Monday, 2017-01-09

*** vint_bra has quit IRC00:41
*** jamielennox is now known as jamielennox|away00:41
*** sams-gleb has joined #openstack-swift00:48
*** sams-gleb has quit IRC00:52
*** jamielennox|away is now known as jamielennox00:58
*** chosafine has joined #openstack-swift01:09
*** fbo has quit IRC01:12
*** cschwede has quit IRC01:12
*** jcook has quit IRC01:12
*** cebreidian has quit IRC01:12
*** tmoreira has quit IRC01:12
*** bkeller` has quit IRC01:12
*** timur has quit IRC01:12
*** calebb has quit IRC01:12
*** mcarden has quit IRC01:12
*** mgagne has quit IRC01:12
*** zacksh has quit IRC01:12
*** supamatt has quit IRC01:12
*** mattoliverau has quit IRC01:12
*** dims has quit IRC01:12
*** csmart has quit IRC01:12
*** ningliu has quit IRC01:12
*** darrenc has quit IRC01:12
*** csmart has joined #openstack-swift01:12
*** matt6434 has joined #openstack-swift01:12
*** bkeller` has joined #openstack-swift01:12
*** cschwede has joined #openstack-swift01:12
*** supamatt_ has joined #openstack-swift01:12
*** tmoreira has joined #openstack-swift01:12
*** timur has joined #openstack-swift01:12
*** darrenc_ has joined #openstack-swift01:12
*** zacksh has joined #openstack-swift01:12
*** cebreidian has joined #openstack-swift01:12
*** calebb has joined #openstack-swift01:13
*** ningliu has joined #openstack-swift01:13
*** supamatt_ is now known as supamatt01:13
*** mcarden has joined #openstack-swift01:13
*** Guest96631 has joined #openstack-swift01:13
*** chosafine has quit IRC01:14
*** darrenc_ is now known as darrenc01:14
*** fbo has joined #openstack-swift01:14
*** dims has joined #openstack-swift01:14
*** jcook has joined #openstack-swift01:15
*** acoles_ has quit IRC01:20
*** briancline has quit IRC01:23
*** briancurtin has quit IRC01:23
*** ogelbukh has quit IRC01:23
*** tanee has quit IRC01:23
*** mmotiani has quit IRC01:23
*** acorwin has quit IRC01:23
*** jaakkos has quit IRC01:23
*** jistr has quit IRC01:23
*** thurloat has quit IRC01:23
*** zigo has quit IRC01:23
*** rsFF has quit IRC01:23
*** amit213 has quit IRC01:23
*** kmARC has quit IRC01:23
*** blair has quit IRC01:23
*** zigo has joined #openstack-swift01:23
*** rsFF has joined #openstack-swift01:23
*** acorwin has joined #openstack-swift01:23
*** tanee has joined #openstack-swift01:23
*** thurloat has joined #openstack-swift01:23
*** briancline has joined #openstack-swift01:24
*** jistr has joined #openstack-swift01:24
*** acoles_ has joined #openstack-swift01:25
*** ChanServ sets mode: +v acoles_01:25
*** acoles_ is now known as acoles01:25
*** mmotiani has joined #openstack-swift01:25
*** kmARC has joined #openstack-swift01:28
*** briancurtin has joined #openstack-swift01:28
*** ogelbukh has joined #openstack-swift01:29
*** amit213 has joined #openstack-swift01:29
*** mrda has quit IRC01:41
*** AbyssOne has quit IRC01:45
*** mrda has joined #openstack-swift01:46
*** matt6434 is now known as mattoliverau01:48
*** ChanServ sets mode: +v mattoliverau01:48
*** AbyssOne has joined #openstack-swift01:51
*** AnxiousGarlic has quit IRC02:16
*** blair has joined #openstack-swift02:23
*** samueldmq has quit IRC02:42
*** samueldmq has joined #openstack-swift02:42
*** zigo has quit IRC02:43
*** nikivi has joined #openstack-swift02:43
*** cppforlife_ has quit IRC02:43
*** sudorandom has quit IRC02:43
*** jroll has quit IRC02:44
*** briancline has quit IRC02:45
*** calebb has quit IRC02:45
*** jroll has joined #openstack-swift02:46
*** sudorandom has joined #openstack-swift02:46
*** briancline has joined #openstack-swift02:46
*** calebb has joined #openstack-swift02:47
*** zigo has joined #openstack-swift02:48
*** sams-gleb has joined #openstack-swift02:50
*** janonymous has quit IRC02:52
*** janonymous has joined #openstack-swift02:53
*** sams-gleb has quit IRC02:54
*** david-lyle has quit IRC02:55
*** nikivi has quit IRC02:59
*** andymccr has quit IRC03:01
*** andymccr has joined #openstack-swift03:02
*** sudorandom has quit IRC03:16
*** Jeffrey4l_ has quit IRC03:18
*** sudorandom has joined #openstack-swift03:19
*** chosafine has joined #openstack-swift03:22
*** links has joined #openstack-swift04:02
*** chosafine has quit IRC04:03
*** sanchitmalhotra1 has joined #openstack-swift04:09
*** sanchitmalhotra has quit IRC04:10
*** sanchitmalhotra1 is now known as sanchitmalhotra04:10
*** sheel has joined #openstack-swift04:11
*** chosafine has joined #openstack-swift04:11
*** psachin has joined #openstack-swift04:13
*** chosafine has quit IRC04:18
*** chsc has joined #openstack-swift04:22
*** chsc has joined #openstack-swift04:22
*** chsc has quit IRC04:33
*** chsc has joined #openstack-swift04:34
*** chsc has joined #openstack-swift04:34
*** zul has quit IRC04:37
*** chsc has quit IRC04:38
*** cppforlife_ has joined #openstack-swift04:39
*** nikivi has joined #openstack-swift04:51
*** sams-gleb has joined #openstack-swift04:52
*** sams-gleb has quit IRC04:57
*** ppai has joined #openstack-swift05:38
*** ediardo has quit IRC06:32
*** ediardo has joined #openstack-swift06:32
*** hoonetorg has quit IRC06:38
*** oshritf has joined #openstack-swift06:39
*** AnxiousGarlic has joined #openstack-swift06:39
*** sams-gleb has joined #openstack-swift06:54
*** ChubYann has quit IRC06:56
*** geaaru has joined #openstack-swift06:57
*** sams-gleb has quit IRC06:59
*** rcernin has joined #openstack-swift07:08
*** tesseract has joined #openstack-swift07:12
*** nikivi- has joined #openstack-swift07:23
*** nikivi has left #openstack-swift07:24
*** sams-gleb has joined #openstack-swift07:27
*** pcaruana has joined #openstack-swift07:41
*** hseipp has joined #openstack-swift07:43
*** nikivi- has quit IRC07:43
*** nikivi has joined #openstack-swift07:44
*** d0ugal_ has quit IRC08:03
*** d0ugal has joined #openstack-swift08:03
*** d0ugal has quit IRC08:03
*** d0ugal has joined #openstack-swift08:03
*** hoonetorg has joined #openstack-swift08:07
*** rledisez has joined #openstack-swift08:08
*** openstackgerrit has quit IRC08:18
*** hoonetorg has quit IRC08:25
*** cbartz has joined #openstack-swift09:21
hugokuoA question here10:12
hugokuo ```10:12
hugokuoadd e36c400f4c1ba28a555a0de0ed4a1254 2 1483956539 24610:12
hugokuo{"status": 200, "sync_key": null, "storage_policy": "0", "meta": {}, "cors": {"allow_origin": null, "expose_headers": null, "max_age": null}, "sysmeta": {}, "read_acl": null, "object_count": "3", "write_acl": null, "versions": null, "bytes": "0"}10:12
hugokuoWhat's this information been stored in the memcached ?10:13
hugokuoby list_endpoints middleware ??10:13
*** mvk has joined #openstack-swift10:20
ahalelooks like a container stat10:21
*** jerrygb has joined #openstack-swift10:26
*** sams-gleb has quit IRC10:28
*** sams-gleb has joined #openstack-swift10:28
*** sams-gleb has quit IRC10:33
*** asettle has joined #openstack-swift10:41
*** _JZ_ has joined #openstack-swift10:44
*** sams-gleb has joined #openstack-swift10:46
*** silor has joined #openstack-swift10:48
*** vinsh has quit IRC10:50
*** jordanP has joined #openstack-swift10:58
*** jerrygb_ has joined #openstack-swift11:18
*** jerrygb has quit IRC11:21
*** silor has quit IRC11:28
*** silor has joined #openstack-swift11:29
*** silor1 has joined #openstack-swift11:34
*** silor has quit IRC11:35
*** silor1 is now known as silor11:35
*** ppai has quit IRC11:48
*** hseipp has quit IRC12:01
*** jerrygb_ has quit IRC12:07
*** jerrygb has joined #openstack-swift12:14
*** jerrygb has quit IRC12:20
*** silor has quit IRC12:23
*** vint_bra has joined #openstack-swift12:30
*** zul has joined #openstack-swift12:31
*** zul has quit IRC12:39
*** zul has joined #openstack-swift12:40
*** donagh has left #openstack-swift12:40
*** zul has quit IRC12:41
*** catintheroof has joined #openstack-swift12:42
*** zul has joined #openstack-swift12:42
*** vinsh has joined #openstack-swift12:48
*** jerrygb has joined #openstack-swift12:49
*** vinsh has quit IRC12:53
*** _JZ_ has quit IRC13:05
*** jerrygb_ has joined #openstack-swift13:07
*** jerrygb has quit IRC13:08
*** hseipp has joined #openstack-swift13:09
*** jerrygb has joined #openstack-swift13:10
*** jerrygb_ has quit IRC13:12
*** jerrygb_ has joined #openstack-swift13:12
*** AnxiousGarlic_ has joined #openstack-swift13:13
*** AnxiousGarlic has quit IRC13:15
*** jerrygb has quit IRC13:15
*** jerrygb_ has quit IRC13:24
*** asettle has quit IRC13:35
*** asettle has joined #openstack-swift13:36
*** jamielennox is now known as jamielennox|away13:39
*** rledisez1 has joined #openstack-swift13:57
*** rledisez has quit IRC13:58
*** jerrygb has joined #openstack-swift14:01
*** klamath has joined #openstack-swift14:02
*** klamath has quit IRC14:02
*** klamath has joined #openstack-swift14:02
*** aleph1 is now known as agarner14:18
*** jerrygb has quit IRC14:22
*** jerrygb has joined #openstack-swift14:22
*** jerrygb has quit IRC14:23
*** jordanP has quit IRC14:32
*** sams-gleb has quit IRC14:34
*** sams-gleb has joined #openstack-swift14:34
*** jerrygb has joined #openstack-swift14:35
hugokuoit is but not regular one. There's an extra key cors14:35
*** jordanP has joined #openstack-swift14:35
*** sams-gleb has quit IRC14:39
*** psachin has quit IRC14:41
*** jerrygb_ has joined #openstack-swift14:41
*** jerrygb has quit IRC14:43
*** remix_auei is now known as remix_tj14:46
*** links has quit IRC14:50
*** sams-gleb has joined #openstack-swift14:50
*** mvk has quit IRC15:03
*** jerrygb has joined #openstack-swift15:11
*** chosafine has joined #openstack-swift15:11
*** jerrygb_ has quit IRC15:13
*** jerrygb has quit IRC15:15
*** mvk has joined #openstack-swift15:18
*** _JZ_ has joined #openstack-swift15:21
*** jerrygb has joined #openstack-swift15:23
*** asettle has quit IRC15:29
*** asettle has joined #openstack-swift15:30
*** Jeffrey4l_ has joined #openstack-swift15:35
*** jistr is now known as jistr|biab15:40
*** jistr|biab is now known as jistr15:43
*** mvk has quit IRC15:44
*** d0ugal has quit IRC15:44
*** d0ugal has joined #openstack-swift15:44
*** d0ugal has quit IRC15:44
*** d0ugal has joined #openstack-swift15:44
*** d0ugal has quit IRC15:44
*** d0ugal has joined #openstack-swift15:45
*** d0ugal has joined #openstack-swift15:45
*** jerrygb has quit IRC15:46
*** jerrygb has joined #openstack-swift15:48
*** chosafine has quit IRC15:51
*** jerrygb has quit IRC15:51
*** sgundur_ has joined #openstack-swift15:52
*** vinsh has joined #openstack-swift15:54
*** jordanP has quit IRC16:05
*** jordanP has joined #openstack-swift16:05
*** klrmn has quit IRC16:09
*** sgundur_ has quit IRC16:14
*** sgundur_ has joined #openstack-swift16:14
*** sgundur_ has quit IRC16:19
*** Jeffrey4l_ has quit IRC16:22
*** Jeffrey4l_ has joined #openstack-swift16:24
*** klrmn has joined #openstack-swift16:33
*** Jeffrey4l__ has joined #openstack-swift16:35
*** pcaruana has quit IRC16:38
*** rcernin has quit IRC16:38
*** Jeffrey4l_ has quit IRC16:39
notmynamegood morning16:41
notmynameFYI two patches to call your attention to, in the wider openstack realm...16:49
notmynamehttps://review.openstack.org/#/c/386555/ is an API standards proposal for discovering capabilities16:49
patchbotpatch 386555 - api-wg - Add API capabilities discovery guideline16:49
notmynamehttps://review.openstack.org/#/c/398875/ is about adding new languages (ie golang), and it looks like it's starting to get the final approval votes. now's the time to say something if there's anything in it that concerns you16:50
patchbotpatch 398875 - governance - Reference doc for new language additions16:50
*** Jeffrey4l__ has quit IRC16:51
*** sgundur_ has joined #openstack-swift16:56
*** tesseract has quit IRC16:59
*** bkopilov has quit IRC17:03
notmynamehmm... https://twitter.com/joshkodroff/status/818500031732973570 -> "what's Swift like...I've heard not-nice things about OpenStack in general..."17:06
*** klrmn has quit IRC17:09
notmynameif (1) you use twitter and (2) you've got something to say from an ops/user perspective, please respond! :-)17:09
*** sgundur_ has quit IRC17:09
*** oshritf has quit IRC17:12
*** JimCheung has joined #openstack-swift17:13
*** asettle has quit IRC17:14
*** sgundur_ has joined #openstack-swift17:16
pdardeauon the golang patch, "have been proven to not meet the technical requirements" seems odd17:21
pdardeau'proven' sounds like a gotcha word17:22
pdardeaui.e., a hard-core critic might say "you've shown, but you haven't proven"17:22
pdardeaumaybe that's just being a bit cynical, idk17:23
*** hseipp has quit IRC17:32
notmynameoh, and the election has started for the board members https://www.openstack.org/election/2017-individual-director-election/CandidateList17:32
*** sgundur_ has quit IRC17:36
*** mvk has joined #openstack-swift17:42
*** rledisez1 has quit IRC17:43
*** hoonetorg has joined #openstack-swift17:44
*** cbartz has quit IRC17:45
timburkegood morning17:49
*** Guest96631 is now known as mgagne17:51
*** mgagne has quit IRC17:51
*** mgagne has joined #openstack-swift17:51
*** sgundur_ has joined #openstack-swift17:55
timburkeacoles: so clayg made a reasonable case to me for preserving the building and balancing that's going on in those tests, and my conclusions are (1) those assertions are all *terrible* and (2) the seed parameter is a crutch to preserve a bad test18:00
*** klrmn has joined #openstack-swift18:01
acolestimburke: bu the building and balancing is deterministic due to seed right?18:02
acolesbut*18:02
timburkedropping the seed parameter and repeating the test a few thousand times indicates that the spread is actually much worse than 5-13, and can get to 30+. still loads better than the 3000+ that you saw prior to torgomatic's patch, though18:02
acolesuhuh. I was trying to figure out the reason for 13.18:03
timburkeyes... but the determinism is a bug, not a feature. if we want that test to be meaningful, it should use a sampling rather than a single ring that someone claims is representative18:03
timburkeand we should make tests about the general characteristics that we expect, rather than the specific outcomes that we see in this one case18:04
timburke...which means that this is going to take forever to come up with tight-but-not-too-tight constraints :-(18:04
acolesokay, you guys have probably thought more about in than I have. But it does seem that a test that we always get the same answer from the same ring has some value? reminds of the suffix hashing wrinkles that crept in between recent release that might have been caught if we had a test that asserted that given state should result in a specific hash.18:08
acoleslike my comment, the test will now verify that you get same nodes from same ring with py2 and py3. If we build a random ring and then verify it in the test, do we catch weird variations between py2 and py3 of the kind that caused the test to fail in first place?18:10
*** bkopilov has joined #openstack-swift18:11
timburkeif we nix the seed, look at *all* of the parts instead of a (not-so-) random one, and repeat the test several times, i don't think it should be a problem. random() is still random, just has slightly different outcomes, which should be accounted for in our expectations anyway18:15
*** Jeffrey4l_ has joined #openstack-swift18:23
*** lespaul has joined #openstack-swift18:30
*** ChubYann has joined #openstack-swift19:04
*** sgundur_ has quit IRC19:07
*** rcernin has joined #openstack-swift19:15
*** jordanP has quit IRC19:20
*** acoles is now known as acoles_19:24
*** sgundur_ has joined #openstack-swift19:34
jrichlitimburke: You mentioned for symlinks that "We don't have a good way to determine whether this is a POST to a symlink or a HEAD as a result of a POST to an SLO"19:44
timburkejrichli: i did19:44
timburkeoh! maybe we could look at swift_source...19:45
jrichliand then later you and tdasilva talked about having symlinks not support post-as-copy19:45
jrichli... well, yes.  i admit i had thought that at first, but then figured i must not understand something ;-)19:45
jrichlibut i was going to ask about this non-support for post-as-copy19:46
timburkeit wouldn't be great but might be tolerable, provided that we land https://review.openstack.org/#/c/417209/ or something like it19:46
patchbotpatch 417209 - swift - Default object_post_as_copy to False19:46
jrichlithe problem is that one can turn post-as-copy on and off again on a cluster19:46
jrichliand you could take symlinks in and out of pipeline19:48
jrichliso i dont really know  how you can simply not support post-as-copy cases19:48
jrichliwhat am I missing?19:49
tdasilvajrichli: basically i've added a check that will not let the proxy start if post-as-copy is enabled19:49
tdasilvasys.exit like clayg suggested...19:49
*** pcaruana has joined #openstack-swift19:50
jrichliok, right.  i forgot and was focused on the wording that only the default had chagned19:50
timburkeit's the same sort of problem as comes up if you take slo out of the pipeline -- it's not like we stop serving requests to manifest objects; you turned off the feature, so any requests to data that relied on the feature are off in undefined-behavior-land19:50
jrichliok.  i think i am seeing it again now.  i just had to load it in my head again.19:51
*** openstackgerrit has joined #openstack-swift19:57
openstackgerritTim Burke proposed openstack/swift: Make test_get_more_nodes_with_zero_weight_region more robust  https://review.openstack.org/41808219:57
timburkeacoles_: clayg: ^^^ seems reasonable for the one test; doing something similar for test_get_more_nodes will be more interesting...19:58
timburkefigured i'd get feedback on the zero-weight test before going down that rabbit hole19:59
timburkefwiw, it should add something like 30 seconds to the time-in-gate because of the repeats20:00
*** chsc has joined #openstack-swift20:09
*** chsc has joined #openstack-swift20:09
openstackgerritThiago da Silva proposed openstack/swift: Symlink implementation.  https://review.openstack.org/23216220:13
claygtimburke: only in the gate?20:20
timburkeclayg: unless you export SWIFT_REPEAT_TESTS in your shell20:20
claygoic20:20
claygor run tests with tox?20:21
timburkefigured i'd be a little conservative wrt developers' time20:21
timburkeyeah20:21
timburkei guess i could look for some gate-only env var, but this is part of why we have the tox -e py27 vs. ./.unittests distinction, no?20:22
claygtimburke: i have no diea20:24
clayg*idea20:24
claygthe assertions looks good - big improvement really - I think the "run it 1000x times" is not needed and maybe bad/cruft/confusing or something; it's at best meh for me20:26
timburkeclayg: if somebody comes in and makes some changes to the ring code, i want to know that for a reasonably large sample size, those assertions all pass20:28
timburkei *think* this also helps prevent it being a test that mostly passes, but flakes out (say) 1% of the time20:29
claygyeah having it fail sporatically is pretty annoying - I think you said we jsut weren't mocking random like ... enough?20:35
claygi left a comment on the review - thanks for working on those flakey ring tests!20:35
*** sgundur_ has quit IRC20:41
claygtimburke: ok, i'm so close to finsihing reviewing the hashes pkl optimization fixes - you have to stop distracting me with ring stuff ;)20:44
*** catinthe_ has joined #openstack-swift20:59
*** catintheroof has quit IRC21:00
*** jamielennox|away is now known as jamielennox21:00
*** JimCheung has quit IRC21:06
*** vint_bra has quit IRC21:13
*** manous has joined #openstack-swift21:14
openstackgerritJanie Richling proposed openstack/swift: Symlink API documentation  https://review.openstack.org/41594321:15
*** sgundur_ has joined #openstack-swift21:16
*** dfflanders has joined #openstack-swift21:21
*** JimCheung has joined #openstack-swift21:22
mattoliverauMorning21:24
*** JimCheung has quit IRC21:26
*** sgundur_ has quit IRC21:31
*** sgundur_ has joined #openstack-swift21:32
*** Jeffrey4l__ has joined #openstack-swift21:34
*** Jeffrey4l_ has quit IRC21:35
*** sgundur_ has quit IRC21:36
*** sgundur_ has joined #openstack-swift21:44
*** __JZ__ has joined #openstack-swift21:45
*** JimCheung has joined #openstack-swift21:51
*** sams-gleb has quit IRC21:52
*** chosafine has joined #openstack-swift21:53
*** sams-gleb has joined #openstack-swift22:00
*** sams-gleb has quit IRC22:00
*** sams-gleb has joined #openstack-swift22:01
*** sams-gleb has quit IRC22:01
*** sgundur_ has quit IRC22:11
*** rcernin has quit IRC22:12
*** sgundur_ has joined #openstack-swift22:18
*** mattoliverau has quit IRC22:21
*** matt6434 has joined #openstack-swift22:21
*** matt6434 is now known as mattoliverau22:22
*** ChanServ sets mode: +v mattoliverau22:22
*** darrenc is now known as darrenc_afk22:23
*** catinthe_ has quit IRC22:24
*** lespaul has quit IRC22:31
jrichlitdasilva, timburke, notmyname: any thoughts on acoles comment for symlinks here https://review.openstack.org/#/c/415943/3/api-ref/source/parameters.yaml@84422:32
patchbotpatch 415943 - swift - Symlink API documentation22:32
notmynamejrichli: we've got 5 people in a room talking about it right now ;-)22:33
jrichlior here https://review.openstack.org/#/c/415943/3/api-ref/source/storage-object-services.inc@45622:33
patchbotpatch 415943 - swift - Symlink API documentation22:33
jrichlinotmyname: great22:33
*** geaaru has quit IRC22:54
*** JimCheung has quit IRC22:57
*** darrenc_afk is now known as darrenc23:00
*** sams-gleb has joined #openstack-swift23:02
*** sams-gleb has quit IRC23:06
*** sgundur_ has quit IRC23:07
*** asettle has joined #openstack-swift23:12
*** JimCheung has joined #openstack-swift23:13
*** JimCheung has quit IRC23:17
openstackgerritPete Zaitcev proposed openstack/swift: Improve the "Server started on port 6200" message  https://review.openstack.org/41815023:18
*** JimCheung has joined #openstack-swift23:21
*** asettle has quit IRC23:22
*** asettle has joined #openstack-swift23:23
*** asettle has quit IRC23:27
*** chsc has quit IRC23:34
*** kei_yama has joined #openstack-swift23:37
*** mminesh has quit IRC23:45
*** mminesh has joined #openstack-swift23:45
tdasilvanotmyname, jrichli: i remember discussing this earlier with ppai and torgomatic; can't remember the reasons atm why it was better to follow this 2 header approach.23:50
tdasilvai think one of the reason is that it it's easier to parse, since x-symlink-target could be 'container/dir1/dir2/obj', so you can't just split path on / and expect the first one to be account23:51
notmynameit mirrors the existing api for copy, so it's fine to me23:51
tdasilvaand versioning right?23:51
tdasilvaIIRC23:52
tdasilvano, version is only one header, only copy23:53

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