Thursday, 2016-02-25

*** Jeffrey4l has joined #openstack-swift00:00
*** haomaiwang has quit IRC00:01
*** haomaiwang has joined #openstack-swift00:01
*** diogogmt has quit IRC00:10
openstackgerritBing Hu proposed openstack/python-swiftclient: Support uploading to an object in swift from stdin  https://review.openstack.org/25941000:16
*** mingdang1 has joined #openstack-swift00:31
*** takashi has joined #openstack-swift00:43
*** trifon has quit IRC00:48
*** delatte has joined #openstack-swift00:58
openstackgerritTim Burke proposed openstack/python-swiftclient: Doc follow-up to patch 282363  https://review.openstack.org/28449001:00
*** delattec has quit IRC01:00
patchbotopenstackgerrit: https://review.openstack.org/#/c/282363/ - python-swiftclient - Do not reveal auth token in swiftclient log messag...01:00
*** haomaiwang has quit IRC01:01
*** haomaiwang has joined #openstack-swift01:01
*** gyee has quit IRC01:04
*** diazjf has joined #openstack-swift01:04
*** diazjf1 has joined #openstack-swift01:06
*** diazjf has quit IRC01:06
*** mmuffinman has quit IRC01:07
*** mwheckmann has joined #openstack-swift01:09
*** mingdang1 has quit IRC01:13
*** garthb has quit IRC01:14
*** garthb_ has quit IRC01:14
openstackgerritTravis Tripp proposed openstack/swift: WIP Oslo.messaging middleware  https://review.openstack.org/24947101:19
*** haomaiwang has quit IRC01:22
*** sgundur has left #openstack-swift01:29
*** hezhiqiang has joined #openstack-swift01:46
openstackgerritMerged openstack/python-swiftclient: Do not reveal auth token in swiftclient log messages by default  https://review.openstack.org/28236301:49
*** hezhiqia_ has joined #openstack-swift01:50
openstackgerritOpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements  https://review.openstack.org/8925001:53
*** hezhiqiang has quit IRC01:53
*** dmorita has quit IRC01:56
*** dmorita has joined #openstack-swift01:58
*** haomaiwang has joined #openstack-swift01:59
*** haomaiwang has quit IRC02:01
*** haomaiwang has joined #openstack-swift02:01
*** diogogmt has joined #openstack-swift02:05
*** mingdang1 has joined #openstack-swift02:05
*** lyrrad has quit IRC02:05
*** mingdang1 has quit IRC02:05
*** mingdang1 has joined #openstack-swift02:06
openstackgerritDaisuke Morita proposed openstack/swift: Improvement of logging for invalid policies  https://review.openstack.org/14123802:11
*** klrmn has quit IRC02:18
*** natarej has joined #openstack-swift02:32
*** rminmin has joined #openstack-swift02:33
*** links has joined #openstack-swift02:47
openstackgerritDaisuke Morita proposed openstack/swift: Show each policy's information on audited results in recon  https://review.openstack.org/13869702:57
*** diazjf1 has quit IRC02:58
*** haomaiwang has quit IRC03:01
*** haomaiwang has joined #openstack-swift03:01
*** dmorita has quit IRC03:02
*** dmorita has joined #openstack-swift03:09
*** dmorita has quit IRC03:14
*** diogogmt has quit IRC03:23
*** diogogmt has joined #openstack-swift03:26
*** links has quit IRC03:34
*** hezhiqia_ has quit IRC03:34
*** hezhiqiang has joined #openstack-swift03:35
*** hezhiqiang has quit IRC03:41
*** hezhiqiang has joined #openstack-swift03:42
*** diogogmt has quit IRC03:43
*** ho_ has joined #openstack-swift03:45
*** chlong_ has quit IRC03:47
*** sgundur has joined #openstack-swift03:48
*** sgundur has left #openstack-swift03:48
*** chlong_ has joined #openstack-swift03:49
*** mingyu has quit IRC03:53
*** mingyu has joined #openstack-swift03:53
*** haomaiwang has quit IRC04:01
*** haomaiwa_ has joined #openstack-swift04:01
*** mingyu has quit IRC04:02
*** mingyu has joined #openstack-swift04:04
*** hezhiqiang has quit IRC04:11
*** hezhiqiang has joined #openstack-swift04:11
*** links has joined #openstack-swift04:29
notmynamegood evening04:29
kota_notmyname: everning04:29
ho_notmyname: good evening!04:30
*** haomaiwa_ has quit IRC04:30
*** nadeem has joined #openstack-swift04:31
notmynamewhat's going on?04:33
*** mwheckmann has quit IRC04:37
*** haomaiwa_ has joined #openstack-swift04:39
kota_starting to look at auditor fix04:42
mahaticgood morning04:44
notmynamekota_: thanks04:45
*** ho_away has joined #openstack-swift04:45
notmynamemahatic: good morning04:45
notmynamemahatic: when do you get to bristol?04:45
kota_mahatic: morning04:46
*** ho_away has quit IRC04:46
*** ppai has joined #openstack-swift04:47
*** mingyu has quit IRC04:49
mahaticnotmyname: hello, on sunday evening-ish. Will be arriving in London saturady noon though, hanging out in London until sunday!04:50
notmynamecool04:50
mahatictdasilva: i'm off to stamford bridge on Sunday morning ;) my brother's gf will accompany me04:51
mahaticnotmyname: and you?04:51
*** mingyu has joined #openstack-swift04:51
notmynameI'm getting to london on saturda, but I'm going straight to bristol04:51
*** dmorita has joined #openstack-swift04:51
mahaticah I see04:51
*** mingyu has quit IRC04:52
mahaticI'm planning on taking a taxi back from Bristol to heathrow on Friday morning (8 am i think). If someone needs a ride at the same time, they could join in04:54
notmynameah, cool. yeah I'll be making the same trip. I don't have anything set up yet. was planning on taking the train04:54
notmynameI may take you up on that ;-)04:54
mahaticsure04:55
*** dmorita has quit IRC04:55
mahatictrain needs us getting off somewhere and taking another one, bleh04:56
kota_I am planning to take National Express coach between Bristol and Heathrow.04:58
kota_direct path for Bristol, the coach station is close to our hotel (15 minutes in walking)04:59
*** hezhiqiang has quit IRC05:00
*** hezhiqiang has joined #openstack-swift05:00
notmynameyeah, I need to do some research on LHR->bristol05:00
notmyname(and back)05:00
*** haomaiwa_ has quit IRC05:01
*** haomaiwang has joined #openstack-swift05:01
mattoliveraunotmyname: hire a car, that's what our peeps did for the nova midcycle.05:02
mattoliveraunotmyname: tho the train system is pretty good in the UK05:02
notmynameyeah, I think train is what we're doing05:02
mattoliverauIf i was going I'd hire a car and drive, cause its the same highway I used to drive everyday (except a little further) so would feel like old times  :)05:04
mattoliverauAlso: curl -4 http://wttr\.in/melbourne05:04
mattoliverauie run that in a terminal :)05:04
notmynamenice05:05
*** Gordoni has quit IRC05:06
*** nadeem has quit IRC05:08
zaitcevTrain takes too long05:12
*** diazjf has joined #openstack-swift05:12
zaitcevHas to go through London.05:12
*** mingyu has joined #openstack-swift05:16
*** mingyu has quit IRC05:16
openstackgerritHisashi Osanai proposed openstack/swift: WIP: Replace the http library in SimpleClient for Ccontainer-Sync  https://review.openstack.org/27941705:21
*** trifon has joined #openstack-swift05:22
notmynamezaitcev: looks like a little more than 2 hours05:26
zaitcevNo way.05:27
zaitcevWell, okay. Google says so.05:28
notmynamethat's all I have to go on right now :-)05:28
*** mingyu has joined #openstack-swift05:30
*** mingyu has quit IRC05:31
*** mingyu has joined #openstack-swift05:31
*** haomaiwang has quit IRC06:01
*** haomaiwang has joined #openstack-swift06:01
mahatickota_: right, i just checked, there are coaches too (bristol - LHR), right from early in the morning. thanks for that info06:11
kota_mahatic: you're welcome :-)06:12
*** mmuffinman has joined #openstack-swift06:13
*** mrmoje has joined #openstack-swift06:16
mrmojegood morning06:27
*** ChubYann has quit IRC06:29
*** kei_yama has quit IRC06:34
notmynamemrmoje: good morning06:35
mattoliveraumrmoje: morning06:35
*** SkyRocknRoll has joined #openstack-swift06:36
*** rminmin has quit IRC06:39
*** rminmin has joined #openstack-swift06:39
*** dmorita has joined #openstack-swift06:39
*** mrmoje has quit IRC06:41
*** kei_yama has joined #openstack-swift06:42
*** dmorita has quit IRC06:43
*** mrmoje has joined #openstack-swift06:43
*** ygl has joined #openstack-swift06:46
yglHi All06:46
ygli am having problems in uploading images of sizes 3GB in to siwft06:47
yglcan anyone help please ?06:47
yglanyone here06:50
kota_ygl: what's your problem?06:55
yglkota_: i am unable to upload imges of size 3 GB into swift06:56
kota_notmyname: review done for patch 27944006:56
patchbotkota_: https://review.openstack.org/#/c/279440/ - swift - Skip already checked partitions when auditing obje...06:56
kota_ygl: what response code did you get?06:56
yglkota_: failure and no ability to reset contents for reupload.06:57
yglkota_: but i can upload cirros images though06:58
kota_ygl: what client do you use? swift commands? or curl or another?07:00
yglkota_: swift --debug   upload demo-container2  vsbc-V05.00.01A604.qcow207:00
kota_if you could tell me the *HTTP* status code from swift, it would be helpful07:00
kota_alright, swift command in swift-client07:00
*** haomaiwang has quit IRC07:01
yglkota_: INFO:swiftclient:RESP STATUS: 503 Service Unavailable07:01
*** haomaiwang has joined #openstack-swift07:01
kota_ok, 503 means something wrong happen in Swift.07:01
yglkota_: how to check it ?07:02
zaitcevTime to look at proxy log.07:02
yglkota_: but its working well for smaller images07:02
kota_zaitcev: exactly!07:02
zaitcevProbably out of space, although that one is a 509 or 507.07:03
yglzaitcev: has enough space07:03
yglkota_: how to proceed now ?07:03
kota_ygl: coud you see your proxy log?07:04
yglkota_: should I take a running log while uploadin ?07:05
zaitcevIt's the log that your server writes. It's already at the proxy server somewhere.07:06
kota_ygl: if you already have the log when you got the problem, it's enough to look at.07:07
yglzaitcev: mine is configured in syslog07:07
yglkota_: Feb 25 06:42:46 BASE-CCP-T1-M1-NETCLM proxy-server: ERROR with Object server 10.177.0.13:6000/d1 re: Trying to write to /v1/AUTH_c9554ffbccdd4c16947aa14f964986a0/demo-container2/vsbc-V05.00.01A604.qcow2: ChunkWriteTimeout (10s)07:08
zaitcevAh, so it's overloaded to death then.07:09
zaitcevShorter files fit into RAM, but as soon it spills on disk, it keels over.07:10
kota_ygl: thanks, looks like the object server stalled while transferring the data stream.07:10
*** nakagawamsa has joined #openstack-swift07:12
yglkota_: so what to do now07:12
*** diazjf has quit IRC07:12
yglkota_: i will be back shortly07:12
yglkota_: i will ping u again07:12
*** mrmoje has quit IRC07:13
*** dmorita has joined #openstack-swift07:14
*** mrmoje has joined #openstack-swift07:15
yglkota_: i am back07:17
yglkota_: so how to proceed ?07:17
*** dmorita has quit IRC07:18
yglkota_: any idea ?07:20
kota_how about the object server log?07:21
kota_I am realizing zaitcev already has the answer i.e. running out of disk spaces.07:21
yglkota_: but i have enough disk space07:21
yglkota_: where can I find the object server log ?07:22
zaitcevkota_: No, ygl says it's not that and in any case it's a timeout. It's too slow. One may confirm by looking at disk service times with iostat.07:22
*** mrmoje has quit IRC07:23
kota_zaitcev: yup, I'm not sure for swift client for now, it could send transfer-encoding chunked?07:23
kota_ygl: maybe you can see that in syslog if you configure same setting with proxy sever.07:24
yglkota_: how to configure and where ?07:25
kota_not sure for now, we got disk full with transfer-encoding chunked.07:25
yglkota_: u mean the hard disk space ?07:26
zaitcevkota_: I don't think it's the chunk it means. It's just with ChunkWriteTimeout(self.app.client_timeout): yield buf07:26
zaitcevnot encoding related07:26
yglkota_: zaitcev :  root@BASE-CCP-T1-M1-NETCLM:~# df -h Filesystem      Size  Used Avail Use% Mounted on /dev/dm-0       3.6T  106G  3.3T   4% /07:26
kota_ygl: object-sever.conf07:27
kota_zaitcev: alright. thanks.07:27
zaitcevTimeout is settable in proxy-server.conf07:27
yglzaitcev: kota_ : so what should I do now ?07:28
zaitcevThe problem is, if you start raising it, clients will time out whenever there's a handoff.07:28
*** mingyu has quit IRC07:28
yglzaitcev: kota_ : http://pastebin.com/mv2vbPJk07:28
yglzaitcev: kota_ : thats my object-server.conf07:28
zaitcevThat's why it's so ridiculously short by default (10s)07:28
kota_ygl: you looks to use default setting for object-server log so check your07:30
-kota_- # You can specify default log routing here if you want:07:30
-kota_- # log_name = swift07:30
-kota_- # log_facility = LOG_LOCAL007:30
-kota_- # log_level = INFO07:30
-kota_- # log_address = /dev/log07:30
-kota_- # The following caps the length of log lines to the value given; no limit if07:30
-kota_- # set to 0, the default.07:30
-kota_- # log_max_line_length = 007:30
zaitcevBah, usually default is fine07:31
zaitcevEven if it ends in /var/log/messages, still readable07:31
kota_ygl: so your object-server probably dump the log into syslog.07:31
yglzaitcev: kota_ so what should  I do now ?07:32
zaitcevygl: Anyway, your overall task is to find why writing a 1MB chunk takes more than 10 seconds, then correnct the problem.07:32
kota_zaitcev: right07:32
openstackgerritCharles Hsu proposed openstack/swift: Fix account-reaper unable to delete all containers.  https://review.openstack.org/28456207:32
yglzaitcev: i am suspecting the hard disk. becos i have another env with the same configuration. its working fine with this configuration07:33
yglzaitcev: may be i can try using the dd command and write it to hard disk07:34
zaitcevygl: not a bad idea. also look into iostat(1), it's a part of sysstat package.07:36
*** nakagawamsa has quit IRC07:36
zaitcevanyway, good night07:37
yglzaitcev: how to measure the time of dd command07:37
zaitcev"time dd" of course, but GNU dd reports the speed anyway07:38
yglzaitcev: ok thanks07:38
*** zaitcev has quit IRC07:38
*** zul has quit IRC07:41
*** zul has joined #openstack-swift07:47
yglkota_: is there anything else I can check ?07:50
*** rminmin has quit IRC07:53
kota_ygl: sorry I'm running out of time to spend here for now but the dd result and object-server log would be helpful to look at.07:54
yglkota_: i am able to create a 1 MB file within a second07:55
*** haomaiwang has quit IRC08:01
*** jmccarthy1 has joined #openstack-swift08:01
*** haomaiwa_ has joined #openstack-swift08:01
yglhi all08:02
yglcan anyone here help me with my swift issue ?08:02
*** rledisez has joined #openstack-swift08:05
*** ygl has quit IRC08:08
*** jmccarthy1 has quit IRC08:08
ho_ygl: what is your problem now?08:12
*** Oku_OS has quit IRC08:18
*** jmccarthy1 has joined #openstack-swift08:21
*** _JZ_ has joined #openstack-swift08:30
*** geaaru has joined #openstack-swift08:32
*** acoles_ is now known as acoles08:33
*** mingyu has joined #openstack-swift08:35
*** SkyRocknRoll_ has joined #openstack-swift08:39
*** geaaru has quit IRC08:40
*** SkyRocknRoll has quit IRC08:41
*** ygl has joined #openstack-swift08:43
yglHi All08:43
yglcan anyone help me with the swift issue ?08:43
yglkota_: how to increase the timeout value of 10 seconds in the object server ?08:44
*** ygl has quit IRC08:51
*** geaaru has joined #openstack-swift08:54
openstackgerritAlistair Coles proposed openstack/swift: Stop staticweb revealing container existence to unauth'd requests  https://review.openstack.org/24886709:00
*** haomaiwa_ has quit IRC09:01
*** asettle has joined #openstack-swift09:01
*** haomaiwang has joined #openstack-swift09:01
*** jistr has joined #openstack-swift09:01
*** mrmoje has joined #openstack-swift09:02
*** ygl has joined #openstack-swift09:07
yglHi All09:07
yglis anyone available to help me with the swift issue please ?09:07
*** mvk has joined #openstack-swift09:09
*** Oku_OS has joined #openstack-swift09:10
onovyygl: hi, what's the problem?09:21
yglonovy: i am experiencing timeouts in the proxy log, when uploading 1 GB images09:22
yglonovy: i have found that there is high  write waiting times in iostat for that disk09:23
onovycopy exact error please09:23
yglonovy: ok09:23
onovywhat swift version do you have?09:23
yglonovy: how to check that ?09:24
yglonovy: Feb 25 08:04:33 BASE-CCP-T1-M1-NETCLM proxy-server: ERROR with Object server 10.177.0.12:6000/d1 re: Trying to get final status of PUT to /v1/AUTH_c9554ffbccdd4c16947aa14f964986a0/demo.container6/file.txt: Timeout (10s)09:25
mrmojeswift --version09:25
yglonovy: i have check writing a 1 GB file using dd command. it worked but  not working through swift09:25
mrmojeo sorry thats for client09:25
onovyygl: how you installed swift? rpm/deb/by hand?09:26
yglonovy: swift 2.4.0.post2909:26
onovyah, 2.4.0, ok09:26
yglonovy: no through automation09:26
onovyhttps://github.com/openstack/swift/blob/stable/liberty/etc/proxy-server.conf-sample#L9709:26
yglonovy: i want to know something. why the dd command is working but not swift ?09:26
onovythis is option you are searching for09:27
yglonovy: i have another identical environment. there swift is working fine09:27
onovymaybe bad disk?09:27
onovydd is just local, but swift is over network, so you could be network problem too09:28
onovyso it09:28
yglonovy: yes but why dd command is able to copy the 1 GB file  but not swift ?09:28
yglonovy: oh ok09:28
onovyswift is much more complex09:28
onovyit need to write to container-server info about new object09:29
onovyso maybe problem is with container-server09:29
yglonovy: how does swift write to hard disk ? through drivers ?09:29
onovylook into object-server log09:29
onovy"drivers"? :)09:29
yglonovy: i mean  how does  it send commands to write to hard disk ? through kernel ?09:30
onovyos.write09:30
onovyeverything you write to disk must be done through kernel09:30
yglonovy: oh ok through python os command09:31
onovyos.write = python function, which calls glibc, which calls kernel09:31
yglonovy: oh ok09:31
onovyisn't container "demo.container6" really big? >100k objects?09:32
onovyif you have container server on rotational disk, containers should be small ~ <1M09:32
onovyif it's big, latency goes up09:32
onovyno problem at all on SSD09:32
yglonovy: i see large objects doing well in another env I told u abt09:33
onovyyes i know09:34
yglonovy:probably its the disks I think are bad09:34
onovyso what is difference? :)09:34
onovytry to run smartctl and look for errors and reallocated sectors09:34
yglonovy: also while uploading, i see iostats displays high write wait timings above 1000 + millisecs09:34
yglonovy: ok09:34
onovydisk latency should go up when you writing to it (fast)09:35
onovyso this could be "normal"09:35
yglonovy: so what to check out in the iostat output ?09:35
onovyi'm looking to time and iops09:36
*** hezhiqia_ has joined #openstack-swift09:36
yglonovy: which time ?09:36
onovyrequest time, same as you wrote09:36
onovyi just want to point that, that >1s disk latency can be "ok"09:36
yglonovy: oh ok09:37
yglonovy: how to increase that timeout value in swift as shown in the error ?09:37
onovy10:26:48           onovy | https://github.com/openstack/swift/blob/stable/liberty/etc/proxy-server.conf-sample#L9709:37
onovyand btw: can you upgrade to 2.5.0, which is liberty release? you are using "one version before final liberty release"09:38
yglonovy: what value should be changed ?09:38
onovyygl: i sent you link to exact line09:38
yglonovy: i cant find it ?09:38
onovyi have yellow background on github09:38
onovyit09:38
onovy# node_timeout = 1009:38
*** hezhiqiang has quit IRC09:39
onovysorry must go to meeting now. i will try to reply later09:39
yglonovy: i could n't find the value09:39
yglonovy: oh ok09:39
yglonovy: i have increased node_timeout and cheked09:40
yglonovy: it did not work09:40
*** mingyu has quit IRC09:42
yglany one here who can help ?09:42
*** haomaiw__ has joined #openstack-swift09:48
*** haomaiwang has quit IRC09:49
*** hezhiqia_ has quit IRC09:56
*** haomaiw__ has quit IRC10:01
*** haomaiwang has joined #openstack-swift10:01
*** mvk has quit IRC10:01
*** ho_ has quit IRC10:03
*** stantonnet has quit IRC10:03
*** stantonnet has joined #openstack-swift10:04
*** daemontool has joined #openstack-swift10:07
*** admin6 has joined #openstack-swift10:13
*** ygl has quit IRC10:20
*** daemontool has quit IRC10:21
*** daemontool has joined #openstack-swift10:24
*** esikachev has joined #openstack-swift10:33
esikachevhey folks!10:34
esikachevwho can help me with python-swiftclient?10:34
*** pcaruana has joined #openstack-swift10:42
*** kei_yama has quit IRC10:56
*** haomaiwang has quit IRC11:01
*** haomaiwang has joined #openstack-swift11:01
*** mmuffinman has quit IRC11:16
*** joeljwright has joined #openstack-swift11:16
*** ChanServ sets mode: +v joeljwright11:16
onovyesikachev: hi, what's the problem?11:21
esikachevI'm looking for an easy way to remove pseudo-folder with files. is this possible? i am tried remove all files from pseudo-folder and than remove pseudo-folder11:23
esikachevor force removing container with data11:24
acolesnotmyname: timburke joeljwright - liberty backport for bug 1516692 - patch 28464511:24
openstackbug 1516692 in python-swiftclient "swiftclient logs sensitive info in http requests" [Critical,Fix released] https://launchpad.net/bugs/1516692 - Assigned to Joel Wright (joel-wright-k)11:24
patchbotacoles: https://review.openstack.org/#/c/284645/ - python-swiftclient (stable/liberty) - Do not reveal auth token in swiftclient log messag...11:24
*** jordanP has joined #openstack-swift11:25
*** mingdang1 has quit IRC11:25
*** mingdang1 has joined #openstack-swift11:25
*** mingdang1 has quit IRC11:30
*** hezhiqiang has joined #openstack-swift11:40
*** mingyu has joined #openstack-swift11:43
*** asettle has quit IRC11:46
*** mingyu has quit IRC11:47
*** haomaiwang has quit IRC12:01
*** SkyRocknRoll_ has quit IRC12:01
*** haomaiwang has joined #openstack-swift12:01
*** fthiagogv has quit IRC12:10
*** mmuffinman has joined #openstack-swift12:11
*** mingdang1 has joined #openstack-swift12:19
*** haomaiwang has quit IRC12:31
*** NM has joined #openstack-swift12:31
*** daemontool_ has joined #openstack-swift12:43
*** mingyu has joined #openstack-swift12:43
*** zul has quit IRC12:43
*** zul has joined #openstack-swift12:44
*** daemontool has quit IRC12:46
*** diogogmt has joined #openstack-swift12:46
*** daemontool_ has quit IRC12:47
*** mingyu has quit IRC12:47
*** fthiagogv has joined #openstack-swift12:53
*** pcaruana has quit IRC12:55
*** km has quit IRC12:55
*** joeljwright has quit IRC12:57
*** daemontool has joined #openstack-swift13:01
*** loth has joined #openstack-swift13:03
*** loth1 has quit IRC13:03
*** ppai has quit IRC13:04
*** pcaruana has joined #openstack-swift13:09
*** diogogmt has quit IRC13:10
*** ppai has joined #openstack-swift13:15
*** joeljwright has joined #openstack-swift13:16
*** ChanServ sets mode: +v joeljwright13:16
*** joeljwright has quit IRC13:18
*** asettle has joined #openstack-swift13:20
*** asettle has quit IRC13:20
*** asettle has joined #openstack-swift13:21
asettlenotmyname: thank you for your help :) patch looks good.13:26
*** links has quit IRC13:26
*** mmuffinman has quit IRC13:30
*** ppai has quit IRC13:32
tdasilvagood morning13:48
admin6hello there. I’m currently trying to migrate authentication from tempauth to keystone api v3. It works but where I had answers under 50ms with tempauth to get a token then get an object. I know have duration around 300ms. Plus randomly, some requests are stuck for more than 1 second.13:55
*** mwheckmann has joined #openstack-swift13:56
*** daemontool_ has joined #openstack-swift13:59
*** ho_away has joined #openstack-swift14:01
*** daemontool has quit IRC14:02
*** links has joined #openstack-swift14:07
*** petertr7_away is now known as petertr714:15
*** Jeffrey4l has quit IRC14:33
*** mingdang1 has quit IRC14:34
*** joeljwright has joined #openstack-swift14:39
*** ChanServ sets mode: +v joeljwright14:39
*** hezhiqiang has quit IRC14:41
*** petertr7 is now known as petertr7_away14:45
*** Jeffrey4l has joined #openstack-swift14:47
*** mingyu has joined #openstack-swift14:54
*** diogogmt has joined #openstack-swift14:59
*** petertr7_away is now known as petertr715:01
*** sgundur has joined #openstack-swift15:01
*** PsionTheory has joined #openstack-swift15:04
*** davideagle_ has quit IRC15:06
*** daemontool__ has joined #openstack-swift15:10
*** links has quit IRC15:12
*** davideagle_ has joined #openstack-swift15:13
*** daemontool_ has quit IRC15:14
*** garthb_ has joined #openstack-swift15:18
*** garthb has joined #openstack-swift15:18
*** trifon has quit IRC15:20
*** ho_away has quit IRC15:21
*** takashi_ has joined #openstack-swift15:24
*** PsionTheory has quit IRC15:26
*** takashi has quit IRC15:26
*** davideagle_ has quit IRC15:27
*** mingdang1 has joined #openstack-swift15:30
*** Jeffrey4l has quit IRC15:31
*** davideagle_ has joined #openstack-swift15:33
*** ajiang has left #openstack-swift15:35
*** mingdang1 has quit IRC15:35
*** petertr7 is now known as petertr7_away15:39
*** bjkeller has joined #openstack-swift15:39
*** jmccarthy has quit IRC15:51
*** jmccarthy has joined #openstack-swift15:51
asettleacoles: you aroundddddd?15:53
acolesasettle: hi15:53
asettleacoles: do you have a preference between Hewlett Packard Enterprise and HPE? Do you guys have a standard?\15:57
*** nadeem has joined #openstack-swift15:58
*** nadeem has quit IRC15:58
*** nadeem has joined #openstack-swift15:59
*** chsc has joined #openstack-swift16:03
*** garthb has quit IRC16:06
*** garthb_ has quit IRC16:06
notmynamegood morning16:06
asettleO/16:07
acolesasettle: thats a good question16:08
*** diogogmt has quit IRC16:08
acolesasettle: I think HPE is ok, and briefer!16:09
asettleYess, it probably should reflect what you guys want and use as a standard (since we're very clearly stealing the old HP docs)16:09
asettleOkay, as long as HPE is okay with that, I can go through and standardise that.16:09
notmynamehttps://static.googleusercontent.com/media/research.google.com/en//pubs/archive/44830.pdf is google's recent paper on the state of HDD tech for large storage16:09
asettlenotmyname: thanks for your patch overnight, I appreciate the help :)16:09
notmynameasettle: happy to help. anything that gets it landing faster :-)16:10
*** esikachev has quit IRC16:10
notmynameasettle: I was still seeing some sphinx formatting issues. with tables, mostly, IIRC16:10
asettlenotmyname: should be pretty much on its way now, other than the HPE thing. Got to go through and standardise that. I was having troubles with my build system (new laptop woes).16:11
asettleThe tables were a nightmare, tbh. I had to do most of them manually.16:11
notmynameI changed several HP to HPE references. actually, I thought I got them all. note I didn't change hardware model numbers16:12
asettlenotmyname: oh good :) there's Hewlett Packard Enterprise referenced in the beginning in the index file, but I can quickly fix that (or at least, say Hewlett Packard Enterpise and ensure (HPE) is there too for referential sake)16:12
*** hezhiqiang has joined #openstack-swift16:13
*** jmccarthy1 has quit IRC16:13
asettlenotmyname: did you get any build issues with the links?16:15
*** daemontool__ has quit IRC16:16
*** trifon has joined #openstack-swift16:17
*** etienneme has quit IRC16:20
*** haomaiwa_ has joined #openstack-swift16:24
*** urth has quit IRC16:25
*** etienneme has joined #openstack-swift16:26
*** lpabon has joined #openstack-swift16:28
*** urth has joined #openstack-swift16:29
notmynameasettle: yeah, I think there might have been one or two. I don't remember specifically16:30
acolesasettle: having Hewlett Packard Enterpise (HPE) and then HPE onwards sounds good16:30
asettleacoles: rad. notmyname gracias16:30
asettleWill do16:30
*** arch-nemesis has joined #openstack-swift16:30
acolesasettle: when the companies split I thought HPE was verboten but I see it all over our public websites16:31
asettleCool, well, I mean, they'd know, right? ;)16:31
*** davideagle_ has quit IRC16:32
*** mrmoje has quit IRC16:40
*** gyee has joined #openstack-swift16:41
*** acoles is now known as acoles_16:47
*** etienneme has quit IRC16:48
*** etienneme has joined #openstack-swift16:48
*** diazjf has joined #openstack-swift16:51
notmynamejoeljwright: timburke: we've got the auth token hiding thing landed in swiftclient. that was the only thing we were waiting on for a swiftclient release, right?16:53
*** Razva has joined #openstack-swift16:54
openstackgerritBing Hu proposed openstack/python-swiftclient: download method shouldn't download all object  https://review.openstack.org/28480516:54
notmynameacoles_: thank you for doing the swiftclient backport16:56
*** garthb has joined #openstack-swift16:58
*** garthb_ has joined #openstack-swift16:59
joeljwrightnotmyname: I think so16:59
joeljwrightnotmyname: it needs a very small patch for the test, but that shouldn;t hold it up17:00
notmynameI'll pick up the final release note patches work today17:00
joeljwrightnotmyname: just about tot push a new version of the backport to fix unicode issues17:00
notmynamegreat17:00
joeljwrightI'll have the test update patch up within the hour17:00
joeljwrightfor master17:00
*** haomaiwa_ has quit IRC17:01
*** petertr7_away is now known as petertr717:01
*** esker has quit IRC17:01
*** haomaiwang has joined #openstack-swift17:01
openstackgerritCharles Hsu proposed openstack/swift: Fix account-reaper unable to delete all containers.  https://review.openstack.org/28456217:05
joeljwrightnotmyname: any advice on updating another person's patch?17:06
joeljwrightI have checked out 284645, but after amending it tells me I'm about to submit multiple commits...17:07
joeljwrightnot seen this before17:07
notmynamejoeljwright: there's nothing special. use the same change-id, like you would for updating your own. and leave a comment about what you changed, like you would for your own17:07
joeljwrightah! I understand17:08
joeljwrightthere's a dependent patch17:08
notmynameoh yeah. that would do it17:08
*** jistr has quit IRC17:08
notmynameeven though it warns you of that, if you aren't changing the first one in the chain, it will not update it17:08
joeljwrightkk, thanks17:09
openstackgerritCharles Hsu proposed openstack/swift: Fix account-reaper unable to delete all containers.  https://review.openstack.org/28456217:13
*** geaaru has quit IRC17:15
*** lyrrad has joined #openstack-swift17:15
openstackgerritAlexandra Settle proposed openstack/swift: operational procedures guide  https://review.openstack.org/27824517:15
openstackgerritJoel Wright proposed openstack/python-swiftclient: Fix test for redacting sensitive data in client.http_log()  https://review.openstack.org/28482217:15
asettlenotmyname: I think I've managed my last round of edits for now. I keep finding spelling mistakes and other things throughout the doc though. So it's going to need another read over at some point.17:16
notmynameasettle: ok17:16
asettlenotmyname: was this originally an internal doc?17:17
notmynameasettle: what's next? do you have more work you want to do on it now? or is the next step to review and land what's there now?17:17
notmynameasettle: yes17:17
asettlenotmyname: review and land for now, I think :) I believe it's okay to go up, unless there's still significant build issues?17:17
notmynameok17:17
notmynamethat sounds like a great goal for next week :-)17:18
asettleIt's in relatively good-ish shape considering the horrific conversion (let's never do word doc again, promise? :P)17:18
notmynameheh, ok :-)17:18
notmynamethank you for doing it. wouldn't have happened otherwise17:19
asettleNot a problem at all :) it's been an interesting learning experience ;)17:19
Zyric_Good Morning17:23
notmynamehello Zyric_17:23
openstackgerritJohn Dickinson proposed openstack/python-swiftclient: authors/changelog updates for 2.8 release  https://review.openstack.org/28197417:24
notmynamejoeljwright: timburke: there's the authors/changelog updates for a 2.8 swiftclient release17:24
notmynamejoeljwright: I do want to see your patch too about the unicode stuff17:25
notmynamebut my patch is there to look at for accuracy17:25
*** zhill has joined #openstack-swift17:26
*** barker has joined #openstack-swift17:27
notmynameoops17:28
notmynameI opened up a patch to review and discovered that I had a few pending inline comments on it that I never submitted17:28
*** admin6 has quit IRC17:30
*** barker has quit IRC17:32
*** jordanP has quit IRC17:35
*** davideagle has joined #openstack-swift17:36
*** lyrrad has quit IRC17:36
timburkenotmyname: are you going to address joeljwright's and my comments on the CHANGELOG?17:42
notmynameyeah I should probably do that17:43
notmyname...yeah...I was totally going to do that...absolutely didn't just forget that I had WIP submitted it before...I'm all over it17:44
notmynamehas anyone here used wercker (http://wercker.com)? it looks to be a container deployment tool17:45
joeljwrightnotmyname: sorry I was picking up my wife from the station17:45
joeljwrightnotmyname: will check over the the changelog17:45
notmynamejoeljwright: looks like I need to address your existing comments first ;-)17:46
joeljwright:)17:46
*** asettle has quit IRC17:47
*** petertr7 is now known as petertr7_away17:47
timburkenotmyname: how much of a time crunch are we under to get the release out? acoles_ made a good point about token vs. header in patch 284490, which makes me want to change the boolean option name from redact_sensitive_tokens to redact_sensitive_headers. but if we're going to have a release before my follow-up lands, it's probably not worth it (as we'd want to preserve compatibility with the old option)17:52
patchbottimburke: https://review.openstack.org/#/c/284490/ - python-swiftclient - Doc follow-up to patch 28236317:52
notmynametimburke: we should have a release next week17:54
*** hezhiqiang has quit IRC17:56
openstackgerritTim Burke proposed openstack/python-swiftclient: Follow-up to patch 282363  https://review.openstack.org/28449017:57
patchbotopenstackgerrit: https://review.openstack.org/#/c/282363/ - python-swiftclient - Do not reveal auth token in swiftclient log messag... (MERGED)17:57
joeljwrightwoohoo! thanks patchbot!17:57
timburkenotmyname: if you want patchbot to stop doing that^, push out the latest code17:58
notmynameI've been worried about getting into some botstorm17:58
notmynameoh? did you fix something?17:58
notmynameah yes17:58
joeljwrightfwiw I agree with timburke and acoles_ - the mixed use of token and header is confusing17:59
timburkego +2 it!17:59
*** rledisez has quit IRC18:00
*** haomaiwang has quit IRC18:01
*** haomaiwang has joined #openstack-swift18:01
joeljwrighttimburke: sorry, had to -118:02
Razvais Liberty ready for production, or should I go with Kilo?18:02
*** dmorita has joined #openstack-swift18:02
joeljwrighttimburke: but if you fix shell.py I'll give it +2 :)18:03
notmynameRazva: for swift?18:03
timburkejoeljwright: gah! good catch. thanks18:03
notmynameRazva: I'd recommend using 2.6.0. (all swift releases are production ready)18:04
notmynameRazva: but certainly liberty (which was swift 2.5.0) is production ready18:04
timburkealthough there is a known issue with 2.6.0 if you have separate middleware that use Swift's logging infrastructure18:05
openstackgerritTim Burke proposed openstack/python-swiftclient: Follow-up to patch 282363  https://review.openstack.org/28449018:08
notmynamenotice how patchbot didn't respond to openstackgerrit this time18:08
notmynamethanks timburke18:08
*** mingyu has quit IRC18:09
timburkeyay!18:09
joeljwright:)18:09
*** mingyu has joined #openstack-swift18:09
openstackgerritBing Hu proposed openstack/python-swiftclient: make download --prefix working  https://review.openstack.org/28485118:10
*** gyee has quit IRC18:11
*** mingyu has quit IRC18:16
*** mingyu has joined #openstack-swift18:16
*** barker has joined #openstack-swift18:19
notmynamejoeljwright: still here?18:19
*** klrmn has joined #openstack-swift18:21
notmynamejoeljwright: why shoul li cheng be added to swiftclient authors file? `git shortlog -nes | grep -i cheng` returns zero results18:21
timburkenotmyname: co-author on joeljwright's patch18:21
joeljwrightnotmyname: still here18:21
joeljwrighthe's the 'real' author of one of the header redaction patches (using a colleague's account)18:22
*** mmuffinman has joined #openstack-swift18:22
notmynameah, ok18:23
notmynamemy current "detect new authors" script doesn't look at co-authored. just what's in the git metadata18:23
joeljwright:)18:23
*** AndreiaKumpera has quit IRC18:25
openstackgerritJohn Dickinson proposed openstack/python-swiftclient: authors/changelog updates for 2.8 release  https://review.openstack.org/28197418:25
*** bjkeller has quit IRC18:25
notmynamedone18:25
*** barker has quit IRC18:29
*** barker has joined #openstack-swift18:30
*** mmuffinman has quit IRC18:30
joeljwrighttimburke: just to complete the header redaction saga - https://review.openstack.org/#/c/28482218:31
*** mingdang1 has joined #openstack-swift18:31
joeljwrighttimburke: it's only 1 line18:31
*** gyee has joined #openstack-swift18:33
*** mingdang1 has quit IRC18:35
*** gyee has quit IRC18:37
*** haomaiwang has quit IRC18:39
timburkeah, yeah, right. will review it shortly18:41
*** mmuffinman has joined #openstack-swift18:41
*** haomaiwang has joined #openstack-swift18:44
*** haomaiwang has quit IRC18:49
notmynamejoeljwright: I changed my mind after thingking about it more18:49
timburkejoeljwright: is it worth adding a non-utf-8 header value, eg for a tempurl key? it'd be a bit of a pain to deal with on a utf-8 command line, but could happen18:49
notmynameI want to keep the authors file as just the people in the git metadata18:49
notmynamein part because we enforce CLA on those who are listed as the author, but you can put anything in the commit message18:49
notmynameshould we add authors if I have a commit message including "this patch would not be possible without the support of my loving wife"?18:50
notmyname(extreme example)18:50
*** mmuffinman has quit IRC18:51
openstackgerritJohn Dickinson proposed openstack/python-swiftclient: authors/changelog updates for 2.8 release  https://review.openstack.org/28197418:52
notmynamehowever, if someone has gone throught he other process to be counted as ATC (ie being individually approved in a vote by the TC), then that's different. we've got a few people like that in swift's authors file18:52
notmynameanyway, I want to keep it simple. if you're listed as an author by git, you're in the authors file18:53
*** klrmn has quit IRC18:53
*** klrmn has joined #openstack-swift18:54
*** mvk has joined #openstack-swift18:54
*** ChubYann has joined #openstack-swift18:55
*** Razva has left #openstack-swift18:56
*** lpabon has quit IRC18:56
*** barker has quit IRC18:59
*** zaitcev has joined #openstack-swift19:00
*** ChanServ sets mode: +v zaitcev19:00
*** davideagle has quit IRC19:00
*** esker has joined #openstack-swift19:01
*** mingyu has quit IRC19:01
*** davideagle_ has joined #openstack-swift19:02
*** loth1 has joined #openstack-swift19:02
*** mingyu has joined #openstack-swift19:02
*** loth has quit IRC19:02
*** tamizh_geek has quit IRC19:04
*** tamizh_geek has joined #openstack-swift19:07
openstackgerritSteve McLellan proposed openstack/swift: WIP Oslo.messaging middleware  https://review.openstack.org/24947119:09
openstackgerritJoel Wright proposed openstack/python-swiftclient: Fix test for redacting sensitive data in client.http_log()  https://review.openstack.org/28482219:10
openstackgerritSamuel Merritt proposed openstack/swift: Let developers/operators add watchers to object audit  https://review.openstack.org/21282419:13
*** bjkeller has joined #openstack-swift19:13
*** mrmoje has joined #openstack-swift19:29
*** petertr7_away is now known as petertr719:35
*** chsc has quit IRC19:37
*** _JZ_ has quit IRC19:43
*** haomaiwang has joined #openstack-swift19:45
*** mmuffinman has joined #openstack-swift19:45
*** haomaiwang has quit IRC19:53
mrmojegood evening/morning!19:57
mrmojeor afternoon even19:57
*** joeljwright has quit IRC19:57
*** trifon has quit IRC20:04
*** petertr7 is now known as petertr7_away20:04
*** mwheckmann has quit IRC20:04
mrmojejust incase anyone gets the chance, i'll be grateful for a review:- https://review.openstack.org/#/c/281700/20:05
patchbotmrmoje: patch 281700 - python-swiftclient - Fix object/path-names for pseudo-folders on winnt20:05
*** esikachev has joined #openstack-swift20:16
*** zhill has quit IRC20:16
*** esker has quit IRC20:21
*** chsc has joined #openstack-swift20:26
hroukota_, poke :D  Take a look at the email I sent, hoping you'll get a change to chat about symlink a bit during the hackaton20:34
*** petertr7_away is now known as petertr720:35
*** proteusguy__ has quit IRC20:46
*** porunov has joined #openstack-swift20:48
*** haomaiwang has joined #openstack-swift20:50
*** loth1 is now known as loth20:53
*** haomaiwang has quit IRC20:58
*** asettle has joined #openstack-swift20:58
mattoliverauGood morning21:00
*** gyee has joined #openstack-swift21:00
*** asettle has quit IRC21:02
*** proteusguy__ has joined #openstack-swift21:05
*** daemontool has joined #openstack-swift21:05
openstackgerritThiago da Silva proposed openstack/python-swiftclient: adding .manpages script to swift client  https://review.openstack.org/28492821:07
tdasilvamattoliverau: good morning!21:08
notmynametdasilva: oh hey. that looks simple21:09
tdasilvanotmyname: :)21:09
tdasilvajust needed it while reviewing copy command21:09
notmynametimburke: have you run current master on py33?21:12
*** davideagle_ has quit IRC21:15
*** mmuffinman has quit IRC21:16
*** NM has quit IRC21:17
*** NM has joined #openstack-swift21:17
*** trifon has joined #openstack-swift21:19
*** rickyrem has joined #openstack-swift21:20
openstackgerritJohn Dickinson proposed openstack/python-swiftclient: authors/changelog updates for 2.8 release  https://review.openstack.org/28197421:21
kota_hrou: ack, sure. Will send the response today21:24
hroukota_, Awesome thanks so much, and I'm going to put a bit more thought into it over the next few days and will update that ehterpad for you all to you use21:27
*** esikachev has quit IRC21:32
*** lpabon has joined #openstack-swift21:33
*** esker has joined #openstack-swift21:36
tdasilvahrou: do you have anything on a spec that is updated?21:38
tdasilvaI was also curious about symlinks21:38
*** ajiang has joined #openstack-swift21:38
mattoliveraume too, and seeing I wont be there, I need something to read.. though I guess I can wait till after y'all talk about it at the hackathon21:39
*** mmuffinman has joined #openstack-swift21:39
*** esker has quit IRC21:41
timburkenotmyname: yes; tests pass fine on my mac. i've not used it against an actual swift cluster much on py33, though; if i'm not using py27, i tend to go for py3521:41
*** esker has joined #openstack-swift21:46
*** daemontool has quit IRC21:48
claygi don't know how to use the new gerrit to trace a patch series down to the root21:48
*** zhill has joined #openstack-swift21:48
claygeverything is "related" - but it seems un-ordered?21:48
claygthe parent sha's aren't clickable21:48
claygoh wait... maybe it's just the one on the bottom - maybe it's no un-ordered21:49
claygif there's multiple patches that share a root tho it seems to get garbled up - and I'm not sure how deep it goes?21:49
*** esker has quit IRC21:51
*** esker has joined #openstack-swift21:51
*** _JZ_ has joined #openstack-swift21:54
*** haomaiwang has joined #openstack-swift21:55
notmynameclayg: I've been wondering the same thing. if you figure it out, please teach me!21:58
*** delatte has quit IRC22:00
claygnotmyname: i have a working theory - bottom one in the list is the bottom of the patch chain - happy to have a counter example?22:00
*** haomaiwang has quit IRC22:01
*** porunov has quit IRC22:01
notmynamenope. so far I'm still at the "not sure what's going on here, better download it and use real tools" stage22:01
notmynamewhich is ok for a single review. but bad for tracking stuff across patches22:01
*** CaioBren_ has quit IRC22:02
claygtry out the last in the list thing - let me know when it fails you - i'll keep you posted22:06
*** dmorita has quit IRC22:11
*** dmorita has joined #openstack-swift22:14
*** daemontool has joined #openstack-swift22:17
*** petertr7 is now known as petertr7_away22:22
*** delatte has joined #openstack-swift22:25
*** StraubTW has quit IRC22:27
*** dmorita has quit IRC22:32
*** dmorita has joined #openstack-swift22:33
*** delattec has joined #openstack-swift22:33
*** delatte has quit IRC22:36
*** arch-nemesis has quit IRC22:44
*** nadeem has quit IRC22:50
*** esker has quit IRC22:52
*** haomaiwang has joined #openstack-swift22:59
*** arch-nemesis has joined #openstack-swift22:59
*** haomaiwang has quit IRC23:01
*** delatte has joined #openstack-swift23:01
*** delattec has quit IRC23:03
*** bjkeller has left #openstack-swift23:06
*** arch-nemesis has quit IRC23:07
*** zhill has quit IRC23:07
*** diazjf has quit IRC23:08
*** dmorita has quit IRC23:12
*** km has joined #openstack-swift23:17
mrmojeclayg: good [ morning | afternoon | evening ].23:17
mrmojeon your previous review on https://review.openstack.org/#/c/271728/,23:18
patchbotmrmoje: patch 271728 - python-swiftclient - Fix segmented upload to pseudo-dir via <container> (MERGED)23:18
mrmojestill think the urljoin cleanup bug worth opening?23:19
*** mrmoje has quit IRC23:20
hroutdasilva, mattoliverau hey guys !  Yea the spec isn't exactly up to date now but I'm not sure I'd go about fixing it as the big unknown in the room so to speak (i.e. how to handle post), is the part that needs refining :)23:20
*** dmorita has joined #openstack-swift23:20
mattoliverauhrou: no stress, talk about it and solve it at hackathon and then update the spec :)23:21
*** trifon has quit IRC23:26
*** mwheckmann has joined #openstack-swift23:27
*** davideagle has joined #openstack-swift23:29
*** dmorita has quit IRC23:30
*** kei_yama has joined #openstack-swift23:32
*** dmorita has joined #openstack-swift23:35
notmynamereviewer and owner timings on http://not.mn/swift/swift_community_dashboard.html were slightly updated to trim outliers23:35
*** chsc has quit IRC23:38
*** davideagle has quit IRC23:38
mattoliverauoh no owners are winning!23:39
*** dmorita has quit IRC23:45
*** ho_away has joined #openstack-swift23:46
*** dmorita has joined #openstack-swift23:46
*** dmorita has quit IRC23:50

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