Friday, 2016-03-18

*** haomaiwang has quit IRC00:01
*** haomaiwang has joined #openstack-swift00:01
*** garthb has quit IRC00:14
*** lyrrad has quit IRC00:24
*** daemontool has joined #openstack-swift00:28
*** NM has quit IRC00:33
*** nadeem has joined #openstack-swift00:34
*** nadeem has quit IRC00:39
*** nadeem has joined #openstack-swift00:40
*** rickyrem has joined #openstack-swift00:44
kota_good morning00:44
openstackgerritMerged openstack/swift: Update ring overview for reuse of device ids  https://review.openstack.org/29431000:48
*** daemontool has quit IRC00:48
*** takashi has joined #openstack-swift01:00
takashigood morning01:00
pdardeautakashi: good morning01:00
*** haomaiwang has quit IRC01:01
*** Jeffrey4l_ has joined #openstack-swift01:01
*** haomaiwa_ has joined #openstack-swift01:01
*** CaioBrentano has joined #openstack-swift01:03
*** cdelatte has quit IRC01:07
takashipdardeau: o/01:08
*** nadeem has quit IRC01:08
*** CaioBrentano has quit IRC01:08
*** StraubTW has quit IRC01:11
*** nadeem has joined #openstack-swift01:11
*** mingdang1 has joined #openstack-swift01:11
*** MVenesio has joined #openstack-swift01:15
*** MVenesio has quit IRC01:19
*** nadeem has quit IRC01:20
*** nadeem has joined #openstack-swift01:20
*** nadeem has quit IRC01:21
*** NM has joined #openstack-swift01:22
*** NM has quit IRC01:25
*** dmorita has quit IRC01:33
*** dmorita has joined #openstack-swift01:35
*** haomaiwa_ has quit IRC01:37
*** dmorita has quit IRC01:40
*** baojg has joined #openstack-swift01:54
*** itlinux has joined #openstack-swift02:12
*** haomaiwang has joined #openstack-swift02:20
*** itlinux has quit IRC02:23
*** itlinux has joined #openstack-swift02:29
*** rickyrem has quit IRC02:38
*** itlinux has quit IRC02:40
*** links has joined #openstack-swift02:40
*** haomaiwang has quit IRC02:41
*** sgundur has left #openstack-swift02:44
*** haomaiwang has joined #openstack-swift02:45
*** sgundur has joined #openstack-swift02:46
*** sgundur has left #openstack-swift02:47
*** haomaiwang has quit IRC03:01
*** haomaiwang has joined #openstack-swift03:01
*** nadeem has joined #openstack-swift03:11
*** MVenesio has joined #openstack-swift03:16
*** MVenesio has quit IRC03:20
*** itlinux has joined #openstack-swift03:25
*** itlinux has quit IRC03:40
*** proteusguy__ has quit IRC03:41
*** itlinux has joined #openstack-swift03:47
*** itlinux has quit IRC03:48
*** Jeffrey4l_ is now known as Jeffrey4l03:49
*** proteusguy__ has joined #openstack-swift03:53
mahaticgood morning03:56
*** haomaiwang has quit IRC04:01
*** haomaiwang has joined #openstack-swift04:01
notmynamehello mahatic04:01
mahatichello04:01
mahaticnotmyname: good evening04:02
takashimahatic, notmyname: good morning and evening04:09
*** treaki__ has joined #openstack-swift04:28
*** baojg has quit IRC04:28
*** treaki_ has quit IRC04:32
*** baojg has joined #openstack-swift04:40
*** nadeem has quit IRC04:55
*** lifeless has quit IRC04:55
*** lifeless has joined #openstack-swift04:56
*** haomaiwang has quit IRC05:01
*** haomaiwang has joined #openstack-swift05:01
*** mingdang1 has quit IRC05:03
*** nadeem has joined #openstack-swift05:03
*** itlinux has joined #openstack-swift05:05
*** cebruns has quit IRC05:08
openstackgerritMerged openstack/swift: Add a note for functional tests with Keystone.  https://review.openstack.org/26266405:08
*** jrichli has quit IRC05:09
*** cebruns has joined #openstack-swift05:09
*** jrichli has joined #openstack-swift05:14
*** MVenesio has joined #openstack-swift05:16
*** MVenesio has quit IRC05:21
*** zaitcev has quit IRC05:27
*** ppai has joined #openstack-swift05:47
*** trifon has joined #openstack-swift05:52
*** asettle has quit IRC05:57
*** nadeem has quit IRC05:59
*** nadeem has joined #openstack-swift05:59
*** haomaiwang has quit IRC06:01
*** haomaiwang has joined #openstack-swift06:01
*** pcaruana has quit IRC06:18
*** aswadr_ has joined #openstack-swift06:21
*** itlinux has quit IRC06:26
*** ChubYann has quit IRC06:27
*** dmorita has joined #openstack-swift06:33
*** rcernin has joined #openstack-swift06:37
*** dmorita has quit IRC06:38
*** mahatic has quit IRC06:46
*** mahatic has joined #openstack-swift06:47
*** baojg has quit IRC06:55
*** haomaiwang has quit IRC07:01
*** haomaiwang has joined #openstack-swift07:01
*** klrmn has quit IRC07:06
*** Zyric has quit IRC07:07
*** mingdang1 has joined #openstack-swift07:09
*** kaleta has joined #openstack-swift07:14
*** baojg has joined #openstack-swift07:14
*** peterlisak has joined #openstack-swift07:14
*** onovy has joined #openstack-swift07:14
*** MVenesio has joined #openstack-swift07:17
*** mingdang1 has quit IRC07:19
*** MVenesio has quit IRC07:22
*** Zyric has joined #openstack-swift07:22
*** baojg has quit IRC07:25
*** BobBall_ has joined #openstack-swift07:33
*** BobBall_ is now known as Guest9218507:34
*** BobBall has quit IRC07:36
*** BobBall has joined #openstack-swift07:38
*** Guest92185 has quit IRC07:38
*** BobBall is now known as Guest3632607:39
openstackgerritKota Tsuyuzaki proposed openstack/swift: Don't merge: Replace duplicated code by iotop lib  https://review.openstack.org/29443207:43
kota_acoles: looking at patch 238799, looks like we can use iotop lib instead of using the similar (duplicated? inspired?) code, right?07:45
patchbotkota_: https://review.openstack.org/#/c/238799/ - swift - Change schedule priority of daemon/server in config07:45
kota_acoles: I tried to write for that at patch 294432 but I'm still not sure if it is right way (or not) and also I don't know if python have also the library for 'nice'.07:46
patchbotkota_: https://review.openstack.org/#/c/294432/ - swift - Don't merge: Replace duplicated code by iotop lib07:46
kota_If you could suggest something about that, it will be really helpful for me :)07:47
kota_acoles: ^^07:47
*** pcaruana has joined #openstack-swift07:57
*** mmcardle has joined #openstack-swift07:59
*** haomaiwang has quit IRC08:01
*** haomaiwang has joined #openstack-swift08:01
*** jmccarthy has quit IRC08:05
*** jmccarthy has joined #openstack-swift08:06
*** geaaru has joined #openstack-swift08:12
*** rledisez has joined #openstack-swift08:16
openstackgerritKota Tsuyuzaki proposed openstack/swift: Don't merge: Replace duplicated code by iotop lib  https://review.openstack.org/29443208:16
kota_ah..."import iotop" makes all swiftstack tox tests as failed... :(08:17
*** silor has joined #openstack-swift08:17
*** mingdang1 has joined #openstack-swift08:19
*** tesseract has joined #openstack-swift08:20
*** tesseract is now known as Guest5995408:21
kota_ugh... it looks to need psutil08:21
*** mingdang1 has quit IRC08:24
*** Guest59954 is now known as new08:25
*** new has quit IRC08:26
*** tesseract- has joined #openstack-swift08:26
*** tmoreira has quit IRC08:32
openstackgerritKota Tsuyuzaki proposed openstack/swift: Don't merge: Replace duplicated code by iotop lib  https://review.openstack.org/29443208:33
*** flaper87 has quit IRC08:33
*** tmoreira has joined #openstack-swift08:34
*** flaper87 has joined #openstack-swift08:34
kota_ah, perhaps it is included distro package.08:39
*** takashi has quit IRC08:45
charzkota_: that's odd, did you run `pip install iotop` or something?08:53
charzkota_: I can't find `iotop` on pypi repo.08:53
*** ho_ has quit IRC08:55
*** cdelatte has joined #openstack-swift08:59
*** daemontool has joined #openstack-swift08:59
*** haomaiwang has quit IRC09:01
*** haomaiwang has joined #openstack-swift09:01
*** jmccarthy1 has joined #openstack-swift09:01
*** stantonnet has quit IRC09:02
*** stantonnet has joined #openstack-swift09:06
kota_charz: you're right, I got mistake.09:07
kota_charz: I'm still searching the dependencies09:08
charzkota_: How to install that? Or it just a simple python file? iotop.py?09:08
kota_maybe, from package-repo09:09
kota_i.e. sudo apt-get install iotop09:09
kota_but not sure right now, I noticed, it's already in my local before I was trying the patch09:09
openstackgerritKota Tsuyuzaki proposed openstack/swift: Don't merge: Replace duplicated code by iotop lib  https://review.openstack.org/29443209:10
kota_charz: I'm sure for now, that is inlucded iotop in package repo.09:13
charzkota_: maybe you just `pip install https://github.com/kahing/iotop`?09:13
kota_charz: I made sure that will be unavailable via 'apt-get remove iotop' and then 'apt-get install iotop' enabes to be available.09:13
kota_charz: might be09:14
kota_ah, it looks fork09:14
*** acoles_ is now known as acoles09:15
charzkota_: https://gist.github.com/charz/c15c231cb0a2a028de6709:16
charzThat'll install iotop to system path, when we run tests with tox, that won't able to import that.09:17
*** MVenesio has joined #openstack-swift09:18
openstackgerritNguyen Hung Phuong proposed openstack/python-swiftclient: Removes redundant "to"  https://review.openstack.org/29446909:18
kota_charz: exactly09:19
kota_charz: so good catch... no idea to add the dependency which is not mantained in pypi...09:20
kota_maybe it's the reason, Peter didn't so in the patch 238799.09:21
*** silor has quit IRC09:21
patchbotkota_: https://review.openstack.org/#/c/238799/ - swift - Change schedule priority of daemon/server in config09:21
kota_acoles: good morning09:22
acoleskota_: i doubt i will get chance to look at that today ^^ I am only here for a short time09:22
charzkota_: maybe ask the author of iotop, is possible upload the it to pypi09:22
charzacoles: good morning!!09:22
acoleskota_: but thanks for looking into it09:22
acolescharz: hi!09:22
*** MVenesio has quit IRC09:22
kota_acoles: alright, I think you can help me to solve that :)09:23
kota_acoles: and sorry, I might make it tangled :/09:23
kota_charz: sounds good.09:24
kota_and sorry, I'm running out of my time at office today. So I will look at that again in the weekend (or at the beggining of next week, at least).09:26
*** jistr has joined #openstack-swift09:28
*** haomaiwang has quit IRC09:30
*** haomaiwang has joined #openstack-swift09:32
*** links has quit IRC09:37
*** Guest36326 is now known as BobBall09:40
*** panda has quit IRC09:40
*** panda has joined #openstack-swift09:41
*** aswadr_ has quit IRC09:50
*** baojg has joined #openstack-swift09:53
*** haomaiwang has quit IRC10:01
*** mvk has joined #openstack-swift10:01
*** haomaiwa_ has joined #openstack-swift10:01
*** jordanP has joined #openstack-swift10:11
*** km has quit IRC10:14
*** links has joined #openstack-swift10:34
*** mvk has quit IRC10:35
openstackgerritMerged openstack/python-swiftclient: Initialize delete_object mock *before* creating all the threads  https://review.openstack.org/29431610:43
openstackgerritMerged openstack/swift: Make keystone middleware options consistent in docs  https://review.openstack.org/29338210:43
*** flaper87 has quit IRC10:49
*** flaper87 has joined #openstack-swift10:49
openstackgerritMerged openstack/swift: Keystone middleware deprecated option is_admin removed  https://review.openstack.org/25815810:51
*** haomaiwa_ has quit IRC10:52
*** baojg has quit IRC10:58
*** haomaiwa_ has joined #openstack-swift10:58
*** BobBall_ has joined #openstack-swift10:59
*** BobBall_ is now known as Guest8647310:59
*** haomaiwa_ has quit IRC11:01
*** haomaiwang has joined #openstack-swift11:01
*** BobBall has quit IRC11:02
*** mvk has joined #openstack-swift11:02
*** Guest86473 has quit IRC11:03
openstackgerritAlistair Coles proposed openstack/swift: decouple versioned writes from COPY  https://review.openstack.org/26017911:04
*** BobBall has joined #openstack-swift11:04
acolestdasilva: ^^ I addressed torgomatic's comment, please check it over11:04
*** BobBall is now known as Guest7227011:04
*** Guest72270 is now known as BobBall211:07
*** wanghua has quit IRC11:14
*** MVenesio has joined #openstack-swift11:19
*** cdelatte has quit IRC11:21
openstackgerritAlistair Coles proposed openstack/swift: Re-format the SLO manifest file on new multipart-manifest GET call  https://review.openstack.org/26390211:22
*** MVenesio has quit IRC11:23
*** haomaiwang has quit IRC11:26
*** juzuluag has joined #openstack-swift11:27
pdardeaugood morning11:37
*** Razva has left #openstack-swift11:43
*** cdelatte has joined #openstack-swift11:51
*** nadeem has quit IRC11:53
mahaticpdardeau: morning already?12:04
pdardeaumahatic: kinda, only 1 cup of coffee so far :)12:05
mahaticpdardeau: :) good morning12:05
*** sbezverk has joined #openstack-swift12:12
*** MVenesio has joined #openstack-swift12:20
openstackgerritAlistair Coles proposed openstack/swift: Refactor server side copy as middleware  https://review.openstack.org/15692312:21
openstackgerritAlistair Coles proposed openstack/swift: Re-format the SLO manifest file on new multipart-manifest GET call  https://review.openstack.org/26390212:21
openstackgerritAlistair Coles proposed openstack/swift: decouple versioned writes from COPY  https://review.openstack.org/26017912:21
acoles^^ update to 156923, the others are simply rebases12:26
*** acoles is now known as acoles_12:27
*** delattec has joined #openstack-swift12:29
*** cdelatte has quit IRC12:33
*** diogogmt has quit IRC12:37
sbezverkGood morning.. I see an issue with the latest swift when keystone does not see authentication token in swift requests and denies authorization.. I checked config and it looks good to me, but I suspect I am missing something. I would appreciate if somebody could take a look at it here: http://paste.openstack.org/show/491086/12:39
*** diogogmt has joined #openstack-swift12:42
*** openstackgerrit has quit IRC12:48
*** openstackgerrit has joined #openstack-swift12:49
*** jistr has quit IRC12:51
*** jistr has joined #openstack-swift12:51
*** CaioBrentano has joined #openstack-swift12:51
*** alejandrito has joined #openstack-swift12:56
*** tesseract- has quit IRC12:59
*** diogogmt has quit IRC12:59
*** delatte has joined #openstack-swift13:03
*** BobBall2 is now known as BobBall13:06
*** delattec has quit IRC13:06
*** tesseract has joined #openstack-swift13:08
*** tesseract is now known as Guest8468813:08
*** angelique85 has joined #openstack-swift13:21
angelique85great thank you!13:21
*** links has quit IRC13:22
*** _JZ_ has joined #openstack-swift13:29
*** mmcardle has quit IRC13:30
*** dmorita has joined #openstack-swift13:34
*** mmcardle has joined #openstack-swift13:34
*** ametts has joined #openstack-swift13:37
*** dmorita has quit IRC13:38
*** panda has quit IRC13:40
*** panda has joined #openstack-swift13:41
*** dmellado is now known as dmellado|lunch13:44
*** haomaiwa_ has joined #openstack-swift13:55
*** ppai has quit IRC13:55
*** esker has joined #openstack-swift13:58
*** haomaiwa_ has quit IRC13:59
*** haomaiwang has joined #openstack-swift13:59
*** haomaiwang has quit IRC14:01
*** sgundur has joined #openstack-swift14:01
*** haomaiwang has joined #openstack-swift14:01
*** rcernin has quit IRC14:01
*** links has joined #openstack-swift14:09
*** diogogmt has joined #openstack-swift14:17
*** sheel has joined #openstack-swift14:18
sheelnotmyname:  hi there14:20
sheelnotmyname: need some pointers for https://bugs.launchpad.net/swift/+bug/155733714:21
openstackLaunchpad bug 1557337 in OpenStack Object Storage (swift) "Command "swift-init" is showing wrong output" [Undecided,Confirmed] - Assigned to Ankur Jain (j-ankur)14:21
*** arch-nemesis has joined #openstack-swift14:22
*** klamath has joined #openstack-swift14:23
*** dmellado|lunch is now known as dmellado14:24
*** sc has joined #openstack-swift14:32
*** baojg has joined #openstack-swift14:33
*** baojg has quit IRC14:35
*** pcaruana has quit IRC14:38
sbezverkAppreciate some attention to this bug https://bugs.launchpad.net/swift/+bug/1559148 it is kind of a show stopper for my activities..14:38
openstackLaunchpad bug 1559148 in OpenStack Object Storage (swift) "swift stat returns account not found" [Undecided,New]14:38
*** Guest84688 has quit IRC14:48
*** links has quit IRC14:54
*** links has joined #openstack-swift14:56
*** links has quit IRC14:57
*** ametts has quit IRC14:57
*** links has joined #openstack-swift14:57
*** rickyrem has joined #openstack-swift14:58
*** tesseract has joined #openstack-swift15:00
*** ajiang has left #openstack-swift15:00
*** tesseract is now known as Guest7688315:00
*** haomaiwang has quit IRC15:01
*** haomaiwa_ has joined #openstack-swift15:01
scIs there any way to manage quotas from Horizon?15:03
*** klrmn has joined #openstack-swift15:04
*** acoles_ is now known as acoles15:07
*** arch-nemesis has quit IRC15:08
*** angelique85 has quit IRC15:09
*** garthb has joined #openstack-swift15:13
*** trifon has quit IRC15:17
*** ppai has joined #openstack-swift15:18
*** ppai has quit IRC15:18
*** links has quit IRC15:19
acolessbezverk: do you have account_autocreate = true in your proxy-server conf ? http://docs.openstack.org/developer/swift/middleware.html#keystoneauth15:22
brianclinemorning folks, any feedback on https://review.openstack.org/292206 ?15:22
brianclineour ops team would love you very much15:23
*** itlinux has joined #openstack-swift15:25
sbezverk+acoles yes15:28
*** baojg has joined #openstack-swift15:30
sbezverk+acoles I tried all sorts of combinations still no go. anything else I should check?15:31
*** arch-nemesis has joined #openstack-swift15:32
*** david-lyle has quit IRC15:32
*** david-lyle has joined #openstack-swift15:33
*** daemontool has quit IRC15:35
acolessbezverk: this line from the swift CLI log that you posted in the bug report does not look right15:35
acolesINFO:swiftclient:REQ: curl -i http://192.168.80.205:8080 -I -H "X-Auth-Token: b036bb1aa2534ff79f2ec2ae332a4d7d"15:35
acolesthe url should end with /v1/AUTH_<a project id>15:36
acolescould be that the swift endpoints in the keystone service catalog are not correctly configured15:36
sbezverk+acoles I am using keystone v3 authentication, here is complete proxy-server.conf http://paste.openstack.org/show/491118/15:36
sbezverk+acoles from this paste you can see v3 api is called.. http://paste.openstack.org/show/491121/15:38
sbezverk+acoles when server catalog get created does it require exaplicitely specify --os-identity-api-version 3 ?15:40
*** admin6 has joined #openstack-swift15:40
acolessbezverk: i am referring to the step 3 in this guide http://docs.openstack.org/liberty/install-guide-ubuntu/swift-controller-install.html15:40
acolesthe curl command in this trace  http://paste.openstack.org/show/491121/ shows the url that swift client is attempting to use, which is what it gets from the keystone response body service catalog15:41
acolesthat url is wrong, which suggests that keystone service catalog may not be configured correctly15:42
admin6Hi there. Is it possible to change the number of replica of my container.builder ring? I’m not familiar with such manipulations.15:42
sbezverk+acoles I will check the link you pasted, thank you..15:43
*** baojg has quit IRC15:43
acolessbezverk: obviously, use the install guide for whatever version you have, the link i pasted is for liberty. pay particular attention to this step 'openstack endpoint create --region RegionOne \15:44
acoles  object-store public http://controller:8080/v1/AUTH_%\(tenant_id\)s'15:44
acolesthe %\(tenant_id\)s syntax tells keystone to insert the tenant id in the endpoint url15:45
acolesif you do 'openstack endpoint list' or somesuch command  you should be able to see what is set15:46
acolesdisclaimer: i don't use openstack cli / keystone commands often15:46
sbezverk+acoles I do :-) and you were right.  ee71fe84d4e747498ecb01ca2061db57 | RegionOne | swift        | object-store | True    | admin     | http://192.168.80.205:8080                  |15:47
sbezverkit seems there is a part missing15:47
acolessbezverk: ok. so when it is setup correctly and you do 'swift --debug stat' you should see a line similar to DEBUG:swiftclient:REQ: curl -i http://saio-1.localdomain:8080/v1/AUTH_cfb8d9d45212408b90bc0776117aec9e -I -H "X-Auth-Token: 683b693bb2d04cd0a53c27c4d9431bc9"15:49
acolessbezverk: note the /v1/AUTH_cfb8d9d45212408b90bc0776117aec9e at end of url15:49
*** itlinux has quit IRC15:51
sbezverk+acoles, got it trying to fix it, thank you very much..15:52
acolessbezverk: np. good luck.15:53
admin6ok, my command is (not documented in the swift-ring-builder man page) swift-ring-builder container.builder set_replicas 3.5  Sorry for the noise.15:54
*** david-lyle has quit IRC16:00
*** haomaiwa_ has quit IRC16:01
*** david-lyle has joined #openstack-swift16:01
*** haomaiwang has joined #openstack-swift16:01
*** itlinux has joined #openstack-swift16:02
*** jordanP has quit IRC16:10
*** jordanP has joined #openstack-swift16:11
*** klrmn1 has joined #openstack-swift16:16
*** klrmn has quit IRC16:18
*** rickyrem has quit IRC16:21
*** klrmn1 has quit IRC16:22
sbezverk+acoles it is better but still failure: Account HEAD failed: http://192.168.80.205:8080/v1/eb32edde924a49c998b5e6aca44f8063 403 Forbidden16:27
*** daemontool has joined #openstack-swift16:28
*** dmorita has joined #openstack-swift16:29
acolessbezverk: your url is still missing the reseller prefix AUTH_ - by default keystoneauth uses that prefix in front of the tenant_id, so endpoint should be http://controller:8080/v1/AUTH_%(tenant_id)s16:32
acolessbezverk: do 'swift --debug stat', copy the curl command from log, insert AUTH_ after ...8080/v1/  and before the tenant id, and see if that works16:33
*** chsc has joined #openstack-swift16:36
*** mittal has joined #openstack-swift16:37
mittalHi all. I am still struggling with my question :(16:38
mittalThe servers which are started with swift-init, on receiving SIGHUP behave differently than when the same server is started from systemctl command16:39
mittalEg. object-server on receiving SIGHUP exits and workers exit only after finishing the I/Os16:39
notmynamegood morning16:40
mittalThis is the case when I start object-server as swift-init object-server start16:40
mittalIf I do service object-server start, on receiving SIGHUP the parent exits. And workers, which are still busy, exit too since they receive a SIGTERM each16:41
*** jmccarthy1 has quit IRC16:44
*** pcaruana has joined #openstack-swift16:47
*** lyrrad has joined #openstack-swift16:51
*** ppai has joined #openstack-swift16:53
notmynamemailing list post this morning (subject: [openstack-dev][all][infra] revert new gerrit) that well-summarizes a lot of the issues we've had with gerrit too16:53
notmynamethe response is "nope, not gonna happen. either use gertty or write a new client on your own time"16:53
*** trifon has joined #openstack-swift17:00
*** haomaiwang has quit IRC17:01
*** haomaiwang has joined #openstack-swift17:01
*** tech004 has joined #openstack-swift17:02
*** rledisez has quit IRC17:04
*** zaitcev has joined #openstack-swift17:04
*** ChanServ sets mode: +v zaitcev17:04
*** mittal has left #openstack-swift17:04
tech004hi anyone please explain  servers_per_port option how can i run two port on same conf?17:05
zaitcevCreate two configurations in /etc/swift/xxxxxx-server/[12].conf17:08
tech004 from the docs same ip adress in a ring using 2 ports 6000 and 6001 m not getting it17:08
zaitcev6000 is object and 6001 is container, so two different servers and two different configuration files.17:08
claygi thought we were going to change all those ports17:09
*** daemontool_ has joined #openstack-swift17:10
notmynamepatch 27484017:11
patchbotnotmyname: https://review.openstack.org/#/c/274840/ - swift - change default ports for servers17:11
tech004  <zaitcev> object-server.conf.d/ ? like this17:11
zaitcevtech004: conventionally there's no .conf.d suffix. In any case, you use /etc/swift/object-server.conf with 6000 (or 6200 now) and /etc/swift/container-server.conf with 6001 (or 6201).17:12
*** daemontool has quit IRC17:12
tech004<zaitcev> http://docs.openstack.org/developer/swift/deployment_guide.html#server-per-port-configuration17:13
tech004I am reading this doc17:13
zaitcevGo go gadget git blame17:17
*** klrmn has joined #openstack-swift17:17
*** admin6 has left #openstack-swift17:17
zaitcevThat dubious piece of port allocation was written by Peter Lisak and approved by Sam and John. Way to go, guys17:18
zaitcevI wish I caught it.17:18
tech004Thanks guys if its  a mistake  please fix it or explain to me coz i dont know how to achieve it17:19
zaitcevWell, it's not a mistake per se. It is, however, an unconventional and confusing numbering.17:19
tech004<zaitcev> Great so its possible so how can i do it with my config I have old conf with same port for object server on four nodes17:21
tech004<zaitcev> if there is another doc I can look into it :)17:23
*** ppai has quit IRC17:27
zaitcevNo, it's fine. I was confused when I mentioned /etc/swift/object-server/1.conf and the like. Do everything as usual and don't set bind_port to any value in the conf. The servers should automatically find all listening ports from the ring for their node.17:27
zaitcevIf servers_per_port is set, that is.17:28
zaitcevAlso, I was looking at wrong commit for the doc.17:28
zaitcevBet I approved that one myself17:29
zaitcevhttps://review.openstack.org/#/c/184189/19/doc/source/deployment_guide.rst17:30
patchbotzaitcev: patch 184189 - swift - Allow 1+ object-servers-per-disk deployment (MERGED)17:30
*** jordanP has quit IRC17:31
tech004<zaitcev> Great thanks for the link I will look into it17:32
openstackgerritMerged openstack/swift: go: fix O_TMPFILE for existing files  https://review.openstack.org/29192817:32
zaitcevtech004: No, just follow the doc. However, I'm not certain you are following the right one for your problem.17:32
*** rickyrem has joined #openstack-swift17:33
openstackgerritMerged openstack/swift: go: fix O_TMPFILE compilation on older OS  https://review.openstack.org/29193417:33
tech004Thanks again guys17:34
*** panda has quit IRC17:40
*** panda has joined #openstack-swift17:41
*** david-lyle_ has joined #openstack-swift17:44
*** david-lyle has quit IRC17:45
*** david-lyle_ is now known as david-lyle17:46
*** rickyrem has quit IRC17:46
*** rickyrem has joined #openstack-swift17:55
zaitcevtech004: Wait, I'm spouting garbage whole day today. The bind_port still needs to be set to something, even if ignored in case of servers_per_port. Or so it appears by reading the code.17:55
*** haomaiwang has quit IRC18:01
*** haomaiwa_ has joined #openstack-swift18:01
*** geaaru has quit IRC18:08
*** Guest76883 has quit IRC18:15
claygzaitcev: for some reason that rings true to me - but of the life of me I can't recall the justification18:18
claygi have some vauge recallection of it having someething to do with replica or what_are_my_ips in general... on all-in-ones?18:19
zaitcevclayg: I think I'm going to change the check so that bind_port does not need to be set if servers_per_port is in force.18:19
claygi really don't remember18:19
claygzaitcev: worth a shot I suppose - you'll want to reconfigure your all-in-one with servers_per_port and make sure probetests still pass18:19
acoleseranrom, oshritf, notmyname: just an fyi, i plan to review container sync stuff early next week18:20
claygacoles: woot!  i'm finally on container sync today18:20
notmynamenice :-)18:20
acolesclayg: cool, well i can slipstream you :)18:20
notmynamesweet. acoles left +2 on patch 11771018:21
patchbotnotmyname: https://review.openstack.org/#/c/117710/ - swift - Add concurrent reads option to proxy18:21
claygacoles: i turned on slowdown and fired up concurrent gets i'm getting a bunch of functest failures - probably too many nodes are timing out writes or something18:21
claygacoles: so i'll probably dig into that, check for socket leaks maybe, pick up my unittest branch see if I can hit a few more test scenarios and call it a day18:21
acolesclayg: oh i noticed an exception when using slowdown with PUTs enabled18:22
claygacoles: ORLY!?18:22
acolesso i took PUT out of the slowdown req verbs in its json conf file18:22
*** mmcardle has quit IRC18:22
claygwas it an OverflowError?18:22
acolesclayg: IIRC the slowdown m/ware uses FileLikeIter or some class that doesn't implement a method to set 100 continue headers, tbh i forget exactly what18:23
acolesgimme a min...18:23
claygacoles: don't worry about it - i figure it out18:25
claygacoles: I like limiting the slowdown to GET HEAD anyway - since I don't expect any of the changes to effect any other verbs18:25
*** McMurlock1 has joined #openstack-swift18:25
claygacoles: it may be that we have some issues when backend nodes are slow to PUT's - but that's a seperate experiment18:26
gmmahaacoles: clayg: guess i will be right behind you :( sorry couldnt get to it in the last two days (Container sync reviews)18:26
*** McMurlock1 has quit IRC18:28
acolesclayg: network dropped out on me just then18:29
acolesi think it was call to set_hundred_continue_response_headers in obj server PUT that blows up with slowdown, but that should only be with EC policy18:29
claygacoles: ok, no worries18:30
clayggmmaha: ok, no worries18:30
*** klrmn has quit IRC18:31
acolesanyway, today i got myself set up to observe number of bytes being sync'd so I'm ready to try out the HEAD before PUTs sync patch next week.18:32
claygacoles: whoa - fancy - how'd you pull that off!?18:32
acolesclayg: nothing fancy, just grepping netstat and using a big object so the increments are obvious18:33
*** jistr has quit IRC18:33
acolesthen swift-init container-sync once, then once again18:33
acolestdasilva: you here?18:34
tdasilvaacoles: hi18:35
tdasilvaacoles: in a meeting, so only sort of here18:35
acolestdasilva: patch 263902 has yours and dfg's +2, and multiple +A but is blocked from merging by patch 260179. However, it does not need to depend on 260179, I can flip the dependency and then 263902 can merge. But I think if I do that (rebase 263902 on master) then I will lose the +2's.18:38
patchbotacoles: https://review.openstack.org/#/c/263902/ - swift - Re-format the SLO manifest file on new multipart-m...18:38
patchbotacoles: https://review.openstack.org/#/c/260179/ - swift - decouple versioned writes from COPY18:38
acolesso would tdasilva and dfg be ok to re-apply +2 on patch 263902?18:38
patchbotacoles: https://review.openstack.org/#/c/263902/ - swift - Re-format the SLO manifest file on new multipart-m...18:38
*** klrmn has joined #openstack-swift18:38
acolesif i were to rebase it onto master18:38
tdasilvaacoles: that would be ok by me18:39
acolesI think its worth it cos it could end up in merge conflict anyway the longer it sits there18:40
tdasilvaacoles: true, just simplifies things18:41
tdasilvaacoles: i noticed torgomatic also reviewed patch 260179 yesterday, so maybe that one is also closer to merging ???18:41
patchbottdasilva: https://review.openstack.org/#/c/260179/ - swift - decouple versioned writes from COPY18:41
tdasilvai think mattoliverau had a +2 on it18:41
*** mvk has quit IRC18:42
acolesyup but he won't be looking at it again in next week (I hope!)18:42
acolestdasilva: when i put those patches in dependency chain i had to choose which one to depend on the other, and I chose wrong, in terms of which got +A'd first18:45
*** ChubYann has joined #openstack-swift18:51
acolesi guess dfg isn't around? ^^18:53
*** dmorita_ has joined #openstack-swift18:58
*** dmorita has quit IRC19:01
*** haomaiwa_ has quit IRC19:01
*** haomaiwang has joined #openstack-swift19:01
*** daemontool_ has quit IRC19:03
*** dmorita_ has quit IRC19:04
openstackgerritAlistair Coles proposed openstack/swift: Refactor server side copy as middleware  https://review.openstack.org/15692319:08
openstackgerritAlistair Coles proposed openstack/swift: decouple versioned writes from COPY  https://review.openstack.org/26017919:08
openstackgerritAlistair Coles proposed openstack/swift: Re-format the SLO manifest file on new multipart-manifest GET call  https://review.openstack.org/26390219:08
*** mmcardle has joined #openstack-swift19:08
*** trifon has quit IRC19:10
acolestdasilva: so it rebased to master without losing +2's (I am surprised!). I +A'd19:11
acolesdfg: sorry for the noise, pls ignore19:11
*** chlong has quit IRC19:12
acolestdasilva: notmyname patch 263902 *should* now merge but I haven't seen it go to the queue in zuul yet. could you keep an eye on it? i have to leave. thx19:15
patchbotacoles: https://review.openstack.org/#/c/263902/ - swift - Re-format the SLO manifest file on new multipart-m...19:15
acolesok its in queue now19:18
*** acoles is now known as acoles_19:18
tdasilvaacoles_: thanks!19:21
sheelacoles_: hi there19:22
* sheel acoles_: need some input on https://bugs.launchpad.net/swift/+bug/155733719:22
openstackLaunchpad bug 1557337 in OpenStack Object Storage (swift) "Command "swift-init" is showing wrong output" [Undecided,Confirmed] - Assigned to Ankur Jain (j-ankur)19:22
sheelacoles_: is it worth fixed or not?19:23
sheelnotmyname: hey.. any input on ^^19:25
notmynameacoles_: willdo19:28
openstackgerritMichael Barton proposed openstack/swift: go: ObjectEngine abstraction layer  https://review.openstack.org/27893219:28
*** Jeffrey4l has quit IRC19:31
notmynamesheel: seems like there could be some improvements in cleaning up the output. so for that, sure, it could be worth fixing19:33
*** rickyrem has quit IRC19:33
sheelnotmyname: thanks for your input19:34
sheelnotmyname: so should it be cleaning of this command19:34
sheelnotmyname: or specifying some error message to user19:34
sheelnotmyname: means what way to fix this?19:34
notmynameeg the traceback output could be captured and turned into something cleaner and more helpful19:35
sheelnotmyname:  means is this command deprecated in kilo and liberty or still in use?19:35
sheelnotmyname: ah ok got it19:36
sheelnotmyname: thank you for direction19:36
notmynamenot it's not a deprecated command19:36
sheelnotmyname: ok, so basically we have to clean the message which is currently displayed19:38
sheelnotmyname:  like "processes may be running, but not able to show"19:38
sheel?19:38
notmynameI'd suggest starting with the traceback and treating that as one change by itself19:39
notmynameand not worry with the pid or pidfile not finding an existing running process19:39
sheelnotmyname: ok, got some clarity...19:41
sheelthanks19:43
claygrly?  test.functional.tests.TestSlo.test_slo_container_listing ... SKIPPED: Only passes with object_post_as_copy=False19:53
*** haomaiwang has quit IRC20:01
*** haomaiwa_ has joined #openstack-swift20:01
*** dmorita has joined #openstack-swift20:04
*** nadeem has joined #openstack-swift20:04
*** nadeem has quit IRC20:04
*** rickyrem has joined #openstack-swift20:04
*** nadeem has joined #openstack-swift20:05
*** dmorita has quit IRC20:11
*** mvk has joined #openstack-swift20:14
onovynotmyname: hi, are you here? still looking someone who can help me with fixing CVE-2016-0737 / CVE-2016-0738 in swift 2.2.020:16
notmynameonovy: I don't have any status updates for you on that20:17
onovyand are you still interested in fixing this?20:18
notmynameit's not a priority, but I'm not opposed to it. I've got it on my todo list, just not close to the top20:20
onovyok, thank you20:21
*** _JZ_ has quit IRC20:25
*** rickyrem has left #openstack-swift20:27
claygdid someone say they added something to do with COPY verbs to swiftclient?20:34
openstackgerritAndreas Jaeger proposed openstack/swift: Avoid translation problem  https://review.openstack.org/29479920:34
*** AJaeger has joined #openstack-swift20:34
timburkeclayg: https://review.openstack.org/#/c/280200/20:34
patchbottimburke: patch 280200 - python-swiftclient - Add copy object method20:34
AJaegerswift cores, https://review.openstack.org/294799 is a minimal fix to avoid a trap that translators run in. Would be nice to get this in...20:35
*** mmcardle has quit IRC20:36
claygtimburke: thanks!20:37
*** alejandrito has quit IRC20:37
*** dmorita has joined #openstack-swift20:43
*** vint_bra has joined #openstack-swift20:44
notmynameAJaeger: seems reasonable20:49
AJaegerthanks, notmyname20:57
*** MVenesio has quit IRC20:57
*** AJaeger has left #openstack-swift20:57
*** haomaiwa_ has quit IRC21:01
*** haomaiwang has joined #openstack-swift21:01
*** david-lyle has quit IRC21:06
*** david-lyle has joined #openstack-swift21:07
*** diogogmt has quit IRC21:21
*** diogogmt has joined #openstack-swift21:24
*** christian has joined #openstack-swift21:24
*** yarkot1 has joined #openstack-swift21:24
*** charz_ has joined #openstack-swift21:25
*** nchristia has quit IRC21:25
*** charz has quit IRC21:25
*** yarkot has quit IRC21:25
*** treyd has quit IRC21:25
*** hugokuo has quit IRC21:25
*** mathiasb has quit IRC21:25
*** sarafraj has quit IRC21:25
*** j_king has quit IRC21:25
*** esker has quit IRC21:26
*** dhellmann has quit IRC21:26
*** swat30 has quit IRC21:26
*** j_king has joined #openstack-swift21:26
*** dhellmann has joined #openstack-swift21:26
*** esker has joined #openstack-swift21:26
*** treyd has joined #openstack-swift21:26
*** hugokuo has joined #openstack-swift21:27
*** swat30 has joined #openstack-swift21:27
*** mathiasb has joined #openstack-swift21:34
openstackgerritMerged openstack/swift: Re-format the SLO manifest file on new multipart-manifest GET call  https://review.openstack.org/26390221:39
tdasilvayay! acoles_ ^^^21:40
*** panda has quit IRC21:40
*** panda has joined #openstack-swift21:41
*** CaioBrentano has quit IRC21:51
*** vint_bra has quit IRC21:56
*** juzuluag has quit IRC22:01
*** haomaiwang has quit IRC22:01
*** haomaiwa_ has joined #openstack-swift22:01
*** rickyrem1 has joined #openstack-swift22:02
*** ajiang has joined #openstack-swift22:09
*** rickyrem1 has quit IRC22:11
openstackgerritMichael Barton proposed openstack/swift: go: hashes.invalid  https://review.openstack.org/28683322:13
openstackgerritMerged openstack/swift: Avoid translation problem  https://review.openstack.org/29479922:16
*** rickyrem has joined #openstack-swift22:18
klamathnotmyname: quick question, for 6 replica, 2 regions, should write affinity node count be = 3 * replicas or 2 * replicas?22:18
notmynameI don't know22:20
notmyname(1) I 've never run a 6 replica cluster22:20
notmyname(2) personally, I'm not a big fan of write affinity22:20
claygoh reviewing concurrent gets is so much fun - i'm going to miss it!22:21
notmyname(3) both 3x and 2x will be expensive with 6 replicas22:21
klamathim seeing proxy writes in from r1 -> r2 with 3 * replica right now22:22
klamathgoing to try 2 * and see if i still see that22:22
*** rickyrem has quit IRC22:31
*** rickyrem has joined #openstack-swift22:32
*** rickyrem has quit IRC22:38
notmynamepatch 117710 is at the top of the merge queue in zuul! woohoo!22:39
patchbotnotmyname: https://review.openstack.org/#/c/117710/ - swift - Add concurrent reads option to proxy22:39
*** sheel has quit IRC22:47
*** haomaiwa_ has quit IRC23:01
*** haomaiwang has joined #openstack-swift23:01
openstackgerritMerged openstack/swift: Add concurrent reads option to proxy  https://review.openstack.org/11771023:11
clayg^ oh snap!23:11
*** nadeem has quit IRC23:13
*** garthb has quit IRC23:27
*** chsc has quit IRC23:36
*** itlinux has quit IRC23:40
*** itlinux has joined #openstack-swift23:43
*** itlinux has quit IRC23:47

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