Tuesday, 2016-09-13

*** klrmn has quit IRC00:01
claygnotmyname: why'd you make me look at patch 348495 - container sync + SLO is horrible :\00:05
patchbothttps://review.openstack.org/#/c/348495/ - swift - Make container sync copy SLO manifests00:05
*** Suyi_ has quit IRC00:07
*** diogogmt has quit IRC00:09
claygsometimes I feel like any metadata that's not in the sysmeta namespace is a bug waiting to happen :\00:09
timburkealmost certainly. there's a reason i'm not returning all of allowed_headers in https://review.openstack.org/#/c/357559/00:11
patchbotpatch 357559 - swift - Include metadata in PUT/POST responses00:11
pdardeaui spent a little time trying to fully understand the makeup of swift data dir file path00:16
pdardeaudecided to document my findings for future reference00:16
pdardeauplease let me know if i got any of it wrong00:16
pdardeauhttps://gist.github.com/pauldardeau/559048ef35ce57cb4fb940bfce139f0d00:16
pdardeau*swift data file path00:16
claygtimburke: what about x-delete-at x-object-manifest and x-static-large-object tho?00:17
*** Guess456787654 has joined #openstack-swift00:17
timburkeclayg: not immediately relevant to what i wanted to do with it, nor to tdasilva's use-case. if you need it, you can add it :-P00:18
kota_good morning00:19
timburkegood morning kota_00:19
claygpdardeau: looks good to me!  i'm sure you've noticed we call "last 3 chars of name hash" the "suffix"00:19
kota_hi timburke!00:19
claygit's just a good way to break up all the hashdirs in the partition since there's a lot of them and they all have the same prefix ;)00:19
kota_timburke: we are talking in dual channel :P00:19
timburkeit happens :)00:20
pdardeauclayg: thx. i'll update00:20
claygpdardeau: oh, you don't need to - everything you stated there was correct AFAIK!00:21
* kota_ is thinking we have had a lot of talks in the last night (in my time), so need to update the backlog00:21
claygtimburke: what is the use-case you and tdasilva have for metadata in responses?00:21
timburkefor me, returning encryption status; for tdasilva, symlinks00:22
timburke(and in particular, noticing that you just POSTed to one)00:22
claygi vaugly remember needing some metadata in a response (x-timestamp?) - but that have been only in the backend - maybe related to a conversation with jrichli?  or someone wrt to symlinks?00:22
claygtimburke: maybe that was it!00:22
*** diogogmt has joined #openstack-swift00:24
jrichliclayg timburke: there was an x-timestamp thing added for "help with ACL-caching in swift3"00:25
*** dmorita_ has quit IRC00:25
jrichliyou know I got notes on everything, right? ;-)00:25
*** dmorita has joined #openstack-swift00:26
*** diogogmt has quit IRC00:27
timburkethanks jrichli :-)00:27
notmynameclayg: sorry :-(00:29
*** dmorita has quit IRC00:30
*** Jeffrey4l has quit IRC00:33
*** vint_bra has joined #openstack-swift00:43
*** chsc has quit IRC00:46
*** Jeffrey4l has joined #openstack-swift00:47
claygit irks me so much that multipart-manifest=get&format=raw doesn't return the raw format of the manifest as stored ondisk - but instead the weird translated format that the client sends - which is different from what gets stored for no good reason (client could easily send raw^Wondisk format)00:47
timburkeclayg: i still kinda feel like we ought to accept either format, with an eye toward deprecating the current client-facing one00:49
claygtimburke: yes yes yes +1000000:49
claygtimburke: i think the only reason that multipart-manifest=get can return the client format is for copy middleware yes?  could have easily gone the other way - let the slo middlware accept the current ondisk format - the "raw" format if you will ;)00:50
timburkeand allowing clients to re-upload without needing to do the transformation themselves -- basically, the container-sync use-case00:51
*** dmorita has joined #openstack-swift00:51
*** zhengyin has quit IRC00:52
*** zhengyin has joined #openstack-swift00:52
claygnope, i seem to be wrong about that -> format=raw exists compleatly orthogonal of the copy middleware -> https://review.openstack.org/#/c/263902/00:53
patchbotpatch 263902 - swift - Re-format the SLO manifest file on new multipart-m... (MERGED)00:53
timburkei think we recognized that it would be a necessary pre-req, though00:53
*** tongli has joined #openstack-swift00:55
timburkeindeed, it looks like acoles_ recommended a second param over multipart-manifest-get-raw specifically for copy's benefit00:59
*** m_kazuhiro has joined #openstack-swift01:12
*** chlong_ has joined #openstack-swift01:14
*** tsg has joined #openstack-swift01:17
*** vint_bra has quit IRC01:22
*** Guess456787654 has quit IRC01:23
*** vint_bra has joined #openstack-swift01:25
*** cdelatte has quit IRC01:26
*** _JZ_ has joined #openstack-swift01:26
*** diogogmt has joined #openstack-swift01:28
claygg'night01:33
*** thumpba has joined #openstack-swift01:33
*** thumpba has quit IRC01:33
zhengyinclayg: thanks a lot :)01:34
redbopdardeau: Also filenames can have multiple timestamps, if that's in scope.01:40
*** vint_bra has quit IRC01:43
openstackgerritMerged openstack/swift: More Test[Account|Container]Controller test updates  https://review.openstack.org/36362801:48
*** chlong_ has quit IRC01:49
*** thumpba has joined #openstack-swift01:52
*** Guess456787654 has joined #openstack-swift01:55
*** tsg_ has joined #openstack-swift01:58
*** tsg has quit IRC01:58
*** klrmn has joined #openstack-swift02:01
zhengyingood morning02:04
*** StraubTW has joined #openstack-swift02:04
*** chlong_ has joined #openstack-swift02:05
*** bkopilov__ has quit IRC02:24
*** bkopilov_ has quit IRC02:26
*** bkopilov has quit IRC02:27
*** tqtran has quit IRC02:27
*** tsg_ has quit IRC02:29
*** dmorita has quit IRC02:45
*** dmorita has joined #openstack-swift02:46
*** dmorita has quit IRC02:50
notmynamegood evening02:52
*** tsg has joined #openstack-swift02:56
*** tongli has quit IRC03:03
mattoliveraunotmyname: o/03:04
*** tsg has quit IRC03:06
*** mingyu has quit IRC03:10
zhengyinnotmyname, mattoliverau: o/03:10
notmynamemattoliverau: so my LCA talk...you know, the one in January--four months from now03:11
notmynamemattoliverau: I realized that I told some people I'd give that talk at a meetup next week. guess I better start writing it ;-)03:11
*** StraubTW has quit IRC03:13
*** mingyu has joined #openstack-swift03:13
mattoliveraunotmyname: lol, tho 4 months means it;ll probably be retuned and a completely different talk by LCA :)03:15
notmynameit will definitely be different :-)03:15
mattoliveraunotmyname: however, you usually only right your LCA talk the week before, so it's not like you've given yourself any less time :P03:17
notmynamehey!03:17
mattoliveraus/right/write/03:17
notmynamethat's not completely true!03:17
mattoliveraulol03:18
*** mingyu has quit IRC03:18
notmynamemattoliverau: zhengyin: did you see that I updated https://wiki.openstack.org/wiki/Swift/PriorityReviews ? That can give you a starting point for reviews this week03:20
notmynamemattoliverau: and youv'e already +a'd https://review.openstack.org/#/c/162243/ yay!03:21
patchbotpatch 162243 - swift - Make object creation more atomic in Linux03:21
mattoliveraunotmyname: yup, the first one is done ;)03:21
zhengyinnotmyname: ok :(03:22
mattoliveraunotmyname: yeah, just needed to build an SAIO on a newer kernel to test it out some more, so that and meetings took up my morning. Time to move to the next review :)03:22
mattoliverauzhengyin: morning (sorry for being late). Been a busy morning03:22
zhengyinmattoliverau: :)03:23
mattoliverauLooks like if I set up my env for some container sync action I'll be set for a few more priority reviews.03:24
notmynamemattoliverau: if I could suggest one to start with, it would be https://review.openstack.org/#/c/355958/03:27
patchbotpatch 355958 - swift - EC - eliminate .durable files03:27
*** Guess456787654 has quit IRC03:29
*** tsg has joined #openstack-swift03:31
*** tsg has quit IRC03:32
mahaticgood morning03:34
notmynamehello mahatic03:38
*** thumpba has quit IRC03:38
*** thumpba has joined #openstack-swift03:38
mahaticnotmyname: o/03:41
jrichlimahatic: o/03:46
*** tsg has joined #openstack-swift03:47
mattoliveraumahatic: o/03:50
*** bkopilov has joined #openstack-swift03:51
mahaticjrichli: mattoliverau: hello!03:51
*** bkopilov_ has joined #openstack-swift03:51
*** bkopilov__ has joined #openstack-swift03:52
zhengyinjrichli,mahatic: o/03:53
jrichlizhengyin: hi!03:54
openstackgerritMerged openstack/swift: Make object creation more atomic in Linux  https://review.openstack.org/16224304:02
mahaticzhengyin: o/04:03
*** jamielennox is now known as jamielennox|away04:05
*** ppai has joined #openstack-swift04:07
*** jamielennox|away is now known as jamielennox04:08
*** m_kazuhiro has quit IRC04:08
*** aswadr_ has joined #openstack-swift04:09
*** m_kazuhiro has joined #openstack-swift04:09
*** klrmn has quit IRC04:10
*** m_kazuhiro has quit IRC04:11
*** m_kazuhiro has joined #openstack-swift04:11
*** mingyu has joined #openstack-swift04:15
charzo/04:17
*** mingyu has quit IRC04:19
*** ppai has quit IRC04:41
*** jamielennox is now known as jamielennox|away04:41
*** tsg has quit IRC04:41
*** tsg_ has joined #openstack-swift04:42
*** psachin has joined #openstack-swift04:43
*** thumpba has quit IRC04:43
*** jamielennox|away is now known as jamielennox04:46
mattoliveraucharz: o/04:50
*** Jeffrey4l_ has joined #openstack-swift05:00
*** Jeffrey4l has quit IRC05:04
*** amitkqed has quit IRC05:16
*** amitkqed has joined #openstack-swift05:16
*** ChubYann has quit IRC05:16
*** takashi has joined #openstack-swift05:23
*** tsg has joined #openstack-swift05:31
*** ppai has joined #openstack-swift05:32
*** tsg_ has quit IRC05:34
*** jamielennox is now known as jamielennox|away05:53
*** jamielennox|away is now known as jamielennox06:00
*** jamielennox is now known as jamielennox|away06:13
*** rcernin has joined #openstack-swift06:19
*** jamielennox|away is now known as jamielennox06:30
*** hseipp has joined #openstack-swift06:43
*** pcaruana has joined #openstack-swift07:02
*** tesseract- has joined #openstack-swift07:08
*** pdardeau has quit IRC07:11
*** sgundur1 has quit IRC07:11
*** mmotiani has quit IRC07:11
*** yarkot has quit IRC07:11
*** ntata has quit IRC07:11
*** smurke has quit IRC07:11
*** rledisez has joined #openstack-swift07:15
*** mmotiani has joined #openstack-swift07:16
*** joeljwright has joined #openstack-swift07:21
*** ChanServ sets mode: +v joeljwright07:21
*** ntata has joined #openstack-swift07:21
oxinaboxSemantically, is the pseudofolder part of the container name, or the object name? I'm not sure if it is actually different between upload and download, but it seems like it is, possibly because one of the two corrects it if you have it wrong, and the other doesn't07:44
*** sams-gleb has joined #openstack-swift07:44
oxinaboxIs it a subcontainer, or is it a prefix to the key?07:44
*** openstackgerrit has quit IRC07:48
*** openstackgerrit has joined #openstack-swift07:49
*** mingyu has joined #openstack-swift07:52
*** jaakkos has quit IRC07:55
*** jaakkos has joined #openstack-swift07:56
*** mingyu has quit IRC07:57
*** kodoku has quit IRC08:03
ahale_its part of the object name08:08
openstackgerritMerged openstack/swift: Add test_pop in the test_header_key_dict  https://review.openstack.org/36773308:10
*** bkopilov__ has quit IRC08:24
*** bkopilov_ has quit IRC08:24
*** mvk has quit IRC08:27
oxinaboxthanks08:27
*** tqtran has joined #openstack-swift08:28
*** tqtran has quit IRC08:32
*** zigo_ is now known as zigo08:34
*** asettle has joined #openstack-swift08:36
*** jistr is now known as jistr|mtgs08:43
*** mvk has joined #openstack-swift08:57
*** takashi has quit IRC08:59
*** tsg has quit IRC09:02
*** tsg has joined #openstack-swift09:03
*** tsg has quit IRC09:06
*** tsg has joined #openstack-swift09:06
*** trananhkma has joined #openstack-swift09:10
*** oshritf has joined #openstack-swift09:14
*** asettle has quit IRC09:21
*** asettle has joined #openstack-swift09:21
*** asettle has quit IRC09:25
*** asettle has joined #openstack-swift09:27
*** tsg_ has joined #openstack-swift09:39
*** tsg has quit IRC09:39
*** acoles_ is now known as acoles09:41
openstackgerritOpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements  https://review.openstack.org/8925009:42
openstackgerritOpenStack Proposal Bot proposed openstack/swift: Updated from global requirements  https://review.openstack.org/8873609:43
*** mingyu has joined #openstack-swift09:54
*** mingyu has quit IRC09:58
oxinaboxWhat can I do to imporve my download speed on big files? I get 65Megabytes/s uploading but on 2Megabytes/s downloading.10:11
oxinaboxIs there a guide for this somewhere?10:11
*** permalac has joined #openstack-swift10:17
*** tsg_ has quit IRC10:24
*** tsg_ has joined #openstack-swift10:24
*** fbo_ is now known as fbo10:28
*** m_kazuhiro has quit IRC10:30
*** ppai has quit IRC10:34
*** permalac has quit IRC10:51
*** ppai has joined #openstack-swift10:51
*** cbartz has joined #openstack-swift11:04
*** kei_yama has quit IRC11:06
*** bkopilov has quit IRC11:27
*** tqtran has joined #openstack-swift11:31
*** tqtran has quit IRC11:35
*** tsg_ has quit IRC11:35
*** tsg_ has joined #openstack-swift11:35
*** dmorita has joined #openstack-swift11:46
*** dmorita has quit IRC11:51
*** mvk has quit IRC11:52
*** asettle has quit IRC11:53
*** asettle has joined #openstack-swift11:54
*** MVenesio has joined #openstack-swift11:55
*** asettle has quit IRC11:58
*** asettle has joined #openstack-swift11:58
*** MVenesio has quit IRC11:59
*** mvk has joined #openstack-swift12:04
*** tongli has joined #openstack-swift12:06
*** tsg_ has quit IRC12:07
*** trananhkma has quit IRC12:10
*** _JZ_ has quit IRC12:12
*** Guess456787654 has joined #openstack-swift12:18
*** dmellado is now known as dmellado|lunch12:27
*** dmellado|lunch is now known as dmellado12:27
*** pbandark has joined #openstack-swift12:29
pbandarkcan we use single swift service on multiple openstack setup? for ex. i have configured swift on one setup which is backend for cinder backup service. I want to use same swift backend on other openstack setup as a backend for cinder backup. I tried with updating "backup_swift_url"(cinder.conf). but i am unable to take backup of volume it fails with  Unauthorized error12:30
*** ppai has quit IRC12:30
pbandarkI want to create shared backend for cinder backup service on 2 openstack setup12:30
*** permalac has joined #openstack-swift12:33
*** Guess456787654 has left #openstack-swift12:39
*** silor has joined #openstack-swift12:45
cbartzIs there a core reviewer available for reviewing: https://review.openstack.org/#/c/333331/ ?12:46
patchbotpatch 333331 - swift - Preserve query params in tempurl12:46
*** StraubTW has joined #openstack-swift12:52
*** klamath has joined #openstack-swift12:53
*** klamath has quit IRC12:53
*** klamath has joined #openstack-swift12:53
*** silor1 has joined #openstack-swift12:54
*** Jeffrey4l_ has quit IRC12:55
*** silor has quit IRC12:55
*** silor1 is now known as silor12:55
*** mingyu has joined #openstack-swift12:56
*** bkopilov has joined #openstack-swift12:56
*** Jeffrey4l_ has joined #openstack-swift12:57
*** mingyu has quit IRC13:00
*** caiobrentano_ has joined #openstack-swift13:01
*** klamath has quit IRC13:11
*** thumpba has joined #openstack-swift13:24
*** pdardeau has joined #openstack-swift13:28
*** _JZ_ has joined #openstack-swift13:30
openstackgerritMathias Bjoerkqvist proposed openstack/swift: WIP: Storing encryption root secret in Barbican  https://review.openstack.org/36487813:44
*** vint_bra has joined #openstack-swift13:54
*** diogogmt has quit IRC14:00
*** baojg has joined #openstack-swift14:01
*** baojg has quit IRC14:01
*** baojg has joined #openstack-swift14:02
*** baojg has quit IRC14:02
*** baojg has joined #openstack-swift14:02
*** baojg has quit IRC14:03
*** baojg has joined #openstack-swift14:03
*** baojg has quit IRC14:03
*** thumpba has quit IRC14:04
*** baojg has joined #openstack-swift14:04
*** baojg has quit IRC14:14
*** baojg has joined #openstack-swift14:16
*** thumpba has joined #openstack-swift14:24
*** baojg has quit IRC14:30
*** smurke has joined #openstack-swift14:52
*** cdelatte has joined #openstack-swift14:53
*** joeljwright has quit IRC15:07
*** joeljwright has joined #openstack-swift15:08
*** ChanServ sets mode: +v joeljwright15:08
*** cnf has quit IRC15:09
*** schizo has quit IRC15:10
*** mingyu has joined #openstack-swift15:10
*** mingyu has quit IRC15:11
*** mingyu has joined #openstack-swift15:11
*** dosaboy has quit IRC15:12
*** mtreinish has quit IRC15:12
*** sams-gleb has quit IRC15:12
*** dosaboy has joined #openstack-swift15:12
*** schizo has joined #openstack-swift15:13
*** schizo has joined #openstack-swift15:13
*** oshritf has quit IRC15:13
ntataGood Morning!15:13
*** psachin has quit IRC15:14
*** mtreinish has joined #openstack-swift15:14
*** cnf has joined #openstack-swift15:14
claygoxinabox: that's opposite of most peoples experience with large files - upload is normally slower than download by some good measure15:16
claygpbandark: i'm not sure that multiple cinder installations sharing a single swift cluster is usecase that's specifically covered15:17
claygpbandark: the authorization issue is something with the keystone service user configuration - you should get a single cinder instance working first15:17
*** caiobrentano_ has quit IRC15:17
claygthen you can try to add the second one as a different service user (that's most likely to isolate things from swift's perspective of being a different account)15:18
claygi'm not sure of the implication of configuring the same cinder service user for both cinder clusters15:19
*** CaioBrentano has joined #openstack-swift15:24
*** diogogmt has joined #openstack-swift15:25
*** zaitcev has joined #openstack-swift15:28
*** ChanServ sets mode: +v zaitcev15:28
*** chsc has joined #openstack-swift15:30
*** chsc has quit IRC15:30
*** chsc has joined #openstack-swift15:30
zaitcevhttps://www.linkedin.com/pulse/hpe-dumps-its-legacy-bets-farm-hybrid-simon-phipps?trk=hp-feed-article-title-comment15:33
notmynamegood morning15:33
zaitcev"[HPE] Combines the strong OpenStack bench SUSE has been building with HPE's OpenStack team in an unspecified way"15:33
zaitcevWell, as long as Alasdair is safe15:33
*** tqtran has joined #openstack-swift15:34
notmynamezaitcev: welcome back. hope your vacation was good15:34
zaitcevnotmyname: fabulous. No Internet for 3 days.15:35
*** sams-gleb has joined #openstack-swift15:36
zaitcevIt's great unless a life-threatening injury occurs, then it could suck. Been buzzed by Navy helicopters flying through a canyon in which we camped.15:36
*** chsc has quit IRC15:38
*** tqtran has quit IRC15:39
ahale_so - hows hiring in the openstack swift world ?15:41
zaitcevHiring is great if you're not a complete idiot like myself... Then they ask things like, what have you done in Swift? And I'm like: "Umm...."15:42
*** sams-gleb has quit IRC15:42
*** mingyu has quit IRC15:45
*** silor has quit IRC15:50
*** zul has joined #openstack-swift15:51
*** takashi has joined #openstack-swift15:53
*** jistr|mtgs is now known as jistr15:57
*** sheel has joined #openstack-swift15:57
*** cbartz has left #openstack-swift16:00
*** zhengyin has quit IRC16:04
*** mvk has quit IRC16:14
*** permalac has quit IRC16:20
*** ChubYann has joined #openstack-swift16:28
*** pbandark has quit IRC16:29
*** hseipp has quit IRC16:29
*** delattec has joined #openstack-swift16:35
*** asettle has quit IRC16:35
*** dmorita has joined #openstack-swift16:35
*** asettle has joined #openstack-swift16:35
*** cdelatte has quit IRC16:37
*** dmorita has quit IRC16:38
*** dmorita has joined #openstack-swift16:38
*** calebb has quit IRC16:40
*** asettle has quit IRC16:41
*** squid has joined #openstack-swift16:42
*** squid is now known as calebb16:42
*** tesseract- has quit IRC16:46
*** ChubYann has quit IRC16:52
*** rledisez has quit IRC16:53
timburkeclayg: if you've got a symmetric connection, i totally expect upload to be better than download. it's all about the connections -- i can upload in parallel, unless i'm willing to grab the manifest and reconstruct client-side, i'm downloading sequentially16:53
claygtimburke: I hadn't considered that oxinabox might be using parallel large object upload16:55
mahaticredbo: hi, could you take a look at this patch 346865 when you have a moment? clayg has provided great inputs, would be nice to know what you think of it.16:57
patchbothttps://review.openstack.org/#/c/346865/ - swift - Delete old tombstones16:57
timburkeiirc, it's swiftclient (and SwiftService in particular) wrapped up in julia, so the parallel upload happens for "free"16:57
timburkei've got some ideas for https://bugs.launchpad.net/python-swiftclient/+bug/1621562 but sadly not enough time :-(16:58
openstackLaunchpad bug 1621562 in python-swiftclient "Download large objects concurrently" [Wishlist,Confirmed]16:58
*** pcaruana has quit IRC16:59
*** klrmn has joined #openstack-swift17:03
*** rcernin has quit IRC17:03
*** chsc has joined #openstack-swift17:04
*** chsc has joined #openstack-swift17:04
*** thumpba has quit IRC17:05
-openstackstatus- NOTICE: The Gerrit service on review.openstack.org is being restarted now to address current performance problems, but should return to a working state within a few minutes17:11
*** mvk has joined #openstack-swift17:13
*** yarkot has joined #openstack-swift17:14
*** yarkot has quit IRC17:14
*** daemontool has joined #openstack-swift17:18
*** tqtran has joined #openstack-swift17:32
claygacoles: donagh is right - if you pull out the ETag thing we can go forward with patch 348495 as-is - i have some broader concerens obviously - but i mostly think teaching container-sync to boss the slo middleware around is fine for the short term?17:33
patchbothttps://review.openstack.org/#/c/348495/ - swift - Make container sync copy SLO manifests17:33
claygIMHO17:33
*** daemontool has quit IRC17:36
*** Suyi_ has joined #openstack-swift17:37
*** aswadr_ has quit IRC17:42
*** takashi has quit IRC17:44
*** klrmn has quit IRC17:49
*** klrmn has joined #openstack-swift17:50
*** joeljwright has quit IRC17:55
*** yarkot has joined #openstack-swift17:56
tdasilvatorgomatic, timburke: i remember a couple of patches you guys worked on get_info a while back, should we mark this bug fixed? https://bugs.launchpad.net/swift/+bug/150193918:00
openstackLaunchpad bug 1501939 in OpenStack Object Storage (swift) "get_info environ caching is leaky and brittle" [Medium,Confirmed]18:00
*** sheel has quit IRC18:06
timburketdasilva: probably? i think between https://review.openstack.org/#/c/280977/ and https://review.openstack.org/#/c/280978/ we took care of most of it18:06
patchbotpatch 280977 - swift - Make info caching work across subrequests (MERGED)18:06
patchbotpatch 280978 - swift - Fix up get_account_info and get_container_info (MERGED)18:06
timburkeand by "we" i mean "torgomatic"18:07
*** rcernin has joined #openstack-swift18:13
jrichlinotmyname: looks like the api-ref docs do get updated with the builds.  I had remembered us talking about docs only getting rebuilt with releases or something - which is why I had asked.18:16
notmynamejrichli: ah ok. I know swiftclient used to be that way, but IMO it should all be built at every commit (with links in the sidebar to older versions)18:17
jrichliok18:17
clarkbnotmyname: clayg: trying to debug why glance + swift + keystone v3 only is unhappy here http://logs.openstack.org/50/369550/1/check/gate-tempest-dsvm-neutron-identity-v3-only-full-ubuntu-xenial-nv/7626439/logs/screen-g-api.txt.gz#_2016-09-13_17_36_57_378 is that a known issue? though rereading that traceback glance might have a better idea18:18
notmynameclarkb: not off of the top of my head18:20
*** delattec has quit IRC18:20
notmynametimburke: no known keystone v3 issues that you know of?18:20
notmynameclarkb: have you been able to talk directly to swift with a v3 keystone endpoint (ie without going through glance)?18:21
clarkbrereading that I think glance is the thing failing to keystone there not swift. I thought swift may have returned the 401 but _get_storage_url is what raises and that should be talking to keystone catalog things18:21
clarkbnotmyname: not yet I am spinning up an env to reproduce t his now18:21
notmynamekk18:21
zaitcevI think API path changed in v3. It used to have /path/v2.0 in there, but now it's just /path to which you do POST18:22
zaitcevI end testing with curl -XPOST every time, because I can never remember where the endpoint is.18:23
*** david-lyle has quit IRC18:23
timburkenotmyname: not that i'm aware of, no18:25
timburkeclarkb: i agree; looks like the 401 happened while trying to get the auth token18:25
timburkeurl looks more or less reasonable, but i thought the port was usually 5000?18:26
*** david-lyle has joined #openstack-swift18:26
timburkelooks like this is the relevant bit from the keystone logs, though there still isn't a clear reason that auth failed: http://logs.openstack.org/50/369550/1/check/gate-tempest-dsvm-neutron-identity-v3-only-full-ubuntu-xenial-nv/7626439/logs/apache/keystone.txt.gz?#_2016-09-13_17_36_57_37418:30
*** pcaruana has joined #openstack-swift18:36
zaitcevOh dear god, while I was frolicking in desert, Clay added whitespace to my ugly code.18:38
*** cdelatte has joined #openstack-swift18:39
*** MVenesio has joined #openstack-swift18:40
*** yarkot has quit IRC18:44
tqtranacoles: ping, are you still around. have another question. does the concept of a user_group make sense in keystoneauth context (where we are just checking for user id)?18:48
acolestqtran: no, not to me. user have roles on projects, I'm not sure what a user_group would be or represent.18:50
acolesclayg: ack, thanks for review! I guess I got confused over slo etags (easily done!), will fix soon, been busy doing CPR on optimistic GETs18:51
openstackgerritAlistair Coles proposed openstack/swift: Enable object server to return non-durable data  https://review.openstack.org/21527618:52
acoleskota_: phew, that was a fantastic review, thank you. ^^ fixes your points and rebases for another merge conflict with master.18:53
*** acoles is now known as acoles_18:54
*** samueldmq has joined #openstack-swift19:02
openstackgerritThai Tran proposed openstack/swift: Supporting Account ACL in keystoneauth  https://review.openstack.org/35671519:09
*** itlinux has joined #openstack-swift19:11
*** zul has quit IRC19:13
*** david-lyle has quit IRC19:16
*** CaioBrentano has quit IRC19:16
*** jlvillal has quit IRC19:16
*** cschwede has quit IRC19:16
*** glange has quit IRC19:16
*** yarkot1 has quit IRC19:16
*** htruta has quit IRC19:16
*** ahale_ has quit IRC19:16
*** notmyname has quit IRC19:16
*** bhakta has quit IRC19:16
*** kmARC has quit IRC19:16
*** klrmn has quit IRC19:16
*** dosaboy has quit IRC19:16
*** amitkqed has quit IRC19:16
*** hoonetorg has quit IRC19:16
*** csmart has quit IRC19:16
*** patchbot has quit IRC19:16
*** cebreidian has quit IRC19:16
*** mattoliverau has quit IRC19:16
*** 17SAA00LA has quit IRC19:16
*** clarkb has quit IRC19:16
*** ctennis has quit IRC19:16
*** itlinux has quit IRC19:17
*** nottrobin has quit IRC19:17
*** neonpastor has quit IRC19:17
*** ediardo has quit IRC19:17
*** cargonza has quit IRC19:17
*** wasmum has quit IRC19:17
*** hogepodge has quit IRC19:17
*** peterlisak has quit IRC19:17
*** dabukalam_ has quit IRC19:17
*** saltsa has quit IRC19:17
*** timss has quit IRC19:17
*** madorn has quit IRC19:17
*** redbo has quit IRC19:17
*** jamielennox has quit IRC19:17
*** CrackerJackMack has quit IRC19:17
*** Guest66666 has quit IRC19:17
*** MooingLemur has quit IRC19:17
*** jamiec has quit IRC19:17
*** mathiasb has quit IRC19:17
*** tanee has quit IRC19:17
*** markstur has quit IRC19:17
*** charz has quit IRC19:17
*** rickflare has quit IRC19:17
*** logan- has quit IRC19:17
*** StraubTW has quit IRC19:17
*** zigo has quit IRC19:17
*** tamizh_geek has quit IRC19:17
*** serverascode has quit IRC19:17
*** ogelbukh has quit IRC19:17
*** HenryG has quit IRC19:17
*** kozhukalov has quit IRC19:17
*** zacksh has quit IRC19:17
*** dstepanenko_ has quit IRC19:17
*** rsFF has quit IRC19:17
*** cnf has quit IRC19:17
*** bkopilov has quit IRC19:17
*** pdardeau has quit IRC19:17
*** d0ugal has quit IRC19:17
*** darrenc has quit IRC19:17
*** ejat has quit IRC19:17
*** StevenK has quit IRC19:17
*** onovy has quit IRC19:17
*** philipw has quit IRC19:17
*** kong has quit IRC19:17
*** vern has quit IRC19:17
*** ChanServ has quit IRC19:17
*** donagh has quit IRC19:17
*** chsc has quit IRC19:17
*** vint_bra has quit IRC19:17
*** tongli has quit IRC19:17
*** vinsh has quit IRC19:17
*** geaaru has quit IRC19:17
*** zhugaoxiao has quit IRC19:17
*** etienneme has quit IRC19:17
*** samueldmq has quit IRC19:17
*** zhiyan has quit IRC19:17
*** mahatic has quit IRC19:17
*** oxinabox has quit IRC19:17
*** zackmdavis has quit IRC19:17
*** hurricanerix has quit IRC19:17
*** bikmak has quit IRC19:17
*** mminesh has quit IRC19:17
*** mcarden has quit IRC19:17
*** sileht has quit IRC19:17
*** alpha_ori has quit IRC19:17
*** cdelatte has quit IRC19:17
*** calebb has quit IRC19:17
*** jaakkos has quit IRC19:17
*** tdasilva has quit IRC19:17
*** briancline has quit IRC19:17
*** timur has quit IRC19:17
*** rcernin has quit IRC19:17
*** mtreinish has quit IRC19:17
*** schizo has quit IRC19:17
*** smurke has quit IRC19:17
*** jistr has quit IRC19:17
*** Anticimex has quit IRC19:17
*** amit213 has quit IRC19:17
*** AndyWojo has quit IRC19:17
*** jeblair has quit IRC19:17
*** cppforlife_ has quit IRC19:17
*** kota_ has quit IRC19:17
*** acorwin has quit IRC19:17
*** MVenesio has quit IRC19:17
*** Suyi_ has quit IRC19:17
*** tqtran has quit IRC19:17
*** dmorita has quit IRC19:17
*** _JZ_ has quit IRC19:17
*** Jeffrey4l_ has quit IRC19:17
*** ndk- has quit IRC19:18
*** balajir has quit IRC19:18
*** timburke has quit IRC19:18
*** gmmaha has quit IRC19:18
*** mvk has quit IRC19:18
*** diogogmt has quit IRC19:18
*** ntata has quit IRC19:18
*** chlong_ has quit IRC19:18
*** Raymii has quit IRC19:18
*** tristanC has quit IRC19:18
*** andymccr has quit IRC19:18
*** bapalm has quit IRC19:18
*** cebruns has quit IRC19:18
*** gabor_antal_ has quit IRC19:18
*** tmoreira has quit IRC19:18
*** stevemar has quit IRC19:18
*** jrichli has quit IRC19:18
*** hugokuo has quit IRC19:18
*** ajiang_ has quit IRC19:18
*** bobby2_ has quit IRC19:18
*** j_king has quit IRC19:18
*** tonyb has quit IRC19:18
*** aj701_ has quit IRC19:18
*** pcaruana has quit IRC19:18
*** zaitcev has quit IRC19:18
*** okdas has quit IRC19:18
*** DuncanT has quit IRC19:18
*** furlongm has quit IRC19:18
*** torgomatic has quit IRC19:18
*** ober37 has quit IRC19:18
*** ningliu has quit IRC19:18
*** bkeller` has quit IRC19:18
*** sanchitmalhotra has quit IRC19:18
*** briancurtin has quit IRC19:18
*** chmouel has quit IRC19:18
*** kragniz has quit IRC19:18
*** M4urice has quit IRC19:18
*** janonymous has quit IRC19:18
*** fbo has quit IRC19:18
*** sudorandom has quit IRC19:18
*** acoles_ has quit IRC19:18
*** blairo has quit IRC19:18
*** blair has quit IRC19:18
*** openstackgerrit has quit IRC19:18
*** mmotiani has quit IRC19:18
*** lifeless has quit IRC19:18
*** portante has quit IRC19:18
*** kencjohnston has quit IRC19:18
*** swifterdarrell has quit IRC19:18
*** a1|away has quit IRC19:18
*** clayg has quit IRC19:18
*** treyd has quit IRC19:18
*** mlanner has quit IRC19:18
*** jroll has quit IRC19:18
*** EmilienM has quit IRC19:18
*** tonyb has joined #openstack-swift19:21
*** StevenK has joined #openstack-swift19:23
*** patchbot has joined #openstack-swift19:23
*** cnf has joined #openstack-swift19:23
*** kmARC has joined #openstack-swift19:23
*** bhakta has joined #openstack-swift19:23
*** notmyname has joined #openstack-swift19:23
*** ahale_ has joined #openstack-swift19:23
*** htruta has joined #openstack-swift19:23
*** yarkot1 has joined #openstack-swift19:23
*** glange has joined #openstack-swift19:23
*** cschwede has joined #openstack-swift19:23
*** jlvillal has joined #openstack-swift19:23
*** CaioBrentano has joined #openstack-swift19:23
*** chlong_ has joined #openstack-swift19:23
*** onovy has joined #openstack-swift19:23
*** ctennis has joined #openstack-swift19:23
*** bobby2 has joined #openstack-swift19:23
*** kragniz has joined #openstack-swift19:23
*** acoles has joined #openstack-swift19:23
*** orwell.freenode.net sets mode: +vvvv notmyname glange cschwede acoles19:23
*** pcaruana has joined #openstack-swift19:23
*** zackmdavis has joined #openstack-swift19:23
*** mahatic has joined #openstack-swift19:23
*** zhiyan has joined #openstack-swift19:23
*** etienneme has joined #openstack-swift19:23
*** oxinabox has joined #openstack-swift19:23
*** zhugaoxiao has joined #openstack-swift19:23
*** geaaru has joined #openstack-swift19:23
*** vinsh has joined #openstack-swift19:23
*** tongli has joined #openstack-swift19:23
*** vint_bra has joined #openstack-swift19:23
*** chsc has joined #openstack-swift19:23
*** samueldmq has joined #openstack-swift19:23
*** zaitcev has joined #openstack-swift19:23
*** yarkot has joined #openstack-swift19:23
*** fbo has joined #openstack-swift19:23
*** acorwin has joined #openstack-swift19:23
*** kota_ has joined #openstack-swift19:23
*** cppforlife_ has joined #openstack-swift19:23
*** jeblair has joined #openstack-swift19:23
*** AndyWojo has joined #openstack-swift19:23
*** Anticimex has joined #openstack-swift19:23
*** jistr has joined #openstack-swift19:23
*** smurke has joined #openstack-swift19:23
*** schizo has joined #openstack-swift19:23
*** mtreinish has joined #openstack-swift19:23
*** orwell.freenode.net sets mode: +v kota_19:23
*** rcernin has joined #openstack-swift19:23
*** philipw_ has joined #openstack-swift19:23
*** j_king has joined #openstack-swift19:23
*** gabor_antal has joined #openstack-swift19:23
*** david-lyle has joined #openstack-swift19:23
*** ChanServ has joined #openstack-swift19:23
*** vern has joined #openstack-swift19:23
*** StraubTW has joined #openstack-swift19:23
*** zigo has joined #openstack-swift19:23
*** tamizh_geek has joined #openstack-swift19:23
*** HenryG has joined #openstack-swift19:23
*** kozhukalov has joined #openstack-swift19:23
*** dstepanenko_ has joined #openstack-swift19:23
*** zacksh has joined #openstack-swift19:23
*** rsFF has joined #openstack-swift19:23
*** cebruns has joined #openstack-swift19:23
*** amitkqed has joined #openstack-swift19:23
*** stevemar_ has joined #openstack-swift19:23
*** aj701 has joined #openstack-swift19:23
*** MVenesio has joined #openstack-swift19:23
*** Suyi_ has joined #openstack-swift19:23
*** tqtran has joined #openstack-swift19:23
*** dmorita has joined #openstack-swift19:23
*** _JZ_ has joined #openstack-swift19:23
*** Jeffrey4l_ has joined #openstack-swift19:23
*** ndk- has joined #openstack-swift19:23
*** balajir has joined #openstack-swift19:23
*** timburke has joined #openstack-swift19:23
*** gmmaha has joined #openstack-swift19:23
*** bapalm has joined #openstack-swift19:23
*** mmotiani has joined #openstack-swift19:23
*** furlongm has joined #openstack-swift19:23
*** ajiang has joined #openstack-swift19:23
*** okdas_ has joined #openstack-swift19:23
*** matt6434 has joined #openstack-swift19:23
*** itlinux has joined #openstack-swift19:23
*** neonpastor has joined #openstack-swift19:23
*** wasmum has joined #openstack-swift19:23
*** hogepodge has joined #openstack-swift19:23
*** peterlisak has joined #openstack-swift19:23
*** dabukalam_ has joined #openstack-swift19:23
*** saltsa has joined #openstack-swift19:23
*** timss has joined #openstack-swift19:23
*** madorn has joined #openstack-swift19:23
*** markstur has joined #openstack-swift19:23
*** redbo has joined #openstack-swift19:23
*** orwell.freenode.net sets mode: +ovv ChanServ timburke redbo19:23
*** jamielennox has joined #openstack-swift19:23
*** CrackerJackMack has joined #openstack-swift19:23
*** Guest66666 has joined #openstack-swift19:23
*** MooingLemur has joined #openstack-swift19:23
*** jamiec has joined #openstack-swift19:23
*** mathiasb has joined #openstack-swift19:23
*** tanee has joined #openstack-swift19:23
*** charz has joined #openstack-swift19:23
*** rickflare has joined #openstack-swift19:23
*** logan- has joined #openstack-swift19:23
*** chmouel has joined #openstack-swift19:23
*** dosaboy has joined #openstack-swift19:23
*** mmotiani is now known as 1JTAA9HZM19:23
*** hugokuo has joined #openstack-swift19:23
*** torgomatic_ has joined #openstack-swift19:23
*** openstackgerrit has joined #openstack-swift19:23
*** mmotiani has joined #openstack-swift19:23
*** lifeless has joined #openstack-swift19:23
*** portante has joined #openstack-swift19:23
*** kencjohnston has joined #openstack-swift19:23
*** swifterdarrell has joined #openstack-swift19:23
*** a1|away has joined #openstack-swift19:23
*** clayg has joined #openstack-swift19:23
*** treyd has joined #openstack-swift19:23
*** mlanner has joined #openstack-swift19:23
*** jroll has joined #openstack-swift19:23
*** EmilienM has joined #openstack-swift19:23
*** orwell.freenode.net sets mode: +vv swifterdarrell clayg19:23
*** mvk has joined #openstack-swift19:23
*** hoonetorg has joined #openstack-swift19:23
*** klrmn has joined #openstack-swift19:23
*** hurricanerix has joined #openstack-swift19:23
*** bikmak has joined #openstack-swift19:23
*** mminesh has joined #openstack-swift19:23
*** mcarden has joined #openstack-swift19:23
*** sileht has joined #openstack-swift19:23
*** alpha_ori has joined #openstack-swift19:23
*** clayg has quit IRC19:23
*** ningliu has joined #openstack-swift19:23
*** andymccr has joined #openstack-swift19:23
*** pdardeau has joined #openstack-swift19:23
*** bkopilov has joined #openstack-swift19:23
*** d0ugal has joined #openstack-swift19:23
*** darrenc has joined #openstack-swift19:23
*** ejat has joined #openstack-swift19:23
*** tristanC has joined #openstack-swift19:24
*** cebreidian has joined #openstack-swift19:24
*** clarkb has joined #openstack-swift19:24
*** Raymii has joined #openstack-swift19:24
*** sudorandom has joined #openstack-swift19:24
*** cdelatte has joined #openstack-swift19:24
*** calebb has joined #openstack-swift19:24
*** jaakkos has joined #openstack-swift19:24
*** tdasilva has joined #openstack-swift19:24
*** briancline has joined #openstack-swift19:24
*** timur has joined #openstack-swift19:24
*** orwell.freenode.net sets mode: +v tdasilva19:24
*** acoles has quit IRC19:24
*** acoles has joined #openstack-swift19:24
*** niven.freenode.net sets mode: +v acoles19:24
*** clayg has joined #openstack-swift19:24
*** ChanServ sets mode: +v clayg19:24
*** kozhukalov has quit IRC19:24
claygwell that's marginally better19:24
*** AndyWojo has quit IRC19:24
claygzaitcev: I really was just trying out my vim-go plugins - redbo said it was ok19:24
* clayg has no idea what's *go*ing on ;)19:24
*** ober37 has joined #openstack-swift19:24
*** bkeller` has joined #openstack-swift19:24
zaitcevclarkb: It's fine, just a reminder that they have their equivalent of pep8, built right into the toolchain.19:24
*** tmoreira has joined #openstack-swift19:24
redboYeah, the CI should have caught the lack of formatting, but it's gotten so unstable I kind of ignore it.  I need to take care of it.19:25
*** jrichli has joined #openstack-swift19:25
redboI wish I could just use travisci or whatever.19:26
*** ChubYann has joined #openstack-swift19:26
*** ChanServ sets mode: +v jrichli19:26
*** sanchitmalhotra has joined #openstack-swift19:26
*** briancurtin has joined #openstack-swift19:38
*** blair has joined #openstack-swift19:39
tqtranhi all, have a question. https://github.com/openstack/swift/blob/master/test/unit/common/middleware/test_tempauth.py#L1090 the config seems to have no effect at all, why do we include it in the tests?19:40
*** sgundur1 has joined #openstack-swift19:50
*** zul has joined #openstack-swift19:52
*** manous has joined #openstack-swift19:52
*** serverascode has joined #openstack-swift19:54
*** DuncanT has joined #openstack-swift19:59
*** asettle has joined #openstack-swift19:59
*** amit213 has joined #openstack-swift19:59
*** kong has joined #openstack-swift20:01
*** asettle has quit IRC20:02
*** ogelbukh has joined #openstack-swift20:02
*** kozhukalov has joined #openstack-swift20:02
*** asettle has joined #openstack-swift20:02
*** cargonza has joined #openstack-swift20:03
*** nottrobin has joined #openstack-swift20:03
*** janonymous has joined #openstack-swift20:05
*** clayg has quit IRC20:06
*** clayg has joined #openstack-swift20:06
*** ChanServ sets mode: +v clayg20:06
*** ediardo has joined #openstack-swift20:07
*** itlinux has quit IRC20:08
*** pcaruana has quit IRC20:11
claygtqtran: looks like cruft to me20:12
tqtrangreat! i'll remove them in my refactor patch :D20:13
claygtqtran: i mean it's sorta nuts the way the test is setup to preseed the request environ cache with the fake cache that has the token already cached with the grouops and stuff20:13
*** zul has quit IRC20:14
tqtranright, but the conf doesnt really play into the outcome, so not sure why we even bother since it just introduces "cruft"20:14
claygsure20:14
claygi think some of the other TestCase classes in the module do things a bit different and acctually use the configured users20:15
claygso maybe it was cargo culted from there?20:15
tqtranyeah thats entirely possible20:15
*** AndyWojo has joined #openstack-swift20:18
*** itlinux has joined #openstack-swift20:21
notmynamehttps://wiki.openstack.org/wiki/Swift/PriorityReviews has been updated with a more restricted set of patches to focus on to get into the release20:33
*** asettle has quit IRC20:36
*** zul has joined #openstack-swift20:37
*** porunov has joined #openstack-swift20:41
*** porunov has quit IRC20:41
*** manous has quit IRC20:43
*** asettle has joined #openstack-swift20:55
*** asettle has quit IRC21:02
*** asettle has joined #openstack-swift21:05
*** sudorandom has quit IRC21:11
*** sudorandom has joined #openstack-swift21:12
*** ober37 has quit IRC21:21
*** ober37 has joined #openstack-swift21:22
claygirc is being weird/quiet for me today21:25
*** ober37 has quit IRC21:27
*** ediardo has quit IRC21:27
*** janonymous has quit IRC21:27
*** kozhukalov has quit IRC21:27
*** amit213 has quit IRC21:27
*** blair has quit IRC21:27
*** tristanC has quit IRC21:27
*** cebreidian has quit IRC21:27
*** Raymii has quit IRC21:27
*** clarkb has quit IRC21:27
*** cdelatte has quit IRC21:27
*** calebb has quit IRC21:27
*** jaakkos has quit IRC21:27
*** sgundur1 has quit IRC21:27
*** tdasilva has quit IRC21:27
*** briancline has quit IRC21:27
*** timur has quit IRC21:28
*** gyee has joined #openstack-swift21:28
*** briancline has joined #openstack-swift21:28
*** tristanC has joined #openstack-swift21:28
*** clarkb has joined #openstack-swift21:28
*** calebb has joined #openstack-swift21:28
*** cebreidian has joined #openstack-swift21:28
*** cdelatte has joined #openstack-swift21:28
*** Raymii has joined #openstack-swift21:28
*** ober37 has joined #openstack-swift21:30
*** ediardo has joined #openstack-swift21:32
*** timur_ has joined #openstack-swift21:33
*** matt6434 is now known as mattoliverau21:33
*** ChanServ sets mode: +v mattoliverau21:33
*** kozhukalov has joined #openstack-swift21:34
*** amit213 has joined #openstack-swift21:34
mattoliverauMorning21:34
*** jlwhite has joined #openstack-swift21:36
*** janonymous has joined #openstack-swift21:36
jlwhitetimburke: Hi, I have a question about python-swiftclient21:37
timburkehey, i know about that!21:37
timburkejlwhite: what's the question?21:37
*** vint_bra has quit IRC21:41
clayglol21:43
claygredbo: what was i going to do about golang and privledge drops again?21:43
notmynamenow I'm curious what Alexandr Porunov's nick is in IRC (from the ML) ;-)21:46
*** asettle has quit IRC21:53
zaitcevhttps://www.linkedin.com/in/%D0%B0%D0%BB%D0%B5%D0%BA%D1%81%D0%B0%D0%BD%D0%B4%D1%80-%D0%BF%D0%BE%D1%80%D1%83%D0%BD%D0%BE%D0%B2-53861a7921:55
notmynamezaitcev: I can't read that. but I'm guessing you can21:55
claygyou guys are creepy ;)21:56
zaitcevWell, he's put OpenStack Swift onto his LinkedIn cred, which seems A Bit Too Soon to me.21:57
notmynamehe said he was joining IRC. I just wanted to say hi21:57
claygrofl21:57
claygterrible21:57
claygjust terrible21:57
claygzaitcev: have you checked if hummingbird supports config directories lately?21:57
notmynameI think the 4k video use case is awesome21:57
*** jaakkos has joined #openstack-swift21:58
claygzaitcev: do you care about config directories?  IME packers like config directories for stuff sometimes21:58
claygnotmyname: it's all about 16K video now21:58
zaitcevclayg: There is some code there to support them, which reads each file with standard config reader then merges values. But I did not test it.21:58
notmynameclayg: *mind blown*21:58
*** myoonnyc has joined #openstack-swift22:01
zaitcevclayg: I do not use config directories in packaging of Swift in RDO, at least not yet. But in general, it's much easier to add a file to /etc/foo.conf.d than to sed/ed/awk a configuration file and hope for the best.22:01
claygredbo: k, yeah so my `hummingbird object-replicator -c /blah/blah.conf` def does *not* honor the "user" value in my config - so I think I can fix that in my shim22:01
clayg... but idk what to do if someone tries to set bind_port to something privledged?22:02
claygfor me I'm happy to call it a limitation of the object-replicator that it can't bind to a privledged port unless you setcap something something?22:03
zaitcevHmm. I never thought to consider what happens. Most daemons allow it, unless ran by a user account from the start, because they drop privileges after bind().22:04
claygi put proxies on < 1024 in a few supported configs but mostly that's where I've got my ssl terminators22:04
claygzaitcev: redbo tells me golang can't drop privileges?22:04
zaitcevclayg: I just don't know anything about it, sorry.22:05
redboYeah, it's a whole thing about it can't drop privileges in all of its system threads.22:05
claygi started to read this bug fell asleep -> https://github.com/golang/go/issues/143522:05
claygredbo: neway - i'm just gunna do it in the shim before I execvp22:06
*** tdasilva has joined #openstack-swift22:06
claygso by the time you try to bind you're going to be the configured user - so .... like don't set a stupid port22:06
claygor it probably won't work22:06
redboYeah.  It kind of sucks.22:07
*** blair has joined #openstack-swift22:08
claygredbo: I bet rust does it *perfectly*22:08
zaitcevSocket activation FTW anyway22:11
zaitcevI'd try to support that firs... waaaait a second. All the people who migrate from swift-init22:12
*** itlinux has quit IRC22:13
openstackgerritClay Gerrard proposed openstack/swift: fix string formatting in manager  https://review.openstack.org/36972822:15
redboProbably22:15
*** itlinux has joined #openstack-swift22:16
redboI mean.. we could make swift-init bind the port.22:16
redboPass it in on the command line.22:16
redboThat can wait for v222:17
claygredbo: no fuck that22:17
claygsetcap is a thing22:17
claygor iptables22:17
claygreally I don't care22:17
claygdon't bind your object server to privledged port - this is fine for me22:17
*** tongli has quit IRC22:22
*** cdelatte has quit IRC22:24
*** itlinux has quit IRC22:26
*** chsc has quit IRC22:37
*** itlinux has joined #openstack-swift22:39
timur_I can't reconcile the Swift returned value of Last-Modified Date with the documentation. The documented value is in the ISO-8601 format (e.g. 2015-08-27T09:49:58-05:00). When doing a HEAD on an object, I get back a string like this "Mon, 12 Sep 2016 22:38:14 GMT". I'm looking at the documentation here:22:43
patchbothttps://review.openstack.org/#/c/2016/ - nova - Replace more cases of instance ids with uuids (ABANDONED)22:43
timur_http://developer.openstack.org/api-ref/object-storage/index.html?expanded=show-object-metadata-detail22:43
timur_is the documentation wrong or is there a bug in the last-modified date presentation?22:43
claygthose docs anchor links never work for me22:45
timur_clayg: click on "detail" next to "HEAD" under the Objects section22:46
clayghttp://developer.openstack.org/api-ref/object-storage/index.html?expanded=show-object-metadata-detail#show-object-metadata22:46
timur_nice! even better22:46
timur_I'm so confused by this. Like, the documentation contradicts itself?22:47
*** dmorita has quit IRC22:47
timur_because it lists an example that clearly has a not-ISO-8601 Last-Modified date?22:47
timur_but then continues to describe the ISO-8601 format in the Response Parameters section22:48
claygit's also listed twice?22:48
timburkeso jlwhite just led me to notice that run_wsgi can leave a lingering child process. if you're in an interactive session and hit ctrl-c, we break out of our loop but never clean up the children -- see https://github.com/openstack/swift/blob/2.9.0/swift/common/wsgi.py#L96322:48
timburkeso my question is, should we call kill_children() or hup()?22:48
*** itlinux has quit IRC22:49
timur_clayg: that's interesting too! "Date" is _also_ documented as ISO 860122:49
timur_(in the Response parameters section)22:49
claygtimburke: neither?  the parent process is supposed to be able to exit and leave the children to die when their sockets are closed up22:50
claygor something22:50
claygwhy are we calling run_wsgi interactively - is there a use-case here?22:50
timburkebin/swift-proxy-server22:51
timburkemaybe we ought to convert devstack to use swift-init, though22:51
timur_I realized that I care about this because I need to stuff the date into elastic search and the date format Swift returns by default is not something that it can parse, so I'll need to read it properly (it understands ISO 8601 though)22:51
claygtimur_: just use x-timestamp - you can also file doc bugs if you want22:52
claygtimburke: undoubtably - i had recently discovered that a number of other services in devstack configure and start themselves quite reasonably and then their screen sessions just tail -f /the/path/to/the/logs22:53
*** sanchitmalhotra1 has joined #openstack-swift22:54
*** sanchitmalhotra has quit IRC22:55
*** sanchitmalhotra1 is now known as sanchitmalhotra22:55
timur_clayg: yea, seems like a documentation bug. I found the HTTP RFC that defines the HTTP date for the Last-Modified header, so Swift is doing the right thing22:56
*** StraubTW has quit IRC22:56
claygtimburke: I still don't really see the problem - i can start `swift-proxy-server /path/to/proxy/config` and ctrl-c terminates all the children22:56
notmynametimur_: IIRC there was some difference between swift and the rfc that broke some clients, so we reverted it. timezone designation or something like that22:57
*** itlinux has joined #openstack-swift22:57
*** MVenesio has quit IRC22:58
notmynameI don't remember. it's faded into myth and legend by now22:59
notmyname"one time, we made a change and clients broke! we had to revert that and swore, 'never again!'. let this be a lesson to you, and pass it on to your children"22:59
timur_notmyname: ha! ok23:01
*** csmart has joined #openstack-swift23:01
notmyname"in the days of openstack public clouds, when giants roamed the earth..."23:01
zaitcevBut it was twice23:02
zaitcevThe date format was one, Etag/ETag was the other.23:02
*** tongli has joined #openstack-swift23:02
notmynameoh yeah. the E[Tt]ag one. *sigh*23:03
claygnotmyname: don't forget the xml quotes!23:03
*** tongli has quit IRC23:04
zaitcevI don't recally that one actually breaking anything, but I can totally believe it.23:04
notmyname"children, listen to me! gather round and learn from our mistakes: you'll never NOT break clients, so just do whatever you think is right"23:04
notmyname;-)23:04
claygtimur_: do we explictly catch the KeyboardInterblahbalhError?23:04
claygsorry timburke ^23:04
claygcause like if want to blast SIGTERMS out to random processes when running swift proxy interactively in a screen session or whatever else might make devstack happy - like no one would notice or care - no one starts swift like that23:05
claygtimburke: OTOH - if you wanna jump on the make devstack+swift awesome train with me - i'm currently working on https://github.com/swiftstack/vagrant-devstack-all-in-one/issues/123:06
claygit's slow going -> https://gist.github.com/clayg/e0f79ef1c892f617fd4179b2b65a6b3423:06
claygi can push up what i have so far tho...23:07
timburketimur_: clayg: that date format thing is a known documentation bug, but it should be fixed Real Soon Now because donagh's great -- https://review.openstack.org/#/c/354767/23:08
patchbotpatch 354767 - swift - Corrections for the API specifications in api-ref23:08
claygrofl - it may not look like much - but trust me - everyone single one of those lines is *gold* -> https://github.com/swiftstack/vagrant-devstack-all-in-one/commit/6c45a46dd10bca2814f1385df22f97cd2965203123:08
claygyay donagh!23:08
claygtimur_: ^^^23:08
timur_great!23:09
clayghttp://docs-draft.openstack.org/67/354767/12/check/gate-swift-api-ref/f18f452//api-ref/build/html/?expanded=show-object-metadata-detail#show-object-metadata <- is that better?23:09
claygstill says Last-Modified twice...23:09
timur_yep!23:10
timburkebut at least it has the right format23:10
clayg+1 for effort!23:11
*** cdelatte has joined #openstack-swift23:20
*** cdelatte has quit IRC23:21
*** rcernin has quit IRC23:23
claygzaitcev: i get the distinct feeling that hummingbird's replicator doesn't log too much?23:24
zaitcevclayg: Don't ask, I only have "hummingbird object"23:25
*** _JZ_ has quit IRC23:26
*** Jeffrey4l_ is now known as Jeffrey4l23:30
claygman i sure wish i would fix this -> https://github.com/swiftstack/vagrant-swift-all-in-one/issues/4523:32
*** kei_yama has joined #openstack-swift23:32
*** zaitcev has quit IRC23:33
claygredbo: damnit - dfg and nadeem are still not in channel!23:36
claygwell, anyway - moving the repconn server to the replicators may have been a terrible idea23:40
claygit breaks the ability to `swift-init object-replciator once -nv` or however you hummingbird cats you used to spell that23:41
claygthe point is that one daemon can "once" and shutdown before the others are finished with whatever they might need to talk to that server about23:41
claygnot that any of this logs :'(23:42
notmynameunless "once" is defined by "only one time figure out your local work to do and do it".23:42
notmynamebut yeah, getting it to still respond to other things in the cluster is important23:43
claygnotmyname: well like we never "stop" rsync23:43
notmynameyeah23:43
notmynamewhat's the hbird api commend to tell it what parts to process?23:44
claygbut we do have the pre-req on the REPLICATE verb - so all of our probetests do just so happen to assume that the object-server can be running and an replicator once can still "push" to any other node23:44
notmynamecommend? is that like "POST /doing/a/great/job/" ?23:44
claygthe api or the command line args?  the -partitions arg on the cli (why goland uses single dashes...) is basically the same as the python guys23:45
notmynamewhat if `swift-init ... once` isn't a thing? but instead we keep the replicators running all the time (ie to respond) but we have something different to tell it what to do?23:45
claygoh, well maybe - i mean if we have to completly rewrite the probe tests where they can't possibly work with either configuration that'll make it harder - but it could be ok too23:46
notmynameso, fantasy world where we get to invent new things ... what if we had `swift-init start ...` and then swift-ctl or swift-mgr or something to feed the beast23:46
notmynameyeah. everything works just fine as long as we rewrite everything so it doesn't work with the earlier config23:46
claygyeah i'm just curious about nadeems workflows when he was working on all this23:48
claygBUT HE'S NOT IN CHANNEL ;)23:49
clayglol at that one time you forgot to recomplie hummingbird (that *one* time; sigh)23:54
openstackgerritMerged openstack/swift: fix string formatting in manager  https://review.openstack.org/36972823:57

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