Wednesday, 2017-05-17

*** d0ugal has quit IRC00:17
*** alenavolk has joined #openstack-swift00:54
*** dims has left #openstack-swift00:55
*** alenavolk has quit IRC00:59
openstackgerritTim Burke proposed openstack/swift master: name_check: better test maximum_length  https://review.openstack.org/46520601:00
*** itlinux has quit IRC01:07
*** NM has joined #openstack-swift01:08
*** d0ugal has joined #openstack-swift01:13
*** itlinux has joined #openstack-swift01:16
*** m_kazuhiro has joined #openstack-swift01:19
m_kazuhirogood morning01:20
*** gyee has quit IRC01:22
*** NM has quit IRC01:24
*** gyee has joined #openstack-swift01:26
*** tovin07_ has joined #openstack-swift01:26
kota_good morning. back to Japan, office after the summit and a few day offs after the summit.01:27
kota_clayg: ack for patcch https://review.openstack.org/#/c/439572/01:29
patchbotpatch 439572 - swift - Limit number of revert tombstone SSYNC requests01:29
kota_acoles: thanks for composite ring builder patch improvement, i will have a time to look at!01:29
*** _JZ_ has joined #openstack-swift01:44
*** m_kazuhiro has quit IRC01:50
*** itlinux has quit IRC01:51
*** Sukhdev has joined #openstack-swift01:55
*** Sukhdev has quit IRC01:56
*** itlinux has joined #openstack-swift02:00
*** bkopilov_ has quit IRC02:21
*** bkopilov has quit IRC02:22
*** itlinux has quit IRC02:25
*** alenavolk has joined #openstack-swift02:27
*** m_kazuhiro has joined #openstack-swift02:30
*** gyee has quit IRC02:35
*** itlinux has joined #openstack-swift02:36
*** vint_bra has joined #openstack-swift02:40
*** alenavolk has quit IRC02:41
*** alenavolk has joined #openstack-swift02:46
*** itlinux has quit IRC02:48
*** itlinux has joined #openstack-swift02:57
*** vint_bra has quit IRC02:58
*** winggundamth has joined #openstack-swift03:14
*** itlinux has quit IRC03:24
*** winggundamth has quit IRC03:26
*** winggundamth has joined #openstack-swift03:27
*** Sukhdev has joined #openstack-swift03:27
*** winggundamth has quit IRC03:27
*** winggundamth has joined #openstack-swift03:31
*** bkopilov_ has joined #openstack-swift03:39
*** bkopilov has joined #openstack-swift03:42
*** links has joined #openstack-swift03:42
*** links has quit IRC03:51
*** links has joined #openstack-swift03:54
*** m_kazuhiro has quit IRC04:22
*** links has quit IRC04:24
*** klrmn has quit IRC04:33
*** links has joined #openstack-swift05:06
*** ChubYann has quit IRC05:11
*** skudlik has joined #openstack-swift05:17
*** Sukhdev has quit IRC05:23
mahaticgood morning05:40
mahatickota_: thanks05:40
*** JimCheung has joined #openstack-swift05:43
*** alenavolk has quit IRC05:46
*** alenavolk has joined #openstack-swift05:47
*** alenavolk has quit IRC05:48
*** pcaruana has joined #openstack-swift05:48
*** alenavolk has joined #openstack-swift05:48
*** alenavolk has quit IRC05:48
*** alenavolk has joined #openstack-swift05:49
*** alenavolk has quit IRC05:49
*** cshastri has joined #openstack-swift05:51
*** skudlik has quit IRC05:59
kota_mahatic: good morning06:17
*** JimCheung has quit IRC06:17
*** rcernin has joined #openstack-swift06:18
kota_mahatic: can i ask you about the patch 439572 ?06:30
patchbothttps://review.openstack.org/#/c/439572/ - swift - Limit number of revert tombstone SSYNC requests06:30
*** janonymous has quit IRC06:42
*** oshritf has joined #openstack-swift06:42
*** mgagne has quit IRC06:53
*** mgagne has joined #openstack-swift06:56
*** mgagne is now known as Guest2410306:56
*** rcernin has quit IRC06:57
*** JimCheung has joined #openstack-swift07:07
*** JimCheung has quit IRC07:11
*** rcernin has joined #openstack-swift07:13
*** geaaru has joined #openstack-swift07:19
*** skudlik has joined #openstack-swift07:27
*** rcernin has quit IRC07:29
*** rcernin has joined #openstack-swift07:35
*** rcernin has quit IRC07:35
*** rcernin has joined #openstack-swift07:35
*** jaosorior_away is now known as jaosorior07:48
*** geaaru has quit IRC08:04
*** geaaru has joined #openstack-swift08:17
*** cbartz has joined #openstack-swift08:20
*** janonymous has joined #openstack-swift08:57
*** links has quit IRC09:48
*** silor has joined #openstack-swift09:54
*** tovin07_ has quit IRC09:56
*** links has joined #openstack-swift10:00
mahatickota_: sorry, I had to step away in the morning10:15
mahatickota_: are you still around?10:15
*** oshritf has quit IRC10:19
*** oshritf has joined #openstack-swift10:20
*** silor1 has joined #openstack-swift10:32
*** silor has quit IRC10:34
*** silor1 is now known as silor10:34
*** bkopilov_ has quit IRC10:39
*** bkopilov has quit IRC10:39
-openstackstatus- NOTICE: gerrit is being restarted to help stuck git replication issues10:55
*** links has quit IRC11:05
*** kjz has joined #openstack-swift11:21
*** links has joined #openstack-swift11:22
*** silor has quit IRC11:24
*** silor has joined #openstack-swift11:25
*** mathiasb has quit IRC11:28
*** zul has quit IRC11:39
*** zul has joined #openstack-swift11:41
*** silor has quit IRC11:44
*** silor has joined #openstack-swift11:44
*** chlong has quit IRC11:57
*** ma9_ has joined #openstack-swift12:37
*** vint_bra has joined #openstack-swift12:46
*** alenavolk has joined #openstack-swift12:49
*** jarbod_ has quit IRC13:07
*** jarbod_ has joined #openstack-swift13:07
*** lan has quit IRC13:11
*** mathiasb has joined #openstack-swift13:13
*** neonpastor has joined #openstack-swift13:14
*** chlong has joined #openstack-swift13:15
*** NM has joined #openstack-swift13:32
*** klamath has joined #openstack-swift13:43
*** klamath has quit IRC13:45
*** klamath has joined #openstack-swift13:45
*** NM has quit IRC13:45
*** NM has joined #openstack-swift13:49
*** ma9_ has left #openstack-swift13:51
*** vint_bra has quit IRC14:03
*** NM has quit IRC14:03
*** NM has joined #openstack-swift14:05
*** vint_bra has joined #openstack-swift14:06
*** itlinux has joined #openstack-swift14:12
*** lcurtis has joined #openstack-swift14:13
*** Guest24103 is now known as mgagne14:32
*** mgagne has quit IRC14:32
*** mgagne has joined #openstack-swift14:32
*** gyee has joined #openstack-swift14:33
*** itlinux has quit IRC14:39
*** cshastri has quit IRC14:49
*** links has quit IRC14:55
*** geaaru has quit IRC14:59
*** rcernin has quit IRC15:04
*** itlinux has joined #openstack-swift15:09
*** geaaru has joined #openstack-swift15:12
*** psachin has joined #openstack-swift15:38
*** Sukhdev has joined #openstack-swift15:39
*** skudlik has quit IRC15:51
*** klrmn has joined #openstack-swift15:53
notmynamegood morning15:58
*** ChanServ changes topic to "Swift is 7 years old! | Logs: http://eavesdrop.openstack.org/irclogs/%23openstack-swift/ | Meetings: https://wiki.openstack.org/wiki/Meetings/Swift | Review Dashboard: http://not.mn/reviews.html"15:58
notmynametoday is Swift's birthday! seven years ago, it was first put into prod15:59
rledisezhappy birthday Swift!16:02
rledisezkudos to the creators, real storage heroes :)16:02
notmynamerledisez: kudos to the operators, the ones who actually give it meaning :-)16:05
* tdasilva wishes more creators and operators would hang out here16:07
mwheckmannhappy bday!16:07
*** klrmn has quit IRC16:09
pdardeauhappy birthday Swift!16:09
pdardeaunotmyname: you gonna provide some interesting stats today?16:09
notmynamepdardeau: no different than I have normally. I haven't mined for anything new (but I did bring corn casserole for the company pot-luck today)16:09
tdasilvaheh16:10
mwheckmannnotmyname: PS after a bunch more testing, it looks the increases in first by write latency that I reported yesterday coincided with changes in TLS certs (move to SHA256 based certs) and some other HAproxy cert/cipher related changes. So false alarm. Now if I can find a way to minimize TLS nego. time.16:10
mwheckmannfirst byte latency I mean16:10
notmynamemwheckmann: oh, interesting. good find!16:10
*** gyee_ has joined #openstack-swift16:11
*** pcaruana has quit IRC16:11
mahatichappy birthday Swift!16:11
mahaticso that makes it my two years with Swift as well!16:11
*** jaosorior is now known as jaosorior_away16:17
notmynameonovy: how'd the daytime traffic go today? did you break the internet? ;-)16:20
*** gyee has quit IRC16:20
*** alenavolk has quit IRC16:26
*** alenavolk has joined #openstack-swift16:27
*** alenavolk has quit IRC16:27
*** alenavolk has joined #openstack-swift16:27
*** alenavolk has quit IRC16:44
*** alenavolk has joined #openstack-swift16:45
*** alenavolk has quit IRC16:45
*** alenavolk has joined #openstack-swift16:45
*** alenavolk has quit IRC16:46
*** mwheckmann has quit IRC17:07
claygman... we need to get some composite ring stuff merged17:07
*** klrmn has joined #openstack-swift17:07
*** cbartz has quit IRC17:12
*** winggundamth has quit IRC17:12
*** oshritf has quit IRC17:13
*** chsc has joined #openstack-swift17:24
*** chsc has joined #openstack-swift17:24
*** Sukhdev has quit IRC17:35
*** geaaru has quit IRC17:41
tdasilvanotmyname: hi, do you happen to have a list/count of how many public clouds use swift as their object storage offering?17:44
*** alenavolk has joined #openstack-swift17:45
*** ChubYann has joined #openstack-swift17:47
*** itlinux has quit IRC17:50
*** geaaru has joined #openstack-swift17:53
*** skudlik has joined #openstack-swift17:57
*** jamielennox is now known as jamielennox|away17:59
*** alenavolk has quit IRC18:05
*** jamielennox|away is now known as jamielennox18:05
*** alenavolk has joined #openstack-swift18:06
*** zul has quit IRC18:06
notmynametdasilva: I do not. I could list of several off the top of my head, but I know I'd be missing a lot in EU and Asia18:07
tdasilvanotmyname: i came across this list: https://docs.openstack.org/developer/os-client-config/vendor-support.html and that got me thinking how many are using swift18:08
notmynametdasilva: clouda, runabove, and softlayer are ones I know of that aren't listed there. (runabove is sortof separate, but also sortof a R&D lab for OVH?)18:10
notmynameI think there's a few more in China maybe, too18:10
*** Sukhdev has joined #openstack-swift18:11
*** silor has quit IRC18:12
notmynametdasilva: as for stats, rerunning them right now. 673 total contributors to swift (commit authors + gerrit reviews)18:15
*** alenavolk has quit IRC18:16
*** psachin has quit IRC18:23
*** geaaru has quit IRC18:23
*** itlinux has joined #openstack-swift18:36
*** zul has joined #openstack-swift18:37
timurshould the python-swifclient default be SLO and not DLO? are there folks that rely on the default behavior being DLO?18:37
timurfollow up question: when using the bulk delete middleware with SLO, it only deletes the manifests. Should it remove the segments as well? While it's consistent with the behavior of DELETE, it seems that it would be confusing to consumers18:39
claygtimur: I think there would need to be a way for the client to explicitly state they want to delete the manifest and segemnts - the bulk middleware can't just change it's behavior to start deleting them18:41
claygtimur: I think there's a good case to be made changing the python-swiftclient cli default from DLO to SLO - it supports both and SLO is better suited to the python-swiftclient cli use-case18:42
claygtimur: I'd be most worried about dropping the --use-slo flag (for scripts that have already adopted the move)18:43
claygtimur: but if someone added a --use-dlo flag and then with some sort of version/release bump flopped over the default when neither flag is specified - I think mostly no one would notice and everyone would be happy18:43
claygtimur: how's google io/offices/w/e?18:44
timurclayg: good point on dropping the flag. A --use-dlo flag would make sense. For the bulk delete, maybe it can understand the ?multipart-manifest=delete flag and do the right thing? would that work?18:45
claygwfm18:45
timurthe SF office is nothing special -- same as all other office buildings. The keynote is a little interesting: they announced Kotlin as the new official Android language and spent a bunch of time talking about how to monetize Youtube (more so than just ads)18:46
claygoh nice, my kids basically all want to be youtubers when they grow up18:47
*** MVenesio has joined #openstack-swift18:47
timurone example was that you could pay to have your comment bubble up to the top and be pinned on a video... which sounds like it could go super wrong18:47
timurI think notmyname said that at least one of his kids wants to be a youtube blogger18:47
timurso, there you go! Google is trying to help them make more money :)18:48
claygsweet - i'm into that18:50
notmynamehow else are we going to be in that retirement home?18:51
*** catintheroof has joined #openstack-swift18:59
*** alenavolk has joined #openstack-swift19:26
*** Sukhdev has quit IRC19:47
notmynameswift team meeting in 3 minutes in #openstack-meeting19:58
tdasilvamoved back 1 hr?19:58
notmynameno. when I said 3 minutes, clearly I meant 63 minutes19:59
notmyname;-)19:59
notmynameI totally misread my calendar. you'd think it was my first time at a swift meeting or something19:59
notmynamewhew. in that case, back to get another plate of food ;-)19:59
tdasilvalol20:00
*** vint_bra has quit IRC20:01
notmynametdasilva: thanks for keeping me honest ;-)20:01
tdasilvanotmyname: i really just assumed that i had missed some announcement. I think there was some convo last week about changing times and thought you guys actually changed it20:02
notmynameno, but that's actually something on the agenda today20:07
*** vint_bra has joined #openstack-swift20:08
*** mwheckmann has joined #openstack-swift20:11
onovynotmyname: we did20:21
onovyhttps://www.nix.cz/cs/port/mrtg2/nix-agr-11420:21
*** vint_bra has quit IRC20:22
notmynameonovy: cool20:22
onovylook to weekly graph what is "normal" for us20:23
notmynameyeah :-)20:23
notmynamequite a spike20:23
onovy:)20:23
*** vint_bra has joined #openstack-swift20:26
*** itlinux_ has joined #openstack-swift20:34
*** oshritf has joined #openstack-swift20:36
*** itlinux has quit IRC20:37
*** oshritf has quit IRC20:42
claygonovy: rledisez: ahale: briancline: you ever notice account/container "temporary" db's show up in <nodes>/<drive>/tmp/<uuid>???20:42
rledisezclayg: yes, when replicator fails to replicate a big db. is that what you're thinking about?20:45
rledisezrsync is done in tmp folder, and then… well, it fails20:45
notmynameany info on *why* it fails?20:46
clayghugokuo: spotted this -> https://github.com/openstack/swift/blob/6e893e228840bc42cfd13546245438832bc2bb46/swift/common/db_replicator.py#L82020:46
hugokuoreading20:47
rlediseznot really. i tried to debug it a bit, increased all possible max diff/timeout values. and never got it working. i'm just waiting for container sharding ;)20:47
claygin this case he has some rsync logs pointing at a race - two remote both trying to rsync onto a newly added container server around the same time - one finishes and calls complete_rsync (which flops the uuid named database into the right spot) - the other calls complete_rysnc and gets a 404 (but the container-server doesn't cleanup the uuid-named database he20:47
claygleft behind)20:47
hugokuoclayg: the db_replicator is called by container-replicator right ?20:48
claygyes, but that section of the module I'm linking at is acctually the ReplicatorRpc which is running and routing requests behind the REPLCIATE verb of the replication container server20:50
*** alenavolk has quit IRC20:51
*** zul has quit IRC20:51
*** chlong has quit IRC20:52
notmynameswift team meeting in 3 minutes (right tdasilva?)20:57
kota_morning20:58
tdasilvaheh20:58
tdasilva:P20:58
notmynamehello kota_20:58
kota_notmyname: o/20:59
tdasilvanotmyname: since you are allowing me to pick on you: t-minus 1.5 wks to Boston ???21:00
mattoliverauA cleanup in the replicator afrer a failure could be handy, or a way of continuing on the 404, but the RPC won't know the uuid of the rsynced file next time. So maybe try and cleanup after itself or maybe insert at least 1 retry in the replicator -> RPC communication21:00
*** oshritf has joined #openstack-swift21:00
mattoliverauOh and morning21:00
notmynametdasilva: yep21:00
*** catintheroof has quit IRC21:01
mattoliverauOh and happy birthday swift!21:02
*** Sukhdev has joined #openstack-swift21:05
*** MVenesio has quit IRC21:11
*** itlinux_ has quit IRC21:14
*** skudlik has quit IRC21:19
rledisezclayg: LAFS or LAWLS? i need translation for these ones (though, i like LAWLS, sounds cool :D)21:36
clayghttp://www.urbandictionary.com/define.php?term=lawl21:36
claygaka "the only reason anyone ever does anything on the internet" "for the lawls"21:36
rledisez:D21:37
*** NM has quit IRC21:37
*** oshritf has quit IRC21:37
*** Sukhdev has quit IRC21:50
*** PavelK has joined #openstack-swift21:52
PavelKclayg: you +1 patch 448480 if failing probe tests will be removed. I noticed on "idea page" that probe tests should be on 8 devices, as you suggested. Do you think that this code cleanup can be merged?21:55
patchbothttps://review.openstack.org/#/c/448480/ - swift - DB replicator cleanup21:55
*** vint_bra has quit IRC22:00
claygshit that was over quick22:02
PavelKit is code cleanup and not updated DB replica is removed if it is on enough replicas (and if it is correct solution then I would like to make simmilar cleanup for objects to patch lp bug 1690791)22:03
openstackLaunchpad bug 1690791 in OpenStack Object Storage (swift) "Objects are not deleted from additional handoff when primary disk fails" [Undecided,New] https://launchpad.net/bugs/169079122:03
claygwe didn't merge that yet!???22:03
PavelKno :-(, you +1 with suggestions and then was summit comming...22:03
* kota_ is leaving to find his breakfast, will be back in the office22:04
claygPavelK: ok, thanks - i got it - i'm sure it will be great!22:04
PavelKthanks22:04
notmynamezaitcev: unfortunately https://review.openstack.org/#/c/427911/ has a merge conflict. can you help resolve that soon?22:15
patchbotpatch 427911 - swift - PUT+POST and its development test22:15
zaitcevnotmyname: I had it resolved locally, just wanted to keep Tim's comments visible for someone else.22:16
*** klamath has quit IRC22:17
notmynameI thought a rebase kept the comments22:20
notmynameI don't remember22:21
zaitcevTrue, they are still available through older versions, but you have to click in. Also, if someone sees the same space and comments on it, they are not in the same place.22:21
notmynameah right. I think it's the votes that persist across a rebase22:23
openstackgerritPete Zaitcev proposed openstack/swift master: PUT+POST and its development test  https://review.openstack.org/42791122:28
*** itlinux has joined #openstack-swift22:33
*** lcurtis has quit IRC22:38
notmynamehttps://wiki.openstack.org/wiki/Swift/PriorityReviews is updated. please take a look22:40
zaitcevyay22:48
*** itlinux has quit IRC22:49
notmynamehttp://lists.openstack.org/pipermail/openstack-dev/2017-May/116947.html22:54
notmynameclayg: you and I are mentioned in that message22:54
*** itlinux has joined #openstack-swift22:59
*** alenavolk has joined #openstack-swift23:10
*** alenavolk has quit IRC23:13
*** alenavolk has joined #openstack-swift23:14
*** alenavolk has joined #openstack-swift23:16
*** alenavolk has quit IRC23:20
*** alenavolk has joined #openstack-swift23:22
*** kei_yama has joined #openstack-swift23:22
claygcalled out!23:25
*** itlinux has quit IRC23:27
*** NM has joined #openstack-swift23:27
*** alenavolk has quit IRC23:38
*** chsc has quit IRC23:41
*** vint_bra has joined #openstack-swift23:58
openstackgerritMerged openstack/swift master: name_check: better test maximum_length  https://review.openstack.org/46520623:59

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