Tuesday, 2017-07-25

*** frank_young has joined #openstack-swift00:18
torgomaticAnyone know what's up with all the use of the logging module in diskfile.py? Seems like in the various places it's called, we've got a logger object that's been configured with all the destinations and formats and whatnot, but we call logging.exception() instead.00:21
*** frank_young has quit IRC00:23
timburketorgomatic: no, but it makes me think of https://bugs.launchpad.net/swift/+bug/1555267 ...00:34
openstackLaunchpad bug 1555267 in OpenStack Object Storage (swift) "swift-account-info and swift-container-info don't set up any loggers" [Undecided,New]00:34
torgomatictimburke: maybe I'll work on that later, then, once I get this other thing working...00:35
torgomaticseems like we ought to use the configured logger instead of the default logging-module behaviour though00:37
*** vint_bra has joined #openstack-swift00:38
*** Renich has joined #openstack-swift00:42
*** itlinux has joined #openstack-swift00:46
*** catintheroof has joined #openstack-swift00:47
*** tovin07_ has joined #openstack-swift00:52
kota_good morning00:54
*** frank_young has joined #openstack-swift00:57
*** kei_yama has quit IRC00:57
timburkegood morning kota_00:59
kota_hello timburke00:59
*** klrmn has quit IRC01:00
*** frank_young has quit IRC01:01
*** kei_yama has joined #openstack-swift01:04
*** JimCheung has quit IRC01:11
*** kei_yama has quit IRC01:11
*** vint_bra has quit IRC01:12
*** kei_yama has joined #openstack-swift01:15
*** catintheroof has quit IRC01:18
*** frank_young has joined #openstack-swift01:26
*** chlong_ has joined #openstack-swift01:45
*** chlong__ has quit IRC01:47
*** tovin07_ has quit IRC01:51
*** tovin07_ has joined #openstack-swift01:54
*** bkopilov_ has quit IRC01:59
*** bkopilov has quit IRC01:59
*** itlinux has quit IRC02:00
*** itlinux has joined #openstack-swift02:00
*** chlong__ has joined #openstack-swift02:06
*** chlong_ has quit IRC02:07
*** JimCheung has joined #openstack-swift02:14
*** klrmn has joined #openstack-swift02:17
*** JimCheung has quit IRC02:18
*** chlong_ has joined #openstack-swift02:25
*** chlong__ has quit IRC02:27
*** winggundamth has joined #openstack-swift02:38
*** cschwede_ has joined #openstack-swift02:57
*** cschwede has quit IRC02:59
*** bkopilov_ has joined #openstack-swift03:15
*** bkopilov has joined #openstack-swift03:15
*** kiennt has joined #openstack-swift03:24
zaitcev"I have just increased partition power openstack swift. But after incrementing the partition power, the objects not found. I can not download previously existing objects in the cluster. Is there a way to fix this and let me know the best way to increase partition power."03:25
zaitcev;_;03:25
*** chlong__ has joined #openstack-swift03:28
*** chlong_ has quit IRC03:28
notmynamezaitcev: ouch03:37
notmynameI'm assuming that's separate from the functionality that will be in the next release?03:37
zaitcevhttps://ask.openstack.org/en/question/107998/object-not-found-when-increase-partition-power-openstack-swift/03:38
notmynamezaitcev: I posted an answer03:47
*** gkadam has joined #openstack-swift03:50
*** xrb_ has quit IRC04:01
*** chlong_ has joined #openstack-swift04:01
*** chlong__ has quit IRC04:01
*** skudlik has joined #openstack-swift04:16
*** chlong__ has joined #openstack-swift04:25
*** chlong_ has quit IRC04:27
*** links has joined #openstack-swift04:30
*** frank_young has quit IRC04:48
*** tonanhngo has quit IRC04:56
*** frank_young has joined #openstack-swift04:58
*** tonanhngo has joined #openstack-swift04:59
*** frank_young has quit IRC05:03
*** tonanhngo has quit IRC05:05
*** frank_young has joined #openstack-swift05:11
*** Venkata has joined #openstack-swift05:12
VenkataHello All, I have a question on public-read ACL of s3 plugin, Generally S3 supports wget download on obj if it is public-read. but s3plugin is not allowing that .05:13
*** frank_young has quit IRC05:16
*** itlinux has quit IRC05:31
*** chlong_ has joined #openstack-swift05:32
*** chlong__ has quit IRC05:34
*** psachin has joined #openstack-swift05:36
*** kiennt has quit IRC05:46
*** cshastri has joined #openstack-swift05:50
*** skudlik has quit IRC05:51
hugokuoAnyone knows what factor would result in this for EC05:59
hugokuo-rw------- 1 swift swift 2.2M Jul 18 13:19 1500409132.72229#2#d.data05:59
hugokuo-rw------- 1 swift swift 2.2M Jul 18 13:20 1500409214.72526#2.data05:59
hugokuoIn the same partiton/suffix06:00
hugokuoThe one w/o #d is invalid data? Is it safe to delete it?06:00
*** kiennt has joined #openstack-swift06:03
*** rcernin has joined #openstack-swift06:10
*** bkopilov has quit IRC06:24
*** xrb has joined #openstack-swift06:26
*** xrb has quit IRC06:28
*** xrb has joined #openstack-swift06:28
openstackgerritPavel Kvasnička proposed openstack/swift master: IDEA: Handle 412 response on expirer when object updated  https://review.openstack.org/48659906:33
*** ukaynar has quit IRC06:36
*** kiennt has quit IRC06:41
*** klrmn has quit IRC06:50
*** noxdafox has quit IRC06:55
*** winggundamth has quit IRC06:55
*** ChubYann has quit IRC06:56
*** noxdafox has joined #openstack-swift07:01
*** kiennt has joined #openstack-swift07:23
*** skudlik has joined #openstack-swift07:25
*** gkadam is now known as gkadam-afk07:31
*** oshritf has joined #openstack-swift07:38
acolesgood morning07:46
acolestorgomatic: I don't know why diskfile is using logging module but I agree it would be better to use the configured logger07:57
mathiasbgood morning08:00
acoleshugokuo: hugokuo the more recent of those two .data has not been made durable, meaning that the object server did not get confirmation that sufficient fragments were written to the cluster for the object to be rebuilt. That may be because either 1. sufficient fragments really were not written, or 2. because the confirmation did not make it to that particular object server.08:14
acoleshugokuo: in case 1 the reconstructor will eventually clean up the newer non-durable .data file. In case 2 the reconstructor will eventually make that newer .data durable and remove the older one.08:15
acoleshugokuo: so in general you should not assume that an EC .data file without the #d in its filename is invalid (the data can be ok it just lack the durable marker) not that it is ok to delete (the object may be durable and the reconstructor will mark this fragment as durable eventually).08:17
acoleshugokuo: BTW a single or even several .data without #d durable markers  does not prevent an object being read - so long as there are sufficient fragments to rebuild then the object may be read.08:18
*** oshritf has quit IRC08:23
*** bkeller`_ has quit IRC08:33
*** redbo has quit IRC08:34
*** jrichli has quit IRC08:36
*** bkeller` has joined #openstack-swift08:38
*** jrichli has joined #openstack-swift08:38
*** geaaru has joined #openstack-swift08:42
*** oshritf has joined #openstack-swift08:42
*** gkadam-afk is now known as gkadam08:46
*** redbo has joined #openstack-swift08:46
*** ChanServ sets mode: +v redbo08:46
xrbnotmyname & timur: thanks for the info09:13
*** Venkata has quit IRC09:19
*** mvk has quit IRC09:25
*** chlong__ has joined #openstack-swift09:33
*** chlong_ has quit IRC09:35
*** mvk has joined #openstack-swift09:54
*** tovin07_ has quit IRC09:58
*** kiennt has quit IRC10:04
*** skudlik has quit IRC10:04
*** openstackgerrit has quit IRC10:17
*** sanchitmalhotra1 has joined #openstack-swift10:27
*** sanchitmalhotra has quit IRC10:28
*** sanchitmalhotra1 is now known as sanchitmalhotra10:28
*** kei_yama has quit IRC10:35
*** skudlik has joined #openstack-swift10:43
*** chlong_ has joined #openstack-swift10:51
*** chlong__ has quit IRC10:53
*** bkopilov_ has quit IRC10:55
*** chlong__ has joined #openstack-swift10:56
*** chlong_ has quit IRC10:59
*** JimCheung has joined #openstack-swift11:15
*** JimCheung has quit IRC11:20
*** openstackgerrit has joined #openstack-swift11:34
openstackgerritjunbo.li proposed openstack/swift master: Unsupported 'message' Exception attribute in PY3  https://review.openstack.org/48701111:34
*** chlong_ has joined #openstack-swift11:45
*** chlong__ has quit IRC11:45
*** chlong_ has quit IRC11:58
*** ChanServ sets mode: +v jrichli12:09
*** links has quit IRC12:27
*** MVenesio has joined #openstack-swift12:28
*** catintheroof has joined #openstack-swift12:39
*** catintheroof has quit IRC12:39
*** catintheroof has joined #openstack-swift12:39
*** links has joined #openstack-swift12:43
*** SkyRocknRoll has joined #openstack-swift12:58
*** gkadam has quit IRC13:02
*** lucasxu has joined #openstack-swift13:05
openstackgerritjunbo.li proposed openstack/swift master: Unsupported 'message' Exception attribute in PY3  https://review.openstack.org/48701113:26
*** rsFF has left #openstack-swift13:26
*** SkyRocknRoll has quit IRC13:27
*** xrb_ has joined #openstack-swift13:34
*** xrb has quit IRC13:35
*** stewie925 has joined #openstack-swift13:35
*** chlong has joined #openstack-swift13:37
*** dancn has quit IRC13:37
*** bkopilov_ has joined #openstack-swift13:42
*** silor has joined #openstack-swift13:57
*** chlong has quit IRC14:03
*** silor has quit IRC14:03
*** vint_bra has joined #openstack-swift14:19
*** links has quit IRC14:26
*** lucasxu has quit IRC14:38
*** cshastri has quit IRC14:39
*** psachin has quit IRC14:40
*** ukaynar has joined #openstack-swift14:55
*** lifeless has quit IRC15:09
*** tovin07 has joined #openstack-swift15:13
*** chlong has joined #openstack-swift15:15
*** rcernin has quit IRC15:17
*** lifeless has joined #openstack-swift15:22
*** ukaynar has quit IRC15:24
*** ukaynar has joined #openstack-swift15:24
*** SkyRocknRoll has joined #openstack-swift15:28
*** lifeless has quit IRC15:29
*** hseipp has joined #openstack-swift15:31
*** Renich has quit IRC15:33
*** lifeless has joined #openstack-swift15:36
*** sanchitmalhotra1 has joined #openstack-swift15:40
*** sanchitmalhotra has quit IRC15:40
*** sanchitmalhotra1 is now known as sanchitmalhotra15:40
*** MVenesio has quit IRC15:42
*** skudlik has quit IRC15:43
*** jistr is now known as jistr|mtg15:58
tdasilvanice: Is possible to "Swift-ize" Neutron to be a standalone Multi-SDN solution in cloud adoption?16:00
tdasilvathe moment swift becomes a verb ;)16:00
*** dr_gogeta86_ has joined #openstack-swift16:03
*** hseipp has quit IRC16:06
*** dr_gogeta86 has quit IRC16:06
*** hseipp has joined #openstack-swift16:06
*** rcernin has joined #openstack-swift16:14
*** tovin07 has left #openstack-swift16:15
*** charz has quit IRC16:16
timburkegood morning16:16
*** charz has joined #openstack-swift16:18
*** jistr|mtg is now known as jistr16:19
*** JimCheung has joined #openstack-swift16:21
*** JimCheung has quit IRC16:21
*** JimCheung has joined #openstack-swift16:22
*** lucasxu has joined #openstack-swift16:26
*** hseipp has quit IRC16:26
*** lucasxu has quit IRC16:28
*** xrb_ has quit IRC16:33
*** nolwenn has joined #openstack-swift16:47
nolwenn@all hello, I have a lot of client disconnects with error "invalid literal for int() with base 16"16:51
nolwennIs it saying something to someone?16:52
timburkenolwenn: sounds like a client that isn't doing chunked transfers correctly to me16:53
*** SkyRocknRoll_ has joined #openstack-swift16:54
timburkeor maybe they hit https://github.com/requests/requests/issues/1648 ?16:54
*** lucasxu has joined #openstack-swift16:54
*** links has joined #openstack-swift16:55
*** SkyRocknRoll has quit IRC16:57
nolwenn@timburke it's possible. I had seen this too : https://bugs.launchpad.net/swift/+bug/66795617:00
openstackLaunchpad bug 667956 in OpenStack Object Storage (swift) "chunked transfer client disconnect causes unhandled value error in object server" [Low,Fix released]17:00
*** mvk has quit IRC17:09
notmynamegood morning17:13
* notmyname has been here a while, just on phone calls17:13
timburkenolwenn: yeah... swift used to clutter logs with tracebacks when there was an error like that. i *think* all of those should be fixed now? but it's come up as recently as https://github.com/openstack/swift/commit/08ac32e17:13
*** ChubYann has joined #openstack-swift17:15
openstackgerritTim Burke proposed openstack/swift master: Nudge users to install extras on ImportErrors in kms_keymanager  https://review.openstack.org/48714517:18
nolwenntimburke : yes but my problem is in proxy server17:22
nolwenn``` in _chunked_read#012 self.chunk_length = int(rfile.readline().split(b";", 1)[0], 16)#012ValueError: invalid literal for int() with base 16: '' ```17:22
timburkenolwenn: what version of swift? mind dropping the full traceback somewhere?17:22
openstackgerritAlistair Coles proposed openstack/swift master: Add multiple worker processes strategy to reconstructor  https://review.openstack.org/47841617:24
nolwenntimburke I send you. I have an very old version : juno :-)17:25
*** klrmn has joined #openstack-swift17:25
*** links has quit IRC17:31
acolestimburke: note fresh version of patch 478416 - needed to fix one issue, plus rejigged some tests. thanks for your additions.17:35
patchbothttps://review.openstack.org/#/c/478416/ - swift - Add multiple worker processes strategy to reconstr...17:35
timburkeacoles: looking now. thanks for that spot; i think i maybe just hadn't put enough objects to notice that i was running on all devices17:35
acolestimburke: it's a little non-intuitive - I think at one point I broke it to have the kwargs look clean of any bogus devices, then put it back, then it slipped again with the squashes :)17:38
acolestimburke: I played with trying to avoid the bogus override device having stats dumped but decided that can wait. IDK if it is noise, harmless or possibly useful as a heads-up that you fat fingered override devices, but in end favored less churn at this point17:39
*** mvk has joined #openstack-swift17:41
timburkeyeah, that's fine. it'll get cleaned up in aggregation, and won't actually *affect* aggregation17:53
timburkei wonder if we could *always* pass in override_devices, so we could treat empty as "do nothing"...17:56
acolestimburke: maybe, but maybe as a follow-up ;)18:03
timburkeyeah, if we want it in for this release (and i think we do -- it's a good improvement), i think it'd be best to iterate later rather than make it perfect now18:06
tdasilvamathiasb: just fyi...when you remove /etc/swift/test.conf, the func tests will run in IN-PROCESS mode, so I'm pretty sure it is not creating new users in keystone, it's not even running the tests against the running swift cluster18:06
mathiasbtdasilva: good to know! i'll have to take another look then18:13
openstackgerritAlistair Coles proposed openstack/swift master: WIP try synchronous recon aggregation  https://review.openstack.org/48716818:33
*** dhellmann has quit IRC18:45
*** itlinux_ has joined #openstack-swift18:46
*** dhellmann has joined #openstack-swift18:47
*** xrb has joined #openstack-swift18:58
*** dhellmann has quit IRC18:58
*** dhellmann has joined #openstack-swift18:58
*** xrb has quit IRC18:59
*** nolwenn has quit IRC19:00
*** nolwenn has joined #openstack-swift19:01
*** nolwenn has quit IRC19:01
*** xrb has joined #openstack-swift19:20
*** SkyRocknRoll_ has quit IRC19:31
*** JimCheung has quit IRC19:38
*** xrb has quit IRC19:40
*** oshritf has quit IRC19:50
*** nolwenn has joined #openstack-swift19:50
*** JimCheung has joined #openstack-swift19:53
*** mtreinish has quit IRC19:55
*** JimCheung has quit IRC19:58
*** noxdafox has quit IRC20:05
*** JimCheung has joined #openstack-swift20:10
*** mtreinish has joined #openstack-swift20:10
*** noxdafox has joined #openstack-swift20:12
*** vinsh has quit IRC20:13
*** JimCheung has quit IRC20:14
*** JimCheung has joined #openstack-swift20:39
*** geaaru has quit IRC21:08
*** cschwede_ has quit IRC21:08
*** lucasxu has quit IRC21:09
*** catintheroof has quit IRC21:11
openstackgerritThiago da Silva proposed openstack/swift master: ready yet? nope, please wait!  https://review.openstack.org/48674521:12
*** JimCheung has quit IRC21:13
*** JimCheung has joined #openstack-swift21:13
openstackgerritAlistair Coles proposed openstack/swift master: fix run forever mode to ignore cli overrides  https://review.openstack.org/48721821:15
notmynametdasilva: it's an emotional journey reading the commit subject on https://review.openstack.org/#/c/486745/21:16
patchbotpatch 486745 - swift - ready yet? nope, please wait!21:16
tdasilvalol21:16
notmynametdasilva: this is to fix the probe tests? you found something after timburke's moving of stdout?21:16
tdasilvanotmyname: yep21:17
*** JimCheung has quit IRC21:18
tdasilvanotmyname: probe tests were still failing for me on master on that spot21:18
notmynamebased on the commit message, I'm assuming you've got more you want to do with it?21:18
tdasilvanotmyname: no, that should do it21:18
notmynameis it something that you think can land this week for the release? should it block a release?21:18
tdasilvai think it's ready to merge :)21:18
notmynameoh, ok then. that sounds good21:18
tdasilva+2 +A21:18
*** JimCheung has joined #openstack-swift21:19
tdasilvanotmyname: not sure it's obvious, but then problem with the probe tests is that it keeps checking the server right after the restart, so we need to signal to wait ;)21:20
notmynameok21:20
tdasilvaso that commit msg was my attempt at making a joke21:20
notmynameyeah, I was trying to load it back in my brain. I looked at that briefly with timburke21:20
tdasilvawhich i'm pretty terrible at21:20
notmynameheh21:20
openstackgerritAlistair Coles proposed openstack/swift master: fix run forever mode to ignore cli overrides  https://review.openstack.org/48721821:21
acolestimburke: I think we need something like that ^^21:22
acolestimburke: I'm not squashing it right now because that may confuse my earlier update on the parent patch, and because I'm too tired to spot any silly mistakes!21:23
acolestimburke: if you think it is ok then please squash into parent21:23
timburkeacoles: thanks, will do!21:23
notmynametdasilva: I'll see what the community cluster does with it.21:27
tdasilvanotmyname: cool!21:27
acolesnotmyname: do we have 0700 meeting tomorrow? do we have a chair?21:27
notmynameyes, it's scheduled. kota_ is the chair. I'm hoping to chat with him when he gets on21:27
tdasilvanotmyname: if anybody has a chance, it would be great to see this go in: https://review.openstack.org/#/c/344693/21:27
patchbotpatch 344693 - swift - Fix swiftdir option and usage of storage policy al...21:27
acolesnotmyname: ok I'll set my alarm !21:28
notmynametdasilva: I see 3 authors and 2 +2s already :-)21:28
tdasilvanotmyname: yeah, i kinda updated for Christian so we are both hesistant to +A, so there's really just the +2 from zaitcev21:32
zaitcevI prefer explicit arguments, easier to follow IMHO than ordering-dependent global setting... We learned that back when we programmed in Fortran. But it wasn't too offensive in case of 344693.21:34
*** mtreinish has quit IRC21:42
*** mtreinish has joined #openstack-swift21:46
*** JimCheung has quit IRC21:47
*** JimCheung has joined #openstack-swift21:47
notmynametdasilva: done!21:49
*** JimCheung has quit IRC22:43
*** JimCheung has joined #openstack-swift22:43
*** JimCheung has quit IRC22:51
*** JimCheung has joined #openstack-swift22:51
*** JimCheung has quit IRC22:56
*** rcernin has quit IRC22:58
openstackgerritMerged openstack/swift master: Fix swiftdir option and usage of storage policy aliases  https://review.openstack.org/34469323:15
openstackgerritJohn Dickinson proposed openstack/swift master: 2.15.0 authors and changelog update  https://review.openstack.org/48583923:21
*** chlong has quit IRC23:32
*** kei_yama has joined #openstack-swift23:33
openstackgerritOpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements  https://review.openstack.org/8873623:33
*** deep-book-gk_ has joined #openstack-swift23:35
*** deep-book-gk_ has left #openstack-swift23:35
*** tonanhngo has joined #openstack-swift23:47
*** JimCheung has joined #openstack-swift23:47

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