Wednesday, 2018-04-11

openstackgerritSamuel Merritt proposed openstack/swift master: Add ability to run specific tests in py35 tox environment.  https://review.openstack.org/56018200:03
*** itlinux has joined #openstack-swift00:07
openstackgerritTim Burke proposed openstack/swift master: swift-(account|container)-info: tolerate LockTimeouts  https://review.openstack.org/56018300:09
openstackgerritMerged openstack/swift master: Add full working example of sharing a container with another user  https://review.openstack.org/47386200:15
*** gyee has quit IRC00:33
*** two_tired has joined #openstack-swift00:42
two_tiredis it possible to generate a tempURL that can list the objects in a container?00:43
two_tiredIOW, is there any way around "Non-object paths will be rejected by tempURL" ?00:44
timburketwo_tired: i'm gonna go with no, not without code changes: https://github.com/openstack/swift/blob/2.17.0/swift/common/middleware/tempurl.py#L587-L59000:48
*** germs has quit IRC00:51
*** germs has joined #openstack-swift00:52
*** germs has quit IRC00:52
*** germs has joined #openstack-swift00:52
two_tiredtimburke: OK thanks, just wanted to make sure I wasn't missing something.00:53
*** germs has quit IRC00:56
*** two_tired has quit IRC01:09
openstackgerritMerged openstack/swift master: Add manpage for swift-object-relinker  https://review.openstack.org/52687501:35
*** germs has joined #openstack-swift01:44
*** germs has quit IRC01:44
*** germs has joined #openstack-swift01:44
*** cshastri has joined #openstack-swift01:44
peluseFYI another multi-proc ASLR failure: https://ci.spdk.io/spdk/builds/review/fbb625370fe2d10bca851f748df592a156d16503.1523404157/ubuntu17.10/build.log01:49
pelusebut it passed on Jenkins :)01:49
pelusecan someone please restart CI on https://review.gerrithub.io/#/c/407222/01:49
mattoliverauhey, peluse, as always great to see you!!! however, I think those messages aren't related to Swift channel.01:53
peluseoh shit, wrong channel :)01:53
mattoliverau:)01:53
pelusemattoliverau, how the hell are ya??01:53
mattoliveraupeluse: yeah, I'm great :) I know I speak for everyone here when I say we miss having you around!01:54
peluseman I miss Swift & you fellas (& gals) too!01:54
mattoliveraupeluse: I now live on a beach, still hacking away01:54
pelusenice!01:54
peluseI gotta say though, I brought many Swift community BKMs to http://spdk.io and things are going great!!01:55
mattoliveraucool. The spdk stuff looks pretty awesome.02:00
kota_good morning02:00
mattoliveraukota_: morning02:01
kota_mattoliverau: o/02:02
kota_we'll be Vancouver almost a whole week.02:03
pelusethanks, so what beach?02:04
pelusekota_, hey man, how are ya?02:04
* peluse is jealous02:04
kota_peluse: good thanks, I'm happy to see you again here!02:04
* kota_ is reading the back logs on irc02:04
pelusewell, yeah it was by accident but that's happening more and more to be lately :)02:04
pelusebe->me02:05
kota_peluse: oic. it was wrong :/02:06
kota_but anyway nice to see.02:07
mattoliveraupeluse: I'm living in a small costal town on the coast of NSW Australia. The place is called Surf Beach.. obviously mostly a vacation spot. But I'm lucky enough to be here all year round.. Love remote work ;)02:07
mattoliverauthough does make travelling to things a little harder.. but hey it always took me ages to get anywhere outside of Oz so what's a little bit longer :P02:08
peluseyeah, things we've adopted from Swift include (a) IRC (b) weekly community meetings (3) bi-annual developer meetups (4) using Trello for both large features and our 'low hanging fruit' backlog (5) getting off of a home grown CI onto Jenkins (6) improved unit testing framework (7) improved test infrastructure.... and the list goes on.  Swift does SO MANY things right, you guys should be proud!02:09
peluse"Surf Beach".... sweet!!!02:10
mattoliverau\o/02:10
mattoliverauhopefully one day we'd also be able to include spdk into Swift. Then the circle will be complete02:11
peluseyeah, when SSDs are super prevalent and latency & CPU efficiencies are critical it's a great solution02:12
*** vint_bra has joined #openstack-swift02:20
*** germs has quit IRC02:32
*** ianychoi has quit IRC02:35
peluse(well super prevalent in Swift that is, they already are in storage appliances and in a lot of ephemeral VM use cases)02:37
*** vint_bra has quit IRC02:40
*** psachin has joined #openstack-swift03:00
*** psachin has quit IRC03:09
*** ianychoi has joined #openstack-swift03:19
openstackgerritMerged openstack/swift master: swob.Match: add __repr__  https://review.openstack.org/55666303:21
*** psachin has joined #openstack-swift03:26
*** kei-ichi has quit IRC03:28
*** kei-ichi has joined #openstack-swift03:28
*** kei_yama has quit IRC04:00
*** kei_yama has joined #openstack-swift04:01
*** d0ugal has quit IRC05:16
*** m_kazuhiro has joined #openstack-swift05:24
*** d0ugal has joined #openstack-swift05:26
*** links has joined #openstack-swift05:29
*** silor has joined #openstack-swift05:32
*** armaan has quit IRC05:42
*** armaan has joined #openstack-swift05:44
*** geaaru has quit IRC05:58
*** hseipp has joined #openstack-swift06:16
*** armaan has quit IRC06:19
*** armaan has joined #openstack-swift06:20
*** armaan has quit IRC06:24
*** armaan has joined #openstack-swift06:25
*** vish_18 has joined #openstack-swift06:38
*** hseipp has quit IRC06:49
*** rcernin has quit IRC07:06
*** ccamacho has joined #openstack-swift07:10
openstackgerritKota Tsuyuzaki proposed openstack/swift feature/s3api: Update S3api Docs  https://review.openstack.org/55285307:10
openstackgerritMerged openstack/swift master: swift-(account|container)-info: tolerate LockTimeouts  https://review.openstack.org/56018307:24
*** tesseract has joined #openstack-swift07:28
*** geaaru has joined #openstack-swift07:40
*** m_kazuhiro has quit IRC07:42
*** mikecmpbll has joined #openstack-swift07:48
*** cbartz has joined #openstack-swift08:05
*** gkadam has joined #openstack-swift08:06
*** pcaruana has joined #openstack-swift08:13
openstackgerritAlistair Coles proposed openstack/swift feature/deep: Minor fix: don't report progress when sharding has completed  https://review.openstack.org/56030908:26
acolespeluse!08:26
acolesgood morning08:26
*** paramite has joined #openstack-swift08:32
*** kei_yama has quit IRC08:40
*** ianychoi has quit IRC08:42
mattoliverauacoles: morning08:47
openstackgerritAlistair Coles proposed openstack/swift feature/deep: Only a leader should set Scan-Done metadata  https://review.openstack.org/56011308:59
openstackgerritAlistair Coles proposed openstack/swift feature/deep: Fix fetching own shard range from root during audit  https://review.openstack.org/56011408:59
*** mikecmpbll has quit IRC09:05
*** ianychoi has joined #openstack-swift09:46
*** silor1 has joined #openstack-swift09:55
openstackgerritAlistair Coles proposed openstack/swift feature/deep: Start from last max row when repeating cleaving  https://review.openstack.org/55949109:55
openstackgerritAlistair Coles proposed openstack/swift feature/deep: Take lock before unlinking db files  https://review.openstack.org/55979609:55
*** silor has quit IRC09:57
*** silor1 is now known as silor09:57
openstackgerritAlistair Coles proposed openstack/swift feature/deep: Shard range updates should not revive deleted container  https://review.openstack.org/55932410:09
*** cshastri has quit IRC10:19
*** armaan has quit IRC10:36
*** armaan has joined #openstack-swift10:36
*** ccamacho has quit IRC10:46
*** ccamacho has joined #openstack-swift10:48
*** mikecmpbll has joined #openstack-swift10:56
openstackgerritKota Tsuyuzaki proposed openstack/swift feature/s3api: Update S3api Docs  https://review.openstack.org/55285311:07
*** mikecmpbll has quit IRC11:15
*** mikecmpbll has joined #openstack-swift11:18
*** armaan has quit IRC11:39
*** armaan has joined #openstack-swift11:39
*** hseipp has joined #openstack-swift11:51
*** gkadam has quit IRC11:54
*** gkadam has joined #openstack-swift11:59
*** hseipp has quit IRC12:02
openstackgerritAlistair Coles proposed openstack/swift feature/deep: Move state alias resolution into ContainerBroker  https://review.openstack.org/56039712:09
openstackgerritAlistair Coles proposed openstack/swift feature/deep: Don't redirect updates to shrinking shards  https://review.openstack.org/56039812:09
*** NM has joined #openstack-swift12:30
*** alee has joined #openstack-swift12:32
*** mvk has quit IRC12:40
fricklerthese staticweb tests are failing when the server address is native IPv6 like "[::1]" because the brackets don't seem to get quoted properly: http://logs.openstack.org/03/505503/6/check/legacy-swift-dsvm-functional/c27b469/job-output.txt.gz#_2018-04-11_06_45_39_786429 maybe someone wants to fix that? ;)12:47
fricklersee https://review.openstack.org/505168 for context12:47
patchbotpatch 505168 - openstack-dev/devstack - Fix running with SERVICE_IP_VERSION=612:47
frickleroohh, I've never seen patchbot. /me wants to have that in other channels, too12:48
*** NM has left #openstack-swift12:54
*** ccamacho is now known as ccamacho|lunch12:57
*** armaan has quit IRC12:58
*** armaan has joined #openstack-swift12:58
*** gkadam_ has joined #openstack-swift12:59
*** gkadam has quit IRC13:02
*** armaan has quit IRC13:03
*** psachin has quit IRC13:05
*** gkadam_ has quit IRC13:08
*** mvk has joined #openstack-swift13:11
*** cshastri has joined #openstack-swift13:13
DHEfrickler: in #zfsonlinux we have one that provides github feedback. It's quite nice, but github's social nature tends to make it noisy13:24
*** cshastri has quit IRC13:26
fricklerDHE: do you run this bot? from it's info I assume it is a sopel module, wondering if it is published somewhere13:26
DHEno it's not mine. I suspect it's based on all of github's API. the other is a project/organization member13:28
openstackgerritMerged openstack/swift feature/deep: Minor fix: don't report progress when sharding has completed  https://review.openstack.org/56030913:50
*** armaan has joined #openstack-swift13:51
*** Supun has joined #openstack-swift13:52
*** links has quit IRC13:58
*** jistr is now known as jistr|mtg14:02
*** ccamacho|lunch is now known as ccamacho14:02
*** Supun has quit IRC14:22
*** Supun has joined #openstack-swift14:24
*** itlinux has quit IRC14:34
notmynamegood morning14:35
notmynamefrickler: I can put patchbot somewhere else, if you want it14:35
*** Supun has quit IRC14:46
*** cshastri has joined #openstack-swift14:50
fricklernotmyname: did you write that module? infra has a plan to integrate things as supybot plugins, so instead of just having that bot just serve additional channels, I'd be even more interesting in seeing the source and trying to port it https://specs.openstack.org/openstack-infra/infra-specs/specs/irc.html15:01
notmynameI wrote patchbot, yes. used to be supybot, but I found and like sopel a lot more. patchbot is currently using sopel15:02
notmynamehttps://github.com/notmyname/Patches215:02
notmynamespoel makes it *vastly* less complex, which makes it much more stable and simpler to debug15:03
notmynamehttps://github.com/notmyname/Patches is the older supyboy version. but it's much worse for setup configs and stuff like that.15:06
*** armaan has quit IRC15:08
*** armaan has joined #openstack-swift15:09
*** gkadam has joined #openstack-swift15:09
*** jistr|mtg is now known as jistr15:10
*** germs has joined #openstack-swift15:11
*** germs has quit IRC15:11
*** germs has joined #openstack-swift15:11
*** germs has quit IRC15:16
*** germs has joined #openstack-swift15:17
*** germs has quit IRC15:17
*** germs has joined #openstack-swift15:17
*** paramite has quit IRC15:21
*** hseipp has joined #openstack-swift15:24
notmynameI'm only going to be available for the first 30 minutes of today's swift meeting. quick check on ongoing stuff, and then we'll end the meeting early15:26
*** alee has left #openstack-swift15:30
*** gkadam_ has joined #openstack-swift15:39
*** gkadam has quit IRC15:40
*** cshastri has quit IRC15:42
*** Supun has joined #openstack-swift15:47
*** itlinux has joined #openstack-swift15:50
*** guimaluf has joined #openstack-swift15:51
openstackgerritAlexandre Lécuyer proposed openstack/swift master: Initial LOSF RPC commit  https://review.openstack.org/56048015:54
*** gyee has joined #openstack-swift15:56
*** gkadam__ has joined #openstack-swift16:01
*** gkadam_ has quit IRC16:04
*** gkadam has joined #openstack-swift16:04
*** gkadam__ has quit IRC16:06
notmynamealecuyer_: whoa16:08
*** cbartz has quit IRC16:09
alecuyer_notmyname: that's the RPC code. As new code, the easiest to rebase ;-)16:11
*** armaan has quit IRC16:12
*** armaan has joined #openstack-swift16:12
*** hseipp has quit IRC16:24
*** Supun has quit IRC16:32
*** mikecmpbll has quit IRC16:36
*** knylander has joined #openstack-swift16:40
*** tesseract has quit IRC16:59
*** gkadam has quit IRC16:59
*** ndk_ has quit IRC17:23
*** portante has quit IRC17:23
*** portante has joined #openstack-swift17:31
*** ndk_ has joined #openstack-swift17:31
*** Supun has joined #openstack-swift17:42
openstackgerritMerged openstack/swift feature/s3api: Update S3api Docs  https://review.openstack.org/55285317:42
tdasilvawow, looks like s3api patches are all done17:46
tdasilva?17:46
tdasilvas/done/merged17:47
notmynameooo, quite interesting17:52
notmynamehttps://review.openstack.org/#/q/project:openstack/swift+branch:feature/s3api17:52
*** geaaru has quit IRC17:53
*** armaan has quit IRC17:54
*** armaan has joined #openstack-swift17:54
*** Supun has quit IRC17:59
notmynametdasilva: what's your opinion on the feature/s3api branch and it's "mergeableness"?18:09
tdasilvai think kota_ is awesome and I've struggled to keep up with him :)18:11
tdasilvaseriously thou, i think he did a great job to get all the worked we had talked about at the PTG merged18:11
tdasilvathere's still remains the question you brought up last time of: "are we ok merging with the currently functionality AS-IS?" or something like that....18:12
notmynameyeah18:12
tdasilvaif we are ok as is, i *think* it's ready to be proposed to master and we can continue making changes there on master for minor things....18:14
notmynamedo we go ahead and merge it to have it merged? or do we wait an undetermined time until after feature/deep initial work gets done?18:14
timburkei'm happy to have gotten https://review.openstack.org/#/c/557526/ in :-)18:14
notmynameif we merge it now, we'll have to deal with long-term migration/support issues of the stuff we want to change (eg changing the way buckets and auth work so that we can add support for more features)18:14
patchbotpatch 557526 - swift (feature/s3api) - Remove auth_host, auth_port, auth_protocol, and au... (MERGED)18:14
timburkehttps://review.openstack.org/#/c/557537/ was good, too18:15
patchbotpatch 557537 - swift (feature/s3api) - Remove zero-byte final segment hack (MERGED)18:15
notmynameif we wait, we've got this body of work that's sort of in limbo without any clear way to make forward progress, even for smaller stuff we'd like to add to it18:15
tdasilvatimburke: did you have other changes in swift3 that you would like to be merged?18:16
notmynamethis time last week I was very much in support of keeping feature/s3api on hold18:16
timburkenot that can't land on master, i don't think18:16
tdasilvaok18:16
notmynamebut yes there are some18:16
notmynameright?18:16
notmynamethe patches we're carrying locally?18:16
tdasilvanotmyname: i guess it all depends on the expectations after feature/deep is merged...18:17
tdasilvanot sure who would be able to keep pushing it forward to where we 'ultimately' want it to be18:17
notmynamecontainer sharding isn't "done" when feature/deep lands. there's definitely more work to be done. eventually, we need to be able to have every container automatically sharded (and unsharded) as needed, transparently to the user18:18
notmynameswiftstack will keep working towards that goal (at least that's the plan now)18:19
timburkenotmyname: yup -- at some point, i'll want pretty much all the parents of https://github.com/swiftstack/swift3-stackforge/commit/3ad529718:19
timburke(or something that addresses the same root issues)18:20
notmynamestuff going on: feautre/deep, consistency engine (replicator/reconstructor performance), LOSF, feature/s3api, policy migration18:20
notmynamefor feature/s3api, future work should include solving the bucket->(account,container) mapping so that (1) we can support more of the bucket api and (2) we can support more ACL forms18:21
*** armaan has quit IRC18:31
*** armaan has joined #openstack-swift18:31
*** armaan has quit IRC18:38
*** armaan has joined #openstack-swift18:39
*** Supun has joined #openstack-swift18:40
*** mvk has quit IRC18:46
*** Supun has quit IRC18:55
*** pcaruana has quit IRC18:55
*** Supun has joined #openstack-swift19:05
*** Supun has quit IRC19:20
*** mvk has joined #openstack-swift19:23
*** mikecmpbll has joined #openstack-swift19:33
*** Supun has joined #openstack-swift19:39
*** knylander has quit IRC19:50
*** Supun has quit IRC19:53
*** Supun has joined #openstack-swift19:54
*** silor has quit IRC20:03
*** Supun has quit IRC20:09
notmynameI've got a weird schedule today. I will do my best to be at the meeting at 2100, but I might be a few minutes late. and, as I said earlier, I'll need to end it at 2130 at the latest20:16
*** paramite has joined #openstack-swift20:20
*** amito_ has joined #openstack-swift20:21
*** zacksh_ has joined #openstack-swift20:22
*** acoles_ has joined #openstack-swift20:27
*** ChanServ sets mode: +v acoles_20:27
openstackgerritAlistair Coles proposed openstack/swift feature/deep: WIP fix ContainerBroker.is_deleted method  https://review.openstack.org/56062920:27
*** zacksh has quit IRC20:28
*** acoles has quit IRC20:28
*** amito has quit IRC20:28
*** acoles_ is now known as acoles20:28
*** amito_ is now known as amito20:28
*** torgomatic has quit IRC20:30
*** torgomatic has joined #openstack-swift20:31
*** ChanServ sets mode: +v torgomatic20:31
*** Supun has joined #openstack-swift20:39
*** paramite has quit IRC20:39
*** Supun has quit IRC20:52
*** Supun has joined #openstack-swift20:53
kota_good morning20:55
mattoliverauMorning20:56
*** m_kazuhiro has joined #openstack-swift20:58
acoleskota_: mattoliverau o/20:59
notmynamewhew, I made it20:59
timburkeyay!20:59
notmynamereally? ;-)21:00
notmynamehad to go get the kids from school and get back21:00
notmynameok, meeting time in #openstack-meeting. will be 30 minutes max, probably less21:00
*** Supun has quit IRC21:05
*** m_kazuhiro has quit IRC21:32
mattoliverauI had an idea to potentially speed up the shard scanning. That should avoid offset and so didn't end up having to scan much of the result sets. But determining the middle string got kinda hard. Maybe find the code and play with it some more today. In the hopes it'll be a little faster.21:35
mattoliverauGot hard in some cases.21:36
mattoliverauAnyway breakfast time21:37
*** rcernin has joined #openstack-swift21:49
*** itlinux has quit IRC22:03
*** itlinux has joined #openstack-swift22:06
*** itlinux has quit IRC22:07
*** mikecmpbll has quit IRC22:16
-openstackstatus- NOTICE: zuul was restarted to updated to the latest code; you may need to recheck changes uploaded or approvals added between 21:30 and 21:4522:31
timburkeacoles: i think we gotta find a way to have manage-shard-ranges not call _commit_puts_stale_ok so much. i tried cranking up broker.timeout to 10 minutes then running find_and_replace (didn't even include --enable!) against a db that i was manually holding the lock on... 40 minutes later i gave up waiting22:35
*** d0ugal has quit IRC22:43
timburkeit hadn't even gotten as far as saying "No shard ranges found to delete." :-(22:46
*** d0ugal has joined #openstack-swift22:48
timburkemaybe it's not even about that, but establishing fewer connections? like, get a connection once, then *hold onto it* until we're done?22:49
*** Raymii has quit IRC23:01
*** Raymii has joined #openstack-swift23:02
DHEWould the sqlite database replicator processes properly handle the situation where all nodes seemed to receive updates properly, but then one crashed and rebooted resulting in the on-disk data rolling back a bit?23:10
*** kei_yama has joined #openstack-swift23:30

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!