Wednesday, 2017-12-06

*** two_tired has joined #openstack-swift00:14
openstackgerritMatthew Oliver proposed openstack/swift feature/deep: Use the same hash and created_at value in shard db  https://review.openstack.org/51648800:22
*** threestrands has quit IRC00:25
timburkemattoliverau: interesting... but... why hasn't this bitten us before??00:27
mattoliverautimburke: because  that patch put the temp shard (preshard) back in the disks tmp folder, so it'll be cleaned in relcaim age so I just be adding some smarts to renamer in case people do have the swift's drive tmp on a different filesystem.00:31
mattoliverauhmm, I might not be actually fixing the main problem. maybe in the xfs tmp tests that fail we must be placing it beyond the xfs filesystem.00:32
mattoliverauwhen I think about, it should never span multiple filesystems if its the devices tmp00:33
mattoliverauI might have to take a look closer as the xfs tmp checks.00:33
* mattoliverau might have to rework that patch, to fix the real issue, not cover it up :)00:34
timburkeyeah, i think that's the trouble -- we need the hash/suffix/partition parts to the path00:37
mattoliverautimburke: yup. I blame myself for being stupid :P00:39
openstackgerritAlistair Coles proposed openstack/swift master: Add --version option to swift-recon man page  https://review.openstack.org/52584000:41
acolesnotmyname: ^^00:43
*** tovin07_ has joined #openstack-swift00:49
*** itlinux has quit IRC01:00
*** tristanC has quit IRC01:08
*** tristanC has joined #openstack-swift01:09
*** gyee has quit IRC01:10
*** cshastri has joined #openstack-swift01:29
*** psachin has joined #openstack-swift01:30
openstackgerritKazuhiro MIYAHARA proposed openstack/swift master: Symlink implementation.  https://review.openstack.org/23216201:35
*** m_kazuhiro has joined #openstack-swift01:36
*** mat128 has joined #openstack-swift01:36
*** tovin07_ has quit IRC01:36
m_kazuhirogood morning01:41
*** saint_ has quit IRC01:44
*** tovin07_ has joined #openstack-swift01:46
*** saint_ has joined #openstack-swift01:55
*** mat128_ has joined #openstack-swift02:02
*** mat128 has quit IRC02:05
kota_good morning02:17
mattoliveraum_kazuhiro, kota_: morning02:22
m_kazuhiromattoliverau: morning02:22
kota_mattoliverau: hi02:24
*** SkyRocknRoll has joined #openstack-swift02:41
*** itlinux has joined #openstack-swift02:45
*** itlinux has quit IRC02:53
*** SkyRocknRoll has quit IRC03:04
openstackgerritMatthew Oliver proposed openstack/swift feature/deep: Use the same hash and created_at value in shard db  https://review.openstack.org/51648803:18
mattoliverautimburke: ^ that adds the extra tempdirs. Tho was lazy and just named the part, suffix and hash.. rather then just giving them some default number.03:19
*** abhinavtechie has joined #openstack-swift03:19
*** abhitechie has quit IRC03:19
*** abhinavtechie has quit IRC03:52
*** links has joined #openstack-swift03:58
*** links has quit IRC04:01
*** links has joined #openstack-swift04:01
*** m_kazuhiro has quit IRC04:30
*** _ix has joined #openstack-swift04:36
*** _ix_ has joined #openstack-swift04:41
*** _ix has quit IRC04:41
*** tovin07_ has quit IRC04:41
*** tovin07_ has joined #openstack-swift04:42
*** SkyRocknRoll has joined #openstack-swift04:50
*** SkyRocknRoll_ has joined #openstack-swift05:02
*** SkyRocknRoll has quit IRC05:03
*** threestrands has joined #openstack-swift05:09
*** threestrands has quit IRC05:09
*** two_tired has quit IRC05:16
*** SkyRocknRoll_ has quit IRC05:24
*** _ix_ has quit IRC05:49
*** mat128_ has quit IRC06:13
*** bkopilov has quit IRC06:35
openstackgerritKota Tsuyuzaki proposed openstack/swift master: Symlink implementation.  https://review.openstack.org/23216206:49
*** gkadam has joined #openstack-swift06:54
*** manous_ has joined #openstack-swift06:55
*** rcernin has quit IRC07:02
*** rcernin has joined #openstack-swift07:39
*** hseipp has joined #openstack-swift07:53
*** hseipp has left #openstack-swift07:54
*** samueldmq has quit IRC08:04
*** samueldmq has joined #openstack-swift08:05
*** zhongjun has quit IRC08:08
*** tesseract has joined #openstack-swift08:19
openstackgerritMatthew Oliver proposed openstack/swift master: Add a -L or --list to recon to list all results  https://review.openstack.org/52503908:34
*** oshritf has joined #openstack-swift08:35
*** kei_yama has quit IRC08:36
*** SkyRocknRoll has joined #openstack-swift08:37
*** cbartz has joined #openstack-swift08:45
*** geaaru has joined #openstack-swift08:58
*** ukaynar has quit IRC09:19
*** ukaynar has joined #openstack-swift09:20
*** armaan has joined #openstack-swift09:21
*** manous_ has quit IRC09:23
*** ukaynar has quit IRC09:24
*** armaan has quit IRC09:31
*** oshritf has quit IRC09:34
*** zhongjun has joined #openstack-swift09:35
*** d0ugal has quit IRC09:37
*** d0ugal has joined #openstack-swift09:41
*** armaan has joined #openstack-swift09:53
openstackgerritKota Tsuyuzaki proposed openstack/swift master: Symlink implementation.  https://review.openstack.org/23216209:55
*** oshritf has joined #openstack-swift09:56
kota_I'm sure that can be the final version of symlink patch before landed.09:56
*** linkmark has joined #openstack-swift09:57
*** oshritf has quit IRC09:58
*** oshritf has joined #openstack-swift10:07
*** oshritf has quit IRC10:09
*** armaan has quit IRC10:16
*** silor has joined #openstack-swift10:16
*** tovin07_ has quit IRC10:27
*** rajaios has joined #openstack-swift10:30
*** oshritf has joined #openstack-swift10:31
*** oshritf has quit IRC10:31
rajaiosWhich is the good libraries for mqtt?10:32
*** rajaios is now known as raja10:32
*** raja is now known as Guest7780710:33
*** d0ugal has quit IRC10:39
*** cshastri has quit IRC10:50
*** oshritf has joined #openstack-swift10:50
*** mvk has quit IRC10:52
*** d0ugal has joined #openstack-swift11:01
*** Guest77807 has quit IRC11:17
*** ukaynar has joined #openstack-swift11:21
*** ukaynar has quit IRC11:25
*** links has quit IRC11:34
*** rcernin has quit IRC11:39
*** links has joined #openstack-swift11:47
*** silor1 has joined #openstack-swift11:49
*** silor has quit IRC11:53
*** silor1 is now known as silor11:53
*** openstackgerrit has quit IRC12:03
*** links has quit IRC12:08
*** silor has quit IRC12:11
*** links has joined #openstack-swift12:13
*** SkyRocknRoll has quit IRC12:17
*** silor has joined #openstack-swift12:20
*** ganders has joined #openstack-swift12:24
*** mvk has joined #openstack-swift12:41
*** tesseract has quit IRC12:53
*** tesseract has joined #openstack-swift12:57
*** armaan has joined #openstack-swift13:09
*** armaan has quit IRC13:20
*** ukaynar has joined #openstack-swift13:21
*** links has quit IRC13:23
*** ukaynar has quit IRC13:26
*** _ix has joined #openstack-swift13:28
*** _ix_ has joined #openstack-swift13:33
*** _ix has quit IRC13:36
*** psachin has quit IRC13:45
*** _ix_ has quit IRC13:47
*** openstackgerrit has joined #openstack-swift13:49
openstackgerritMerged openstack/swift master: dlo: Move conn2 business to the one test that uses it  https://review.openstack.org/52241313:49
*** _ix has joined #openstack-swift13:58
rlediseznotmyname, torgomatic: I got news about swift-objet-updater hanging. it's related to eventlet+logging (obviously…) I used eventlet.backdoor to get a stack of greenlet threads and found that (http://paste.openstack.org/show/628269/). after commenting all self.logger call, I got a lot less hangs, but still some. New stack: http://paste.openstack.org/show/628276/ <- this one I can't fix by just commenting some lines. i don't know14:14
rledisez how to fix that, it's way above my python expertise.14:14
rledisezfor the second stack, logging module is auto registering itself on atexit, so when an object-updater worker is exiting, it calls logging.shutdown => the more concurrency, the more probability it happens at a bad "eventlet timing" I guess14:15
*** SkyRocknRoll has joined #openstack-swift14:17
rledisezany idea welcome. my best one for now is to remove all references to logging in Swift code to use a homemade syslog logger (I think alecuyer already wrote one to work on diskfile without hanging object-server everytime)14:17
*** gkadam has quit IRC14:18
frankkahlegood morning folks, so i ran the tox -e py27 -r command and had only 1 failure.... (FAIL: test_multiphase_put_drains_extra_commit_junk_disconnect (test.unit.obj.test_server.TestObjectServer))14:19
*** SkyRocknRoll has quit IRC14:30
*** ganders has quit IRC14:33
*** tesseract has quit IRC14:37
*** ganders has joined #openstack-swift14:38
*** silor1 has joined #openstack-swift14:38
rledisezmornin14:38
* rledisez typed in the wrong channel, and it's not even morning here…14:39
*** silor has quit IRC14:42
*** silor1 is now known as silor14:42
*** ganders has quit IRC14:42
*** tesseract has joined #openstack-swift14:44
*** mat128 has joined #openstack-swift14:59
*** ukaynar has joined #openstack-swift15:01
openstackgerritMerged openstack/swift master: Added swift version to recon cli  https://review.openstack.org/41399115:01
*** oshritf has quit IRC15:02
*** oshritf has joined #openstack-swift15:02
*** armaan has joined #openstack-swift15:03
*** ganders has joined #openstack-swift15:07
*** ganders has quit IRC15:12
*** oshritf has quit IRC15:12
*** armaan has quit IRC15:16
openstackgerritChristopher Bartz proposed openstack/swift master: X-Delete-At-* headers for all object nodes.  https://review.openstack.org/52454815:18
*** ganders has joined #openstack-swift15:19
*** ganders has quit IRC15:23
*** ganders has joined #openstack-swift15:27
*** ganders_ has joined #openstack-swift15:30
*** ganders has quit IRC15:31
*** ganders_ has quit IRC15:34
*** oshritf has joined #openstack-swift15:39
*** ganders has joined #openstack-swift15:48
*** mvk has quit IRC15:48
*** ganders has quit IRC15:52
*** ianychoi has quit IRC15:56
*** ianychoi_ has joined #openstack-swift15:56
*** ianychoi_ is now known as ianychoi15:57
openstackgerritMerged openstack/swift master: Better handle missing files in _construct_from_data_file  https://review.openstack.org/51740415:59
*** ganders has joined #openstack-swift16:05
*** ganders has quit IRC16:19
*** klrmn_ has joined #openstack-swift16:32
*** ganders has joined #openstack-swift16:36
*** ukaynar has quit IRC16:37
*** ukaynar has joined #openstack-swift16:37
*** ganders has quit IRC16:41
*** ukaynar has quit IRC16:42
*** gyee has joined #openstack-swift16:48
*** ganders has joined #openstack-swift16:52
*** ganders has quit IRC16:57
*** ganders has joined #openstack-swift17:02
*** cbartz has quit IRC17:04
*** ganders has quit IRC17:07
*** itlinux has joined #openstack-swift17:08
*** klrmn_ has quit IRC17:09
openstackgerritAlistair Coles proposed openstack/swift master: Add --swift-versions option to swift-recon man page  https://review.openstack.org/52584017:12
*** ganders has joined #openstack-swift17:21
*** ganders has quit IRC17:24
*** ganders has joined #openstack-swift17:24
timburkerledisez: does that version of swift include https://github.com/openstack/swift/commit/dc8da5b ?17:31
timburkewe've got backports open for pike and ocata in p 512457 and p 517090 (notmyname, wanna go +A?)17:32
patchbothttps://review.openstack.org/#/c/512457/ - swift (stable/pike) - Use "poll" or "selects" Eventlet hub for all Swift...17:32
patchbothttps://review.openstack.org/#/c/517090/ - swift (stable/ocata) - Use "poll" or "selects" Eventlet hub for all Swift...17:32
notmynameyes, it's on my todo list17:32
timburkei think there's some env var you can set to switch the eventlet hub, too, if you wanted to try that with no code change... hold on...17:33
timburke`export EVENTLET_HUB=poll`, say17:35
*** ukaynar has joined #openstack-swift17:50
*** joeljwright has joined #openstack-swift17:59
*** ChanServ sets mode: +v joeljwright17:59
*** tesseract has quit IRC18:07
*** klrmn_ has joined #openstack-swift18:10
openstackgerritAlistair Coles proposed openstack/swift master: Remove repeated text from deployment guide  https://review.openstack.org/52614118:30
*** gkadam has joined #openstack-swift18:32
rlediseztimburke: no, i don't have that patch yet, we are still running 2.15. i'm gonna try that tomorrow. thx a lot for the pointer18:33
frankkahlegood afternoon folks, so i ran the tox -e py27 -r command and had only 1 failure.... (FAIL: test_multiphase_put_drains_extra_commit_junk_disconnect (test.unit.obj.test_server.TestObjectServer))18:43
notmynamethat sounds like a pretty good improvement!19:03
notmynamewhat time is it? it's backport time!19:08
*** saint_ has quit IRC19:12
*** geaaru has quit IRC19:13
notmynamedone!19:26
frankkahle-->notmyname--> Yeah i just dont have any idea how to fix it19:29
notmynamein the test run, you should have a traceback on where the error happened. can you pastebin that and share it?19:30
notmynamehttps://github.com/openstack/swift/blob/master/test/unit/obj/test_server.py#L7570-L7583 makes me smile19:32
notmynamefrankkahle: FWIW, I just reran the tests a couple of times in a couple of different ways, and it all works19:39
*** mvk has joined #openstack-swift19:39
*** ganders has quit IRC19:47
acolesnotmyname: code like that makes git blame redundant :)19:48
*** ganders has joined #openstack-swift19:49
tdasilvaacoles: i guessed torgomatic was i right?19:50
notmynametdasilva: I did too, at first19:50
acolestdasilva: I didn't even bother checking :)19:50
notmynamebut nope! guess again19:50
acolesnotmyname: really !?!19:50
notmynamepeluse19:50
acolesco-authored surely19:51
*** klrmn_ has quit IRC20:06
*** ganders has quit IRC20:07
*** gkadam has quit IRC20:14
*** Anticime1 has quit IRC20:31
*** csmart has quit IRC20:31
*** fyxim has quit IRC20:31
*** kong has quit IRC20:31
*** verto has quit IRC20:31
*** jrichli- has quit IRC20:31
*** gyee has quit IRC20:31
*** ianychoi has quit IRC20:31
*** supamatt has quit IRC20:31
*** peterlisak has quit IRC20:31
*** jarbod_ has quit IRC20:31
*** notmyphone has joined #openstack-swift20:39
*** ChanServ sets mode: +v notmyphone20:39
notmyphoneLunch is likely to run over for me. Meeting is still on, but likely to start a little late20:39
*** itlinux has quit IRC20:40
tdasilvalol, had to do a double take on the irc handle20:46
notmyphone:)20:46
*** Anticime1 has joined #openstack-swift20:47
*** fyxim has joined #openstack-swift20:47
*** kong has joined #openstack-swift20:47
*** verto has joined #openstack-swift20:47
*** jrichli- has joined #openstack-swift20:47
*** gyee has joined #openstack-swift20:47
*** ianychoi has joined #openstack-swift20:47
*** supamatt has joined #openstack-swift20:47
*** peterlisak has joined #openstack-swift20:47
*** jarbod_ has joined #openstack-swift20:47
*** itlinux has joined #openstack-swift20:53
*** m_kazuhiro has joined #openstack-swift20:53
mattoliverauMorning20:58
notmyphoneFinishing up now. Give me 15 minutes?20:58
mattoliverauKk20:58
notmyphoneI’m going to blame it on timburke ;)21:01
mattoliverautimburke: stop forcing notmyname to lunch :p21:01
*** notmyphone has quit IRC21:09
*** itlinux has quit IRC21:13
*** mat128 has quit IRC21:14
*** silor has quit IRC21:14
notmynameI'm back21:16
notmynameok, let's do the swift meeeting in #openstack-meeting21:16
clayglet's do it!21:16
timburkesorry everybody ;-)21:16
*** itlinux has joined #openstack-swift21:26
*** openstackgerrit has quit IRC21:38
*** sai has quit IRC21:38
*** lifeless has quit IRC21:38
*** zaitcev has quit IRC21:38
*** afazekas has quit IRC21:38
*** sc has quit IRC21:38
*** zaitcev has joined #openstack-swift21:40
*** ChanServ sets mode: +v zaitcev21:40
notmynamezaitcev: I was wondering about your protocol patch21:43
notmynamehttps://review.openstack.org/#/c/427911/21:43
patchbotpatch 427911 - swift - PUT+POST and its development test21:43
notmynamecan you rebase it?21:43
zaitcevnotmyname: I started moving Sam's patch forward. Applied his "Step 1" and added a thing known as "DiskFileCache", which is well... a cache that can expire and close stale writers.21:44
zaitcevYes, I can rebase any time. I always keep it up-to-date in my tree.21:44
notmynamewhich one?21:44
zaitcevhttps://review.openstack.org/#/c/509607/21:45
patchbotpatch 509607 - swift - Step 1: make DiskFileWriters able to hang around21:45
zaitcevBut I had some issue with Gnocchi that I need to work on semi-urgently21:46
notmynamezaitcev: ah ok. that's the "sam's patch" you mentioned in the commit message of yours?21:46
notmynameah, I understand :-)21:46
zaitcevyes21:46
notmynametorgomatic: is https://review.openstack.org/#/c/340526/ step two?21:47
patchbotpatch 340526 - swift - Unify Putter and MIMEPutter into a single class.21:47
torgomaticnotmyname: step two of something, probably21:47
notmynameit was under the "related changes" gerrit tab21:47
*** m_kazuhiro has quit IRC21:47
zaitcevnotmyname, torgomatic told me that I'm free to go wild and make it useful, well as long as I'm prompt about it21:47
notmyname:-)21:47
zaitcevoh, wait21:48
zaitcevthe "unify" is not the one21:48
notmynamethe context here is that I've got a conversation with someone later today about where to add some devs to upstream swift work, and I'd like to point them in the direction of patch 42791121:48
patchbothttps://review.openstack.org/#/c/427911/ - swift - PUT+POST and its development test21:48
notmynameand I want to know the right things that are related and the current next steps for it21:48
notmynamenote this is something volunteering some devs. I don't know anything about them, much less have met them yet21:49
zaitcevSo... I have fixed up all the items that guys pointed out in the video meeting that tdasilva has set up a while back. That's done... except 1 thing: I abandoned the use of linkat().21:49
notmynameok. are you planning on bringing it back?21:50
zaitcevYes21:50
*** joeljwright has quit IRC21:50
zaitcevI have to. Or else Sam will -1 it :-)21:50
notmynameheh21:51
timburkehuh. for some reason, i thought i'd added func tests with https://review.openstack.org/#/c/285754/ ...21:53
patchbotpatch 285754 - swift - Require account/container metadata be UTF-8 (MERGED)21:53
notmynamezaitcev: ok, I'll point them towards these patches and you as someone who can help answer questions21:54
notmynamethanks21:55
zaitcevsure21:58
*** itlinux has quit IRC21:59
kota_Sorry, my alarm has been off because of no meeting in the last week22:00
* kota_ woke up now22:01
timburkegood. you *should* be able to sleep in a little :-)22:01
*** openstackgerrit has joined #openstack-swift22:01
*** sai has joined #openstack-swift22:01
*** sc has joined #openstack-swift22:01
*** lifeless has joined #openstack-swift22:01
*** afazekas has joined #openstack-swift22:01
*** rcernin has joined #openstack-swift22:01
*** rcernin has quit IRC22:03
*** rcernin has joined #openstack-swift22:03
kota_timburk: thx! It looks the meeting was going well22:07
*** ukaynar has quit IRC22:45
*** JimCheung has joined #openstack-swift22:58
*** _ix has quit IRC23:22
*** zacksh has quit IRC23:25
*** JimCheung has quit IRC23:26
*** armaan has joined #openstack-swift23:26
*** JimCheung has joined #openstack-swift23:27
*** _ix has joined #openstack-swift23:27
*** openstack has joined #openstack-swift23:29
*** ChanServ sets mode: +o openstack23:29
*** kei_yama has joined #openstack-swift23:32
*** zacksh has joined #openstack-swift23:33
*** zacksh has quit IRC23:38
*** zacksh has joined #openstack-swift23:41

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