Friday, 2014-08-22

*** annegentle has joined #openstack-swift00:14
*** acoles_away has joined #openstack-swift00:17
*** acoles_away is now known as acoles00:17
*** ChanServ sets mode: +v acoles00:17
*** dmorita has joined #openstack-swift00:29
*** aswadr has joined #openstack-swift00:31
*** mwstorer has quit IRC00:52
*** nosnos has joined #openstack-swift00:59
*** annegentle has quit IRC01:02
*** addnull has joined #openstack-swift01:07
*** nexusz99 has joined #openstack-swift01:10
*** annegentle has joined #openstack-swift01:12
*** abhirc has joined #openstack-swift01:17
*** haomaiwang has joined #openstack-swift01:38
*** addnull has quit IRC01:41
*** annegentle has quit IRC01:42
*** addnull has joined #openstack-swift01:44
openstackgerritThiago da Silva proposed a change to openstack/swift: moving object validation checks to top of PUT method  https://review.openstack.org/11599501:48
openstackgerritA change was merged to openstack/swift: Remove fake _get_part and use the real thing  https://review.openstack.org/10396201:50
*** sas_ has joined #openstack-swift01:54
*** zul has joined #openstack-swift01:55
*** nosnos has quit IRC02:03
*** nosnos has joined #openstack-swift02:04
*** zul has quit IRC02:05
*** nosnos_ has joined #openstack-swift02:08
*** nosnos has quit IRC02:08
*** Edward-Zhang has joined #openstack-swift02:12
*** bill_az_ has joined #openstack-swift02:38
*** annegentle has joined #openstack-swift02:50
*** nosnos_ has quit IRC02:56
*** acoles has quit IRC03:15
*** gyee has quit IRC03:24
*** acoles_away has joined #openstack-swift03:25
*** acoles_away is now known as acoles03:25
*** ChanServ sets mode: +v acoles03:25
openstackgerritOpenStack Proposal Bot proposed a change to openstack/python-swiftclient: Updated from global requirements  https://review.openstack.org/8925003:29
openstackgerritOpenStack Proposal Bot proposed a change to openstack/swift: Updated from global requirements  https://review.openstack.org/8873603:29
*** bill_az_ has quit IRC03:36
*** nosnos has joined #openstack-swift03:40
*** annegentle has quit IRC03:52
*** addnull has quit IRC03:56
*** nexusz99 has quit IRC03:58
*** Edward-Zhang has quit IRC04:02
*** mrsnivvel has joined #openstack-swift04:15
*** annegentle has joined #openstack-swift04:16
*** sas_ has quit IRC04:18
*** addnull has joined #openstack-swift04:26
*** addnull has quit IRC04:30
*** tsg has quit IRC04:33
*** cebruns has quit IRC04:33
*** tgohad has joined #openstack-swift04:33
*** cebruns has joined #openstack-swift04:35
*** ppai has joined #openstack-swift04:38
*** Edward-Zhang has joined #openstack-swift04:47
*** annegentle has quit IRC04:50
*** tgohad has quit IRC05:01
*** haomaiw__ has joined #openstack-swift05:02
*** shakamunyi has quit IRC05:03
*** haomaiwang has quit IRC05:06
*** nexusz99 has joined #openstack-swift05:12
*** sas_ has joined #openstack-swift05:16
*** sas_ has quit IRC05:22
*** addnull has joined #openstack-swift05:24
*** tsg has joined #openstack-swift05:25
*** annegentle has joined #openstack-swift05:26
*** kopparam has joined #openstack-swift05:35
*** nshaikh has joined #openstack-swift05:49
*** annegentle has quit IRC06:04
*** annegentle has joined #openstack-swift06:05
*** chandankumar has joined #openstack-swift06:09
*** annegentle has quit IRC06:10
*** shakamunyi has joined #openstack-swift06:29
*** tsg has quit IRC06:31
*** shakamunyi has quit IRC06:34
*** k4n0 has joined #openstack-swift06:35
*** annegentle has joined #openstack-swift06:35
*** annegentle has quit IRC06:42
*** haomaiw__ has quit IRC06:58
*** haomaiwa_ has joined #openstack-swift06:59
*** bkopilov has quit IRC07:01
*** bkopilov has joined #openstack-swift07:03
*** sungju has quit IRC07:18
openstackgerritA change was merged to openstack/swift: Respect device weights when adding replicas  https://review.openstack.org/11541607:18
*** shakamunyi has joined #openstack-swift07:30
*** shakamunyi has quit IRC07:34
ByteSorenotmyname: ping07:39
*** joeljwright has joined #openstack-swift08:15
*** chandankumar has quit IRC08:19
*** shakamunyi has joined #openstack-swift08:30
*** homegrown has joined #openstack-swift08:31
*** shakamunyi has quit IRC08:35
*** annegentle has joined #openstack-swift08:37
openstackgerritTakashi Kajinami proposed a change to openstack/swift: Add process name chacking into swift-init  https://review.openstack.org/11620308:43
*** annegentle has quit IRC08:43
openstackgerritTakashi Kajinami proposed a change to openstack/swift: Add process name checking into swift-init  https://review.openstack.org/11620308:45
*** Edward-Zhang has quit IRC08:48
openstackgerritTakashi Kajinami proposed a change to openstack/swift: Add process name checking into swift-init  https://review.openstack.org/11620308:49
*** shakamunyi has joined #openstack-swift08:59
*** mkollaro has joined #openstack-swift09:07
*** shakamunyi has quit IRC09:12
*** nexusz99 has quit IRC09:12
*** aix has joined #openstack-swift09:14
*** kopparam has quit IRC09:21
*** kopparam has joined #openstack-swift09:22
*** chandankumar has joined #openstack-swift09:25
*** kopparam has quit IRC09:27
*** annegentle has joined #openstack-swift09:37
*** annegentle has quit IRC09:42
*** addnull has quit IRC10:03
*** shakamunyi has joined #openstack-swift10:08
*** shakamunyi has quit IRC10:14
*** ppai has quit IRC10:25
*** ppai has joined #openstack-swift10:28
*** annegentle has joined #openstack-swift10:37
*** annegentle has quit IRC10:42
openstackgerritTakashi Kajinami proposed a change to openstack/swift: Fix connection timeout checking in db_replicator  https://review.openstack.org/11621810:42
*** dmorita has quit IRC11:02
*** nosnos has quit IRC11:17
*** nosnos has joined #openstack-swift11:17
*** nosnos has quit IRC11:21
*** a1|away is now known as JelleB11:25
*** ppai has quit IRC11:26
*** haomaiwa_ has quit IRC11:30
*** geaaru has joined #openstack-swift11:35
*** annegentle has joined #openstack-swift11:37
*** annegentle has quit IRC11:42
*** zul has joined #openstack-swift12:08
*** elambert has quit IRC12:11
*** elambert has joined #openstack-swift12:13
openstackgerritOpenStack Proposal Bot proposed a change to openstack/python-swiftclient: Updated from global requirements  https://review.openstack.org/8925012:34
openstackgerritOpenStack Proposal Bot proposed a change to openstack/swift: Updated from global requirements  https://review.openstack.org/8873612:34
*** annegentle has joined #openstack-swift12:37
openstackgerritA change was merged to openstack/swift: Let admins add a region without melting their cluster  https://review.openstack.org/11544112:38
*** zul has quit IRC12:40
*** annegentle has quit IRC12:42
*** bill_az_ has joined #openstack-swift12:54
*** Midnightmyth has joined #openstack-swift13:03
*** venturi has joined #openstack-swift13:15
*** zul has joined #openstack-swift13:15
*** annegentle has joined #openstack-swift13:17
venturiquestion. what are possible consecuences when storage clusters are not ntp synhronized? is there any bigger problem for Swift to stop working or falls into strange operation?13:20
*** mrsnivvel has quit IRC13:23
*** aswadr has quit IRC13:32
*** dmsimard_away is now known as dmsimard13:35
*** annegentle has quit IRC13:36
tdasilvaventuri: I'm new here, so take what I say with a grain of salt: From what I have seen of the code, there are many operations that are time sync dependent. For example, object_expiration and according to this page (http://docs.openstack.org/openstack-ops/content/network_design.html) object replication13:46
tdasilvaventuri: so I think it would not "stop working", but i would not be surprise if there were strange operations happening13:47
venturitdasilva: ok thx for comment.13:49
tdasilvaanother example is that object server that checks the x-timestamp on a PUT request to make sure that a new request has a newer timestamp than an existing object13:49
tdasilvahttps://github.com/openstack/swift/blob/master/swift/obj/server.py#L40913:49
*** mahatic has joined #openstack-swift13:49
*** venturi has quit IRC13:50
*** k4n0 has quit IRC13:55
*** marcusvrn has joined #openstack-swift14:20
*** annegentle has joined #openstack-swift14:30
*** shakamunyi has joined #openstack-swift14:30
*** mtreinish is now known as mriedem-dplgnr14:35
*** chandankumar has quit IRC14:36
*** zslot has quit IRC14:43
*** sandywalsh has quit IRC14:50
*** nshaikh has quit IRC14:51
*** pberis has quit IRC14:54
*** Midnightmyth has quit IRC15:03
*** tongli has joined #openstack-swift15:07
*** maelfius has joined #openstack-swift15:13
*** Midnightmyth has joined #openstack-swift15:13
*** tongli has quit IRC15:14
*** morganfainberg has quit IRC15:14
*** maelfius is now known as morganfainberg15:14
*** mkollaro has quit IRC15:15
*** tsg has joined #openstack-swift15:16
notmynamegood morning, world15:17
notmynamereminder that patches that are to be included in 2.1.0 need to be approved today.15:27
*** mkollaro has joined #openstack-swift15:33
*** annegentle has quit IRC15:36
*** annegentle has joined #openstack-swift15:41
*** shakamunyi has quit IRC15:41
*** annegent_ has joined #openstack-swift15:45
*** atan8 has joined #openstack-swift15:45
*** annegentle has quit IRC15:46
*** pberis has joined #openstack-swift15:46
*** annegent_ has quit IRC15:46
*** annegentle has joined #openstack-swift15:48
*** nexusz99 has joined #openstack-swift15:55
*** Midnightmyth has quit IRC16:03
*** annegentle has quit IRC16:06
*** atan8 has quit IRC16:06
*** Midnightmyth has joined #openstack-swift16:14
*** nexusz99 has quit IRC16:14
*** Dieterbe has quit IRC16:20
*** Dieterbe has joined #openstack-swift16:21
*** Dieterbe has quit IRC16:21
*** Dieterbe has joined #openstack-swift16:21
*** mwstorer has joined #openstack-swift16:23
*** chandankumar has joined #openstack-swift16:32
*** openstackgerrit has quit IRC16:34
*** homegrown has left #openstack-swift16:35
*** geaaru has quit IRC17:06
notmynamedoes anyone remember if account_autocreate is required for keystone?17:08
notmynameI guess it's a convenience, not a requirement. right?17:08
*** mkollaro has quit IRC17:09
torgomaticnotmyname: yeah, the account holder can always PUT the account themselves, but it'd be nice if they didn't have to17:10
notmynametorgomatic: well, the reseller admin. you can't PUT your own account :-)17:10
torgomaticnotmyname: no? I thought I could, to set metadata. (blah blah POST blah blah)17:10
notmynamealso, as I've been looking in to these things this week, any reason we shouldn't change the default account_autocreate value to true17:11
notmynamenot that I'm looking to change defaults, but it seems pretty useful. I'm just wondering if it would cause issues for people17:12
*** mkollaro has joined #openstack-swift17:12
torgomaticthis is one of those discussions that has the potential to turn into a truly massive bikeshed discussion17:17
torgomaticthe kind that our grandchildrens' grandchildren will still be not caring about 100 years hence17:17
torgomaticbut I, personally, have no problem flipping that to true17:18
*** gyee has joined #openstack-swift17:20
swifterdarrellnotmyname: I'd vote to not change it unless there's some tremendously compelling reason17:20
*** atan8 has joined #openstack-swift17:22
notmynameswifterdarrell: the only reason I can think of is to make it easier to start using faster after an install17:23
*** swills has quit IRC17:30
*** openstackgerrit has joined #openstack-swift17:31
*** zul has quit IRC17:33
*** swills has joined #openstack-swift17:35
*** swills has quit IRC17:38
*** swills has joined #openstack-swift17:38
*** tgohad has joined #openstack-swift17:40
*** tsg has quit IRC17:40
*** cebruns has quit IRC17:41
*** peluse has quit IRC17:41
*** cebruns has joined #openstack-swift17:41
*** peluse has joined #openstack-swift17:42
*** swills has quit IRC17:44
*** swills has joined #openstack-swift17:45
*** swills has quit IRC17:45
*** swills has joined #openstack-swift17:45
*** tgohad has quit IRC18:01
*** tgohad has joined #openstack-swift18:01
*** tongli has joined #openstack-swift18:05
*** Midnightmyth has quit IRC18:09
*** annegentle has joined #openstack-swift18:25
*** chandankumar has quit IRC18:28
claygnotmyname: I'm not sure I agree with the idea that not having to set the value in the config to what you want gets you going quicker - I wouldn't be against making it *required* (like bind_port)18:35
claygwell, bind_port *isn't* required - but it's the reason we can't change the default - because we don't make people acctually pick the values you want18:36
claygyou're gunna have to copy and paste an example config anyway; maybe there's a middle road in having the example configs just have more of the options we want explicitly defined... defined expicitly18:37
claygnotmyname: had a question on swiftservices patch RE: adding new depends on futures>=2.1.318:38
*** tongli has quit IRC18:38
*** shri has joined #openstack-swift18:38
* clayg feels so alone right now18:39
*** zul has joined #openstack-swift18:46
*** mahatic has quit IRC18:49
claygwhat is the right way to change the status of https://bugs.launchpad.net/swift/+bug/1359664 to "this is a feature request that includes a patch, but author doesn't want to gerrit?"  - cause we should totally have a status for that, it's happened twice this week :D18:49
*** annegentle has quit IRC18:59
*** annegentle has joined #openstack-swift18:59
*** achhabra has joined #openstack-swift19:15
*** achhabra_ has joined #openstack-swift19:16
*** mkollaro has quit IRC19:24
notmynameclayg: ya, the swiftservices thing is on my todo today :-)19:24
* notmyname doesn't want clayg to be alone19:24
*** morganfainberg is now known as morganfainberg_Z19:35
notmynameclayg: did you get the futures depend working on your precise box?19:38
*** tgohad__ has joined #openstack-swift19:51
*** tgohad has quit IRC19:55
claygnotmyname: yeah futures ended up working just fine - but I never did get to a point of "understanding" the problem20:07
notmynameclayg: I was just looking at the futures doc to see what the new feature is in this patch that can't be used with the precise version. but maybe he just added that version because that's what in global requirements20:08
claygnotmyname: probably the latter (almost definately)20:08
notmynameok20:09
*** tgohad__ has quit IRC20:13
*** peluse has quit IRC20:14
*** tgohad__ has joined #openstack-swift20:14
*** peluse has joined #openstack-swift20:14
*** cebruns has quit IRC20:14
*** cebruns has joined #openstack-swift20:15
*** fifieldt has quit IRC20:23
*** gyee has quit IRC20:28
*** annegentle has quit IRC20:30
*** joeljwright has quit IRC20:30
*** gyee has joined #openstack-swift20:35
*** fifieldt has joined #openstack-swift20:36
*** annegentle has joined #openstack-swift20:55
*** IRTermite has quit IRC21:00
*** zul has quit IRC21:01
*** atan8 has quit IRC21:10
claygnotmyname: yeah so my question was more about - hey we're adding a new depends; you'll have to package it on most distros (even some that aren't unforgivably old) so... what should happen when we want to do that21:10
notmynameright.21:10
claygnotmyname: I thought it might be as simple as something in the commit message; or as complex as an email to ML asking packagers to chime in about this particular package in their particular situation21:11
notmynameso if we had our druthers, then we'd use the version packaged in precise, right?21:11
claygnotmyname:  I could cirtainly imagine someone saysing well can you USE 2.1.3 since that's what already ships with distro X21:11
notmynameie we'd have a version specifier that complies with global reqs but works with others too. like we do with many of them in swift21:11
claygnotmyname: yeah I think for swiftstack we'd be pretty keen on that; I think RAX is all swiftly all the time these days21:11
claygnotmyname: yeah, and I guess that's still a point of contention - but it used to be something we'd think about21:12
notmynameit's actually the same question about eventlet in swift. https://review.openstack.org/#/c/110815/21:12
claygnotmyname: maybe we'd already given up this battle on python-swiftclient but my brain is stuck in the old world where packgers matter21:12
notmynameso now with 2 actual patches doing this, it's probably time for a conversation with -infra/packagers/ML21:13
*** brnelson has joined #openstack-swift21:13
claygweird py26 test failure on that one :\21:14
notmynameI always get suspicious when the failed test has the word "timings" in it21:15
*** bnelson has quit IRC21:16
*** brnelson has quit IRC21:18
notmynamepeluse: if https://review.openstack.org/#/c/75941/ lands today it will be in 2.1.021:20
notmynametorgomatic: ^ can you look a peluse's question on that patch?21:21
torgomaticnotmyname: yeah, one sec... working on some parser thingy21:21
notmynametorgomatic: thanks21:21
pelusenotmyname:  cool21:21
peluseFYI, going to do the regular weekly master-->feature EC merge here shortly21:22
notmynamepeluse: If there's any other pending stuff for master you're interested in, I'd appreciate if you do that first. the ec merge doesn't affect 2.1.0, and a pending patch on master may21:24
pelusenotmyname:  do which first?21:24
peluseoh, duh, any other master stuff.  sorry :)21:24
claygnotmyname: that patch should really have an explict pool with a tunable concurrency option :\21:25
claygdoes anyone know what the cardinality of the native eventlet pool is?21:25
claygoh, i guess it's just doing the three replicas in parallel and waits on all of them before returning...21:26
notmynamethree patches with an existing +2:21:26
notmynameaccount-to-account copy: https://review.openstack.org/#/c/72157/21:26
notmynamemulti-updater: https://review.openstack.org/#/c/75941/21:26
notmynamekeystone v3 cross-tenant ids to acls: https://review.openstack.org/#/c/86430/21:27
notmynamethe last one IMO isnt' as important for this release, but it is important to land in juno21:27
peluseI've reviewed the 2nd two and honestly don't feel comfortable reviewing the 1st one w/o some more dedicated time21:29
notmynamemakes sense. they are big patches21:30
torgomaticnotmyname: peluse: question answered21:37
notmynametorgomatic: thanks21:37
pelusetorgomatic:  yup, thanks21:38
*** annegentle has quit IRC21:38
peluseclayg:  did you want to push on the explicit pool w/tunable option thingy or are you good if I OK it as is?21:38
*** annegentle has joined #openstack-swift21:39
torgomaticit's moving concurrency from 1 to 3... personally I'm not worried about it, especially since the default GreenPool size is 100021:41
*** occupant has quit IRC21:41
torgomaticnot that you asked me :)21:41
pelusetorgomatic:  heh, only because he made the comment above but yeah I see your point - thanks!21:42
*** occupant has joined #openstack-swift21:42
*** mriedem-dplgnr is now known as mtreinish21:43
openstackgerritpaul luse proposed a change to openstack/swift: Merge master to feature/ec  https://review.openstack.org/11639421:44
openstackgerritJohn Dickinson proposed a change to openstack/swift: authors and changelog updates for 2.1.0 release  https://review.openstack.org/11516721:49
notmynameclayg: torgomatic: peluse: ^^ should now include all the good stuff, including the two recent patches that landed (or will soon land)21:50
torgomaticsuch optimism21:50
notmynameheh21:50
torgomaticnotmyname: do I not understand how mailmap works, or are those new entries all entirely pointless?21:51
notmynametorgomatic: I'd like your opinion on the changelog entry about the ring builder changes21:51
notmynametorgomatic: let me double check the .mailmap21:51
torgomaticthey all seem to be of the form Real Name <X> <X>21:51
notmynametorgomatic: ah. those are necessary21:51
torgomaticnotmyname: what do they do?21:52
notmynamethey are of the form "name x y". which means, any time you see the email "y", set the name and email to "name x"21:52
torgomaticnotmyname: did you just group the ring stuff under "Various other minor bug fixes and improvements"? :)21:52
notmynameall of those people have made multiple commits with different name spellings (mostly capitalization)21:52
notmynameheh21:52
torgomaticnotmyname: ah, that works then21:52
notmynameya, without those mailmap changes, eg we'd have paul luse and Paul Luse in the `git shortlog` output21:53
pelusemy alter ego :)21:54
torgomaticso the first option, then, where I don't understand how mailmap works. Cool. :)21:54
notmynameya.21:54
notmynametorgomatic: nit fixed. waiting to push to se if there are other things21:55
* peluse stuck on phone for a bit...21:55
torgomaticnotmyname: nothing else from me21:55
*** achhabra has quit IRC22:04
*** achhabra_ has quit IRC22:04
*** tgohad__ has quit IRC22:06
notmynamewhoa. the gate is nearly 100 deep22:07
*** dmsimard is now known as dmsimard_away22:23
torgomaticso, release Tuesday or Wednesday then?22:26
notmynameheh22:27
notmynametorgomatic: I hope to get the SHA before I get on a plan on sunday and then send it to ttx.22:28
torgomaticnotmyname: and how many things are you waiting on at the moment?22:28
notmynamethe last one peluse approved for the multithread updater and then the authors/changelog patch (which acorwin is  eviscerating now)22:29
torgomaticyeah, I could see that landing by Sunday22:30
pelusenotmyname:  hey, do you know when they're announcing the talks selected for the summit?22:31
notmynamepeluse: I thought today, but I haven't seen anything. so, really, I have no idea22:31
pelusenotmyname:  cool, that's what I thought I remembered too but wasn't sure22:31
notmynameacorwin: thanks for the review22:32
acorwinnotmyname: my pleasure!22:32
openstackgerritJohn Dickinson proposed a change to openstack/swift: authors and changelog updates for 2.1.0 release  https://review.openstack.org/11516722:40
notmynameacorwin: torgomatic: fixed22:40
*** marcusvrn has quit IRC22:41
*** annegentle has quit IRC22:44
*** annegentle has joined #openstack-swift22:45
notmynameI gonna land the changelog update unless acorwin or peluse are plannign on looking at it22:59
acorwinnotmyname: i'm gonna skim it real quick if i can figure out what button to click23:00
notmynamekk23:00
notmynameacorwin: https://review.openstack.org/#/c/115167 <-- click the filename to get a diff there23:00
acorwinnotmyname: awesome23:00
acorwinnotmyname: can i see a diff from patch set 3 to patchset 4?23:01
acorwinnotmyname: yes i can!23:01
notmynameacorwin: you're lucky that there haven't been any merges to master between to two. otherwise you'd see all those too23:01
acorwinnotmyname: as long as you only ask me to review changelogs, i'll be fine23:01
notmynameheh23:02
acorwinnotmyname: lgtm23:02
notmynameacorwin: thanks23:02
*** annegentle has quit IRC23:05
*** atan8 has joined #openstack-swift23:24
*** atan8_ has joined #openstack-swift23:25
*** cutforth has quit IRC23:28
*** atan8 has quit IRC23:28
*** abhirc has quit IRC23:31
*** abhirc has joined #openstack-swift23:31
openstackgerritOpenStack Proposal Bot proposed a change to openstack/swift: Updated from global requirements  https://review.openstack.org/8873623:36
*** annegentle has joined #openstack-swift23:36
*** annegentle has quit IRC23:41
*** shri has quit IRC23:52
*** aswadr has joined #openstack-swift23:55

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!