstack-meeting13:09
*** mattoliverau has joined #openstack-meeting13:09
*** freefood has joined #openstack-meeting13:09
*** psahoo_ has joined #openstack-meeting13:09
*** psahoo has quit IRC13:12
*** ttx has joined #openstack-meeting13:17
*** Adri2000 has joined #openstack-meeting13:25
*** TrevorV has joined #openstack-meeting13:45
*** whoami-rajat___ has joined #openstack-meeting13:46
*** yasufum has joined #openstack-meeting14:01
*** thgcorrea has joined #openstack-meeting14:03
*** yasufum_ has joined #openstack-meeting14:04
*** yasufum has quit IRC14:05
*** yasufum_ is now known as yasufum14:05
*** macz_ has joined #openstack-meeting14:35
*** icey has quit IRC14:36
*** icey has joined #openstack-meeting14:36
*** macz_ has quit IRC14:39
*** armax has joined #openstack-meeting14:40
*** psahoo_ has quit IRC14:43
*** lpetrut has quit IRC14:45
*** njohnston_ is now known as njohnston14:52
*** priteau has joined #openstack-meeting14:57
*** slaweq_ has joined #openstack-meeting14:58
*** macz_ has joined #openstack-meeting14:58
*** slaweq has quit IRC14:58
*** mlavalle has joined #openstack-meeting15:05
*** yasufum has quit IRC15:16
*** yasufum has joined #openstack-meeting15:24
*** TrevorV has quit IRC15:33
*** TrevorV has joined #openstack-meeting15:34
*** Lucas_Gray has joined #openstack-meeting15:42
*** belmoreira has quit IRC15:44
*** yasufum has quit IRC15:45
*** rpittau is now known as rpittau|afk16:06
*** slaweq has joined #openstack-meeting16:06
*** slaweq_ has quit IRC16:07
*** ociuhandu has quit IRC16:11
*** dsariel has joined #openstack-meeting16:18
*** psachin has quit IRC16:19
*** tosky has quit IRC16:21
*** ociuhandu has joined #openstack-meeting16:43
*** ociuhandu has quit IRC16:49
*** Wryhder has joined #openstack-meeting16:55
*** Lucas_Gray has quit IRC16:55
*** Wryhder is now known as Lucas_Gray16:56
*** yasufum has joined #openstack-meeting17:06
*** yasufum has quit IRC17:09
*** ralonsoh has quit IRC17:12
*** ralonsoh has joined #openstack-meeting17:12
*** dsariel has quit IRC17:13
*** dsariel has joined #openstack-meeting17:14
*** macz_ has quit IRC17:14
*** macz_ has joined #openstack-meeting17:15
*** mahatic has quit IRC17:17
*** Lucas_Gray has quit IRC17:18
*** Lucas_Gray has joined #openstack-meeting17:20
*** Adri2000 has quit IRC17:20
*** e0ne has quit IRC17:23
*** Lucas_Gray has quit IRC17:25
*** Lucas_Gray has joined #openstack-meeting17:26
*** Lucas_Gray has quit IRC17:34
*** Lucas_Gray has joined #openstack-meeting17:38
*** Lucas_Gray has quit IRC17:48
*** mlavalle has quit IRC17:56
*** mlavalle has joined #openstack-meeting18:14
*** armstrong has joined #openstack-meeting18:15
*** lpetrut has joined #openstack-meeting18:18
*** vishalmanchanda has quit IRC18:22
*** lpetrut has quit IRC18:36
*** Adri2000 has joined #openstack-meeting18:43
*** priteau has quit IRC19:12
*** gyee has joined #openstack-meeting19:12
*** tosky has joined #openstack-meeting19:26
*** raildo_ is now known as raildo19:43
*** slaweq has quit IRC20:14
*** gmann is now known as gmann_lunch20:22
*** slaweq has joined #openstack-meeting20:23
*** hyunsikyang__ has joined #openstack-meeting20:30
*** ralonsoh_ has joined #openstack-meeting20:30
*** ralonsoh has quit IRC20:33
*** hyunsikyang has quit IRC20:33
*** ralonsoh_ has quit IRC20:51
*** alecuyer has joined #openstack-meeting20:58
*** slaweq has quit IRC20:58
*** TrevorV has quit IRC21:00
timburke#startmeeting swift21:00
openstacktimburke: Error: Can't start another meeting, one is in progress.  Use #endmeeting first.21:00
timburkewho's here for the swift meeting?21:00
timburke#endmeeting21:00
*** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/"21:00
openstackMeeting ended Wed Oct  7 21:00:49 2020 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)21:00
openstackMinutes:        http://eavesdrop.openstack.org/meetings/scientific_sig/2020/scientific_sig.2020-10-07-11.00.html21:00
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/scientific_sig/2020/scientific_sig.2020-10-07-11.00.txt21:00
openstackLog:            http://eavesdrop.openstack.org/meetings/scientific_sig/2020/scientific_sig.2020-10-07-11.00.log.html21:00
timburke#startmeeting swift21:01
openstackMeeting started Wed Oct  7 21:01:01 2020 UTC and is due to finish in 60 minutes.  The chair is timburke. Information about MeetBot at http://wiki.debian.org/MeetBot.21:01
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.21:01
*** openstack changes topic to " (Meeting topic: swift)"21:01
openstackThe meeting name has been set to 'swift'21:01
timburkewho's here for the swift meeting?21:01
mattoliverauo/ (mostly)21:01
rledisezo/21:01
kota_o/21:01
alecuyero/21:01
timburkeagenda's at https://wiki.openstack.org/wiki/Meetings/Swift21:01
timburke#topic TC election21:02
*** openstack changes topic to "TC election (Meeting topic: swift)"21:02
timburkejust a reminder that there's an election currently being held! vote!21:02
timburkethere are 7 candidates for (i believe) 4 seats21:03
timburke#link https://governance.openstack.org/election/21:03
timburke#topic ptg21:03
*** openstack changes topic to "ptg (Meeting topic: swift)"21:03
timburkewe're also just a couple weeks out from the virtual PTG!21:04
*** whoami-rajat___ is now known as whoami-rajat__21:04
*** gmann_lunch is now known as gmann21:04
timburkeclayg's done a good job of seeding the etherpad21:04
timburke#link https://etherpad.opendev.org/p/swift-ptg-wallaby21:04
timburkei know *i* need to add some words about ALOs21:05
timburkeand probably some other topics21:05
timburke#topic stable releases21:06
*** openstack changes topic to "stable releases (Meeting topic: swift)"21:06
*** zaitcev has joined #openstack-meeting21:06
timburkethere are a couple patches currently working their way through the gate to get changelogs for 2.25.1 and 2.23.221:07
timburkep 756166 and p 75616721:07
patchbothttps://review.opendev.org/#/c/756166/ - swift (stable/ussuri) - Authors/ChangeLog for 2.25.1 - 1 patch set21:07
patchbothttps://review.opendev.org/#/c/756167/ - swift (stable/train) - ChangeLog for 2.23.2 - 1 patch set21:07
timburkeone of the big motivations for them (i feel like) is the backported fix for the py3 crypto bug21:08
timburkejust making sure people are aware of them and some of the great fixes they include :-)21:09
timburkethat's about it for announcements; any questions or comments?21:09
zaitcevnone on my side.21:10
*** raildo has quit IRC21:11
timburkeall right, let's talk about some patches!21:11
claygo/21:11
timburke#topic replication and ring version skew21:11
*** openstack changes topic to "replication and ring version skew (Meeting topic: swift)"21:12
claygyeah, ALOs like s3 MPU!!! 👍21:12
timburke#link https://review.opendev.org/#/c/754242/21:12
patchbotpatch 754242 - swift - Fix a race condition in case of cross-replication - 5 patch sets21:12
timburkerledisez, i said i'd work on getting a dev env where i could actually repro the problem -- sorry, i haven't done that yet :-(21:12
rledisezso, as a reminder, this patch only fix the issue for ssync. i'm working on a patch for rsync but nothing to propose yet21:12
claygtests look good - I don't think I had any specific concerns left over from last week - it might be ready to go?21:13
rledisezi've testing the patch on prod where we we're seing the issue on a very regular basis. so far, so good, but i'll still monitore it closely21:13
rledisezi'd like to share this21:13
rledisez#link https://dl.plik.ovh/file/OIouMcSnLK2W2kwX/v8teIwe3Gapjjl2p/handoff-lock.png21:13
rledisezit's a rebalance of an EC policy. we can see the patch avoided the issue many time21:13
rledisezthe ring distribution started at 18:00 and took 30 minutes21:14
rledisezwhat bothers me is that even after the ring is distributed everywhere, we still get occurences of the reconstructor failing to lock the partition befor reverting it. I don't understand it21:14
timburkecool! i'm loving that visual21:15
rledisezwhat i want to try is to use a short timeout value for the reconstructor to see if it can improve perf (the rebalance without patch was taking 1h, with the patch it's 3h)21:16
rledisezexcept than that, i'm done with it i think21:16
rledisezwaiting for reviews :)21:17
timburkemaybe previous-and-still primaries are doing actual rebuilds to the new primary, so it needs to grab the lock, too?21:17
timburkethat reminds me -- clayg, we need to make sure tsync is on the etherpad, too ;-)21:17
rledisezit should be a perfect explanation, except that in our clusters we don't rebuild partition recently moved (to avoid to reconstruct something that should be moved)21:17
claygtimburke: oh... yeah...21:18
rlediseztimburke: i think you're right, it's the rebuild, but not of moving partitions, of all the other partitions. so yeah, it should be fine. i'll check that but I think you got it21:19
timburkealternatively, maybe there's new data landing in the partition, and the new primary's reconstructor locks it while checking in with neighbors? though you'd think that should be pretty fast21:20
claygtim doesn't even see the code anymore it's just 'primary partition, revert hand-off, orphaned slo segment...'21:20
timburkeall right, sounds like i need to do some reviews and rledisez is awesome21:21
timburke#topic async cleanup of slo segments21:21
*** openstack changes topic to "async cleanup of slo segments (Meeting topic: swift)"21:21
timburkei think that's about ready to go -- thanks for the review, mattoliverau, i'll check to make sure i cover the more-than-one-container case21:22
timburkein large part i left it on here as a segue to talking about ALOs ;-)21:23
zaitcevAnother Large Objects?21:24
timburkei've been thinking "atomic", but yeah, that could work, too :P21:24
timburkethe gist of it is, i want to have something like SLOs, but where: the client API mirrors (basically exactly) S3's, the segments are all in the reserved namespace that we introduced for object versioning, and the segments get cleaned up asynchronously after delete/overwrite21:25
claygYES!!  zaitcev ❤️ 🤣21:25
claygtimburke: can we even make it so it's not racy and because we control the manifest names and segments even if an overwrite thinks it's just a create we still destroy the inaccessable segments at some point?21:27
timburkethat last part may get interesting -- i'm assuming it'll need to get plumbed clear down to diskfile's cleanup_ondisk_files -- basically, before unlinking the old version, scatter a whole bunch of async_pendings to schedule the deletes21:27
zaitcevhopefully not an auditor plugin to clean segments hehe21:27
clayginteresting!  I was thinking it'd mostly happen the container layer!  Can't wait to discuss!21:27
claygauditor plugins 😢21:27
timburkehey, those are making progress! i keep seeing new patchests from david21:28
timburkeall right, that's all i've got on the agenda21:29
timburke#topic open discussion21:29
*** openstack changes topic to "open discussion (Meeting topic: swift)"21:29
timburkeanything else we ought to bring up today?21:29
zaitcevI already engaged your attention about the dark data, so I have nothing. I'm going to look at Romain's patch seriously.21:30
claygtimburke: i was late, so i couldn't say it when you mentioned it - but KUDOS on taking care of all those backports and stable branches man - that's really great21:30
timburkei may have found a new lead on https://bugs.launchpad.net/swift/+bug/1710328 -- i suspect some of the iterable/iterator cleanup in https://review.opendev.org/#/c/755639/ may address it21:31
openstackLaunchpad bug 1710328 in OpenStack Object Storage (swift) "object server deadlocks when a worker thread logs something" [High,Fix released] - Assigned to Samuel Merritt (torgomatic)21:31
patchbotpatch 755639 - swift - New proxy logging field for wire status - 5 patch sets21:31
timburkeer, not *that* deadlock patch... https://bugs.launchpad.net/swift/+bug/189573921:31
openstackLaunchpad bug 1895739 in OpenStack Object Storage (swift) "Proxy server sometimes deadlocks while logging client disconnect" [Undecided,In progress]21:31
zaitcevwow21:33
zaitcevBTW... I don't know if it's going to be helpful to you, but I generally found re-entrant locks to be rather harmful in kernel arena.21:35
zaitcevThey aren't saving from deadlocks if 2 entities are present.21:35
zaitcevAnd, some people clearly have trouble considering how locks work.21:35
zaitcevObviously it was the case 10 years ago and I'm sure you can think about them. I'm just saying that re-entrant locks tend to make the code harder to comprehend for mere mortals.21:36
zaitcevI saw it happen when someone tried to port AFS to Linux21:37
timburkeyeah -- my hope is that if i can clean up the close() calls, we won't be punting to gc to handle the generators and i won't need to touch the _active_limbo_lock thing *at all*21:37
zaitcevAnd in drivers if an rwlock is used for re-entrancy property, it's a signal that the code is out of control and someone is band-aiding around the baggage.21:38
zaitcevOkay.21:38
timburkeeven when i *did* try swapping out the lock, it didn't actually fix the issue -- it'd still come up now and then, i think because of some craziness in eventlet21:39
timburkeit was gross.21:39
timburkebut now i've got that patch applied in the cluster where i first characterized the problem, and i'll try running the same sorts of workloads -- i guess we'll see in a month or whatever if it actually fixed it 🤮21:40
claygtimburke: can you post the link to the jira again (i don't see it from the bug)21:40
clayglp bug #189573921:41
openstackLaunchpad bug 1895739 in OpenStack Object Storage (swift) "Proxy server sometimes deadlocks while logging client disconnect" [Undecided,In progress] https://launchpad.net/bugs/189573921:41
timburkei really wish i had a reliable repro21:41
zaitcevYou know, I cannot see the emoji all that well. Is that face puking?21:41
timburkeyup -- i think that was one of the first emojis where i thought, "well that *is* difficult to express without emojis (at least, while remaining as succinct)"21:42
timburkekota_, clayg, rledisez, mattoliverau how are we feeling about https://review.opendev.org/#/c/739164/ ?21:43
patchbotpatch 739164 - swift - ec: Add an option to write fragments with legacy crc - 3 patch sets21:43
timburkehttps://review.opendev.org/#/c/738959/ landed -- i should cut a libec release21:44
patchbotpatch 738959 - liberasurecode - Be willing to write fragments with legacy crc (MERGED) - 4 patch sets21:44
*** ykatabam has joined #openstack-meeting21:44
timburke(step 1: remember how we do that)21:44
clayghahah21:45
zaitcevYeah. I only know that it ends in tarballs.opendev.org eventually.21:45
rlediseztimburke: it seems right21:45
kota_dependency is always hard problem i'm feeling21:45
timburkei think we can just push a (signed) tag, but i bet tdasilva remembers21:45
zaitcevIf we're done, I need to go.21:47
timburkekota_, the good news is that despite the Depends-On, the new swift code is perfectly happy working with old libec -- it'll set the env var, and nothing will actually look at it21:47
kota_:P21:48
timburkeall right, seems like we're winding down21:50
timburkethank you all for coming, and thank you for working on swift!21:50
timburke#endmeeting21:50
*** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/"21:50
openstackMeeting ended Wed Oct  7 21:50:51 2020 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)21:50
openstackMinutes:        http://eavesdrop.openstack.org/meetings/swift/2020/swift.2020-10-07-21.01.html21:50
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/swift/2020/swift.2020-10-07-21.01.txt21:50
openstackLog:            http://eavesdrop.openstack.org/meetings/swift/2020/swift.2020-10-07-21.01.log.html21:50
*** patchbot has left #openstack-meeting21:51
*** adrianc has quit IRC22:03
*** adrianc has joined #openstack-meeting22:04
*** njohnston has quit IRC22:04
*** yasufum has joined #openstack-meeting22:05
*** manpreet has quit IRC22:17
*** rcernin has joined #openstack-meeting22:21
*** tosky has quit IRC22:22
*** ociuhandu has joined #openstack-meeting22:32
*** ociuhandu has quit IRC22:36
*** yasufum has quit IRC22:39
*** thgcorrea has quit IRC22:47
*** macz_ has quit IRC23:02
*** zaitcev has left #openstack-meeting23:10
*** whoami-rajat__ has quit IRC23:20
*** dsariel has quit IRC23:20
*** dsariel has joined #openstack-meeting23:20
*** dsariel has quit IRC23:51
*** dsariel has joined #openstack-meeting23:51
*** yasufum has joined #openstack-meeting23:55
*** yasufum has quit IRC23:58
*** yasufum_ has joined #openstack-meeting23:58
*** mlavalle has quit IRC23:59

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!