| openstackgerrit | Merged openstack/python-swiftclient: Remove all DLO segments on upload of replacement https://review.openstack.org/161972 | 00:04 |
|---|---|---|
| openstackgerrit | Merged openstack/python-swiftclient: Add connection release test https://review.openstack.org/159076 | 00:04 |
| openstackgerrit | Merged openstack/swift: Increase httplib._MAXHEADERS https://review.openstack.org/197556 | 00:09 |
| openstackgerrit | Merged openstack/python-swiftclient: Add passenv to tox.ini to make functests run with tempauth https://review.openstack.org/188080 | 00:14 |
| openstackgerrit | Merged openstack/python-swiftclient: Run functional tests using keystone auth options https://review.openstack.org/123514 | 00:14 |
| notmyname | yay | 00:16 |
| mattoliverau | nice work! | 00:17 |
| *** dmorita has joined #openstack-swift | 00:21 | |
| notmyname | ok, a bunch of SMR drives are racked. now time to set up swift and do some testing | 00:22 |
| notmyname | "a bunch" == 33 | 00:22 |
| ho | notmyname: expectation for the env is increasing i/o performance? | 00:25 |
| mattoliverau | notmyname: you guys get to play with all the toys, /me jelous | 00:25 |
| ho | mattoliverau: +1 | 00:25 |
| torgomatic | ho: I think the expectation is to experiment with Swift on SMR drives and see how things perform | 00:29 |
| *** zhill has quit IRC | 00:31 | |
| ho | torgomatic: thanks! sound interesting :) | 00:32 |
| openstackgerrit | Alexandra Settle proposed openstack/swift: Correcting minor grammatical errors https://review.openstack.org/195825 | 00:32 |
| *** zhill has joined #openstack-swift | 00:44 | |
| *** zhill has quit IRC | 00:49 | |
| *** nadeem has joined #openstack-swift | 00:58 | |
| *** jasondotstar has joined #openstack-swift | 01:02 | |
| *** nadeem has quit IRC | 01:03 | |
| *** nadeem has joined #openstack-swift | 01:06 | |
| *** blmartin has joined #openstack-swift | 01:07 | |
| blmartin | Hello all | 01:08 |
| ho | blmartin: hello | 01:17 |
| blmartin | how are you doing today, ho? | 01:18 |
| ho | blmartin: good, thanks! | 01:20 |
| ho | blmartin: I re-started to develop function tests for access control. it's tough :) | 01:21 |
| *** blmartin has quit IRC | 01:23 | |
| *** nadeem has quit IRC | 01:26 | |
| *** jasondotstar has quit IRC | 01:27 | |
| *** annegentle has quit IRC | 01:28 | |
| *** zhill has joined #openstack-swift | 01:35 | |
| *** barker has joined #openstack-swift | 01:39 | |
| *** zhill has quit IRC | 01:39 | |
| *** occup4nt has joined #openstack-swift | 01:46 | |
| *** barker has quit IRC | 01:48 | |
| *** mutoulbj has joined #openstack-swift | 01:49 | |
| *** mutoulbj has quit IRC | 01:49 | |
| *** occupant has quit IRC | 01:50 | |
| *** wuhg has joined #openstack-swift | 01:52 | |
| hrou | notmyname: We should touch base, we're planning on doing some work with swift and SMRs as well ! | 01:55 |
| *** blmartin has joined #openstack-swift | 01:55 | |
| blmartin | sorry about that ho, I disconnected. we got a new router and it is not... particularly.... stable? | 02:03 |
| openstackgerrit | Matthew Oliver proposed openstack/swift: Add concurrent reads option to proxy https://review.openstack.org/117710 | 02:05 |
| mattoliverau | acoles_away: ^^ you were right, all I needed was a good night sleep, it was obvious this morning | 02:06 |
| *** janonymous_ has quit IRC | 02:13 | |
| clayg | mattoliverau: 1) nice work 2) acoles is always right | 02:14 |
| *** mutoulbj has joined #openstack-swift | 02:14 | |
| mattoliverau | clayg: it's clay!! yay! and yes he is.. which is way we should have a count down timer for the return of acoles ;) | 02:15 |
| blmartin | clayg, what do you think of the notes on bug 1470576 ? it seems like it could just be a documentation fix. I have just sat down and am going to try and replicate with my saio. | 02:20 |
| openstack | bug 1470576 in OpenStack Object Storage (swift) "mount_check does not prevent writing to root mount" [Medium,Confirmed] https://launchpad.net/bugs/1470576 - Assigned to Ben Martin (blmartin) | 02:20 |
| clayg | blmartin: is that the mount thing? | 02:21 |
| blmartin | it is. where it starts filling up the root drive on unmount | 02:21 |
| clayg | blmartin: I don't think it's *just* a doc fix - permissions or symlinks - both solutions are best practice - but even when mount_check = false, or when the permissions are pervasive - we shouldn't write data onto the root partition - it would be easy to avoid if we tracked down the call(s) to makedirs that can create the device path and fixed them | 02:22 |
| clayg | ultimatly makedirs is just a recursive mkdir that will catch the enoent - we can write that in python and and stop before trying to create the device dir | 02:23 |
| clayg | it'd be better for everyone - the best practices documentation would be good too - zaitcev may have been trying to make the point about the priority of the doc bug over making the code more robust - I'm not entirely sure. | 02:24 |
| zaitcev | You can't make it robust. And it's every 18 months to 2 years someone comes up with this, then finds out that it's inevitably racy. | 02:26 |
| *** blmartin has quit IRC | 02:26 | |
| zaitcev | I remember fixing the docs about it, to make it clearer that the mount monts must be owned by root, but the mounted filesystem's root owned by swift. | 02:26 |
| *** zhill has joined #openstack-swift | 02:32 | |
| *** bill_az has quit IRC | 02:33 | |
| *** zhill has quit IRC | 02:37 | |
| notmyname | hrou: cool | 02:37 |
| notmyname | hi clayg. did you throw the guy into prison? | 02:37 |
| *** zhill has joined #openstack-swift | 02:37 | |
| openstackgerrit | Merged openstack/python-swiftclient: Document missing functional test config option https://review.openstack.org/187503 | 02:39 |
| notmyname | hrou: what company are you with? | 02:39 |
| hrou | notmyname: IBM | 02:40 |
| *** blmartin has joined #openstack-swift | 02:40 | |
| notmyname | hrou: ah, great. I've talked to a few people there already about them. very good info | 02:40 |
| zaitcev | clayg: Unfortunately, the Joe's book relies on the correct order just like any other document: it tells to mount first, then chown. That's fine, but there's no explanation anywhere what happens if you accidentially chown the mount point itself (like blmartin has done). | 02:41 |
| *** zhill has quit IRC | 02:41 | |
| hrou | notmyname: Oh that's great to hear, I can drop you a note and we can touch base further if you like, we're probably just doing some basic profiling to start but have some other ideas longer term in terms of optimizing for SMR use in swift. | 02:42 |
| hrou | notmyname: You're a step ahead, still waiting on the h/w on our end ; - ) | 02:42 |
| notmyname | hrou: what group are you in? I talked with Ann Corrao and Steve Hetzler | 02:42 |
| hrou | notmyname: Ha ; - ) You can potentially ignore that then, I'm working with Ann actually. | 02:43 |
| notmyname | hrou: heh ok :-) | 02:43 |
| hrou | notmyname: Either way ! We'll stay in touch. | 02:44 |
| notmyname | ok :-) | 02:44 |
| *** darrenc is now known as darrenc_afk | 02:45 | |
| *** blmartin has quit IRC | 02:50 | |
| *** blmartin has joined #openstack-swift | 02:51 | |
| *** blmartin has quit IRC | 03:14 | |
| *** silor has joined #openstack-swift | 03:15 | |
| timburke | notmyname: fyi, with that last merge, all the approved python-swiftclient changes have cleared the gate | 03:18 |
| notmyname | yay | 03:18 |
| notmyname | timburke: thanks again for your work on all that :-) | 03:18 |
| timburke | notmyname: np. and with torgomatic's plunge, i think the remaining -1s are legit (save zackmdavis's WIP) | 03:21 |
| *** jkugel has joined #openstack-swift | 03:26 | |
| *** kota_ has joined #openstack-swift | 03:31 | |
| *** ChanServ sets mode: +v kota_ | 03:31 | |
| *** zhill has joined #openstack-swift | 03:37 | |
| *** zhill has quit IRC | 03:41 | |
| *** jkugel has quit IRC | 03:53 | |
| *** jkugel has joined #openstack-swift | 03:54 | |
| *** SkyRocknRoll has joined #openstack-swift | 04:00 | |
| *** jkugel has quit IRC | 04:19 | |
| *** DericHorn-HP has joined #openstack-swift | 04:19 | |
| *** DericHorn-HP has quit IRC | 04:23 | |
| *** ianbrown has joined #openstack-swift | 04:31 | |
| *** zhill has joined #openstack-swift | 04:41 | |
| *** delattec has joined #openstack-swift | 04:43 | |
| *** janonymous has quit IRC | 04:44 | |
| *** cdelatte has quit IRC | 04:45 | |
| *** redbo has quit IRC | 04:46 | |
| *** gvernik has joined #openstack-swift | 04:49 | |
| *** redbo has joined #openstack-swift | 04:52 | |
| *** ChanServ sets mode: +v redbo | 04:52 | |
| *** shlee322 has joined #openstack-swift | 04:54 | |
| *** early has quit IRC | 04:55 | |
| *** shlee322 has quit IRC | 04:56 | |
| *** darrenc_afk is now known as darrenc | 05:00 | |
| *** amit213 has joined #openstack-swift | 05:00 | |
| *** early has joined #openstack-swift | 05:02 | |
| *** zaitcev has quit IRC | 05:11 | |
| *** DericHorn-HP has joined #openstack-swift | 05:18 | |
| *** nadeem has joined #openstack-swift | 05:19 | |
| *** ppai has joined #openstack-swift | 05:19 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Ignore invalid PIDs in swift-init https://review.openstack.org/186703 | 05:20 |
| *** SkyRocknRoll has quit IRC | 05:33 | |
| *** gvernik has quit IRC | 05:43 | |
| *** gvernik has joined #openstack-swift | 05:44 | |
| *** gvernik has joined #openstack-swift | 05:44 | |
| *** DericHorn-HP has quit IRC | 05:45 | |
| *** gvernik has quit IRC | 05:47 | |
| *** gvernik has joined #openstack-swift | 05:48 | |
| *** gvernik has quit IRC | 05:52 | |
| *** gvernik has joined #openstack-swift | 05:52 | |
| *** gvernik has quit IRC | 05:54 | |
| *** gvernik has joined #openstack-swift | 05:54 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:59 | |
| *** gvernik has quit IRC | 06:01 | |
| *** gvernik has joined #openstack-swift | 06:03 | |
| *** gvernik_ has joined #openstack-swift | 06:04 | |
| *** gvernik has quit IRC | 06:08 | |
| *** gvernik_ has quit IRC | 06:10 | |
| *** gvernik has joined #openstack-swift | 06:11 | |
| *** gvernik has quit IRC | 06:20 | |
| *** gvernik has joined #openstack-swift | 06:21 | |
| *** gvernik has quit IRC | 06:22 | |
| *** gvernik has joined #openstack-swift | 06:22 | |
| *** gvernik has quit IRC | 06:24 | |
| *** ianbrown has quit IRC | 06:29 | |
| *** gvernik has joined #openstack-swift | 06:46 | |
| *** gvernik has quit IRC | 06:48 | |
| *** gvernik has joined #openstack-swift | 06:49 | |
| *** gvernik has quit IRC | 06:53 | |
| *** zhill has quit IRC | 06:54 | |
| *** gvernik has joined #openstack-swift | 07:07 | |
| *** nexusz99 has joined #openstack-swift | 07:07 | |
| *** gvernik has quit IRC | 07:09 | |
| *** fifieldt has quit IRC | 07:09 | |
| *** rledisez has joined #openstack-swift | 07:10 | |
| *** hrou has quit IRC | 07:13 | |
| *** occup4nt has quit IRC | 07:21 | |
| *** aix has joined #openstack-swift | 07:24 | |
| *** mutoulbj_ has joined #openstack-swift | 07:32 | |
| *** mutoulbj has quit IRC | 07:34 | |
| *** acoles_away is now known as acoles | 07:39 | |
| *** geaaru has joined #openstack-swift | 07:39 | |
| acoles | ho: congrats and good luck for the final :) | 07:40 |
| *** jistr has joined #openstack-swift | 07:48 | |
| mattoliverau | acoles: morning | 07:50 |
| mattoliverau | acoles: how did your son's critet match go? | 07:50 |
| mattoliverau | cricket | 07:50 |
| *** acoles has quit IRC | 07:51 | |
| * mattoliverau typing is great this evening :P | 07:51 | |
| *** nadeem has quit IRC | 07:55 | |
| *** pksingh has joined #openstack-swift | 07:59 | |
| pksingh | kota_, san, if you have time can you review https://review.openstack.org/#/c/194464/ | 08:00 |
| *** proteusguy has quit IRC | 08:05 | |
| *** jordanP has joined #openstack-swift | 08:06 | |
| *** acoles has joined #openstack-swift | 08:07 | |
| *** ChanServ sets mode: +v acoles | 08:07 | |
| acoles | hey thanks mattoliverau : they lost, but it was hottest day on record in uk, the oppo were good and we think a little older - a touring school team from south africa. i did slip slap slop :) | 08:07 |
| *** dosaboy has quit IRC | 08:10 | |
| *** dosaboy has joined #openstack-swift | 08:11 | |
| *** dosaboy has quit IRC | 08:12 | |
| *** fbo has joined #openstack-swift | 08:14 | |
| *** proteusguy has joined #openstack-swift | 08:16 | |
| *** jamielennox is now known as jamielennox|away | 08:20 | |
| *** dosaboy has joined #openstack-swift | 08:21 | |
| ho | acoles: thanks! and sorry for it! | 08:23 |
| ho | all, do you know why is 'project-domain-id' in sysmeta necessary? Is there any places for using it except keystoneauth? (I didn't find it) | 08:28 |
| acoles | ho: do you mean in account sysmeta? | 08:31 |
| ho | acoles: yes. | 08:33 |
| acoles | ho: it is used by keystoneauth as you say. it is there to deal with a backwards compatibility in container ACLs... | 08:34 |
| acoles | ho: before keystone v3 introduced domains, project-names were assumed to be globally unique, but now with domains they are not. | 08:35 |
| acoles | unfortunately swift deployments may have project names in container ACLs | 08:35 |
| acoles | So we still match ACLs based on names IF the account is in the default domain and the requesting user/proj are alos in default domain. | 08:36 |
| acoles | s/alos/also/ | 08:37 |
| ho | acoles: thanks! i got strange feeling for the code. the project-domain-id is gotten from the request. | 08:37 |
| ho | acoles: i'm not sure but it can be removed??? | 08:38 |
| ho | acoles: background is i try to migrate existing code to oslo policy. | 08:38 |
| acoles | ho: right, so in some/many case we do not know the project domain id for the account because the account existed before this code was written :O So if the sysmeta value is not set we assume account project is in default domain. <-- that means we should now always set it so that we know when an account is not in default domain, and no it cannot be removed. | 08:39 |
| acoles | ho: yes i understand why you are asking but I think the only way to remove is to lose the backwards compatibility i.e. to never match names in acls, and we made a LOT of effort to keep that compatibility | 08:40 |
| *** jamielennox|away is now known as jamielennox | 08:41 | |
| ho | acoles: thanks! you mentioned about this line right? https://github.com/openstack/swift/blob/master/swift/common/middleware/keystoneauth.py#L371 | 08:44 |
| acoles | ho: yes. that method _is_name_allowed_in_acl() is all about deciding if the backwards compatible name matching is allowed. If it returns false then only project-ID's will be matched in ACLs | 08:47 |
| *** aix has quit IRC | 08:50 | |
| *** foexle has joined #openstack-swift | 08:50 | |
| *** cazino has joined #openstack-swift | 08:53 | |
| *** silor has quit IRC | 08:54 | |
| ho | acoles: the logic allows the name if domains' ids are default and cross project access. And the domain id is registered here: https://github.com/openstack/swift/blob/master/swift/common/middleware/keystoneauth.py#L423. (https://github.com/openstack/swift/blob/master/swift/common/middleware/keystoneauth.py#L349) | 08:56 |
| ho | acoles: the logic uses the domain-id from request (yeah, domain is a concept of keystone basically). v2 is none, v3 is valid id if the request is not for cross project access (this case can not get a domain id). | 08:56 |
| acoles | ho: correct | 08:59 |
| ho | acoles: i understand that i need more time to read it :) | 09:00 |
| ho | acoles: that means if there is no request on the account before the logic can not cover the backward compatibility. maybe... | 09:00 |
| acoles | if the account existed before this code was deployed then the sysmeta project id will be missing and default to None, which is interpreted as the account being in the default domain. It is assumed that all pre-existing accounts were in the default domain (swift had a doc warning people nto to use keystone v3 before this change went into code). | 09:03 |
| *** kota_ has quit IRC | 09:03 | |
| acoles | after this change was added, all new accounts should have the sysmeta project id set IF a keystone v3 token is used. If a v2 token is used then again it is assumed the project domain is default (because v2 has no domains). | 09:04 |
| acoles | clarification: If a v2 token is used *when the account gets created* | 09:05 |
| *** openstackgerrit has quit IRC | 09:19 | |
| *** openstackgerrit has joined #openstack-swift | 09:20 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Add notification middleware https://review.openstack.org/196755 | 09:21 |
| *** jasondotstar has joined #openstack-swift | 09:22 | |
| *** pbourke has quit IRC | 09:25 | |
| *** madhuri_ has left #openstack-swift | 09:25 | |
| *** pbourke has joined #openstack-swift | 09:25 | |
| *** aix has joined #openstack-swift | 09:34 | |
| *** proteusguy has quit IRC | 09:38 | |
| *** proteusguy has joined #openstack-swift | 09:41 | |
| *** pksingh has quit IRC | 09:57 | |
| *** mutoulbj_ has quit IRC | 10:19 | |
| *** mutoulbj has joined #openstack-swift | 10:20 | |
| ho | acoles: thanks for the explanation! _is_name_allowed_in_acl returns True if the domain id is same as default domain or same projects (same domain). So this method allows names for acl if the request access to same domain. And v2 case you mentioned "It is assumed that all pre-existing accounts were in the default domain". My rough code image is http://paste.openstack.org/show/335260/ | 10:23 |
| *** mutoulbj has quit IRC | 10:25 | |
| *** silor has joined #openstack-swift | 10:26 | |
| *** dmorita has quit IRC | 10:37 | |
| *** proteusguy has quit IRC | 10:46 | |
| *** proteusguy has joined #openstack-swift | 10:47 | |
| acoles | ho: thats not quite the same - the requirement is that requesting user and project are in the default domain AND the target account/project is in the default domain. Doc here http://docs.openstack.org/developer/swift/overview_auth.html#access-control-using-keystoneauth. | 10:49 |
| acoles | ho: it is not sufficient for requesting user and project to be in the same domain. | 10:49 |
| *** proteusguy has quit IRC | 10:50 | |
| *** proteusguy has joined #openstack-swift | 10:51 | |
| ho | acoles: but in the logic the domain id of the target account/project is created from the request or previous request. I might mis-understand here. my head overheated... I will read the doc tomorrow. thanks for your help. :-) | 10:54 |
| *** proteusguy has quit IRC | 10:55 | |
| *** proteusguy has joined #openstack-swift | 10:56 | |
| acoles | ho: yes but it is only created when the request token's project == account (see _set_project_domain_id) in which case the domain id of the "target" account is known. | 10:56 |
| acoles | ho: and yes it is pretty complicated and hurt my head too!! :) | 10:57 |
| *** proteusguy has quit IRC | 10:58 | |
| *** proteusguy has joined #openstack-swift | 10:59 | |
| ho | acoles: thanks! have a nice day! | 11:00 |
| *** ho has quit IRC | 11:00 | |
| acoles | ho: np. have a good evening | 11:00 |
| *** proteusguy has quit IRC | 11:03 | |
| *** proteusguy has joined #openstack-swift | 11:04 | |
| *** hseipp has joined #openstack-swift | 11:08 | |
| *** adiantum has quit IRC | 11:12 | |
| *** mutoulbj has joined #openstack-swift | 11:14 | |
| *** proteusguy has quit IRC | 11:19 | |
| *** proteusguy has joined #openstack-swift | 11:20 | |
| *** lpabon has joined #openstack-swift | 11:33 | |
| *** proteusguy has quit IRC | 11:37 | |
| *** proteusguy has joined #openstack-swift | 11:40 | |
| *** NM has joined #openstack-swift | 11:40 | |
| *** proteusguy has quit IRC | 11:43 | |
| *** proteusguy has joined #openstack-swift | 11:43 | |
| *** proteusguy has quit IRC | 11:46 | |
| *** proteusguy has joined #openstack-swift | 11:47 | |
| *** jasondotstar has quit IRC | 11:50 | |
| *** proteusguy has quit IRC | 11:50 | |
| *** proteusguy has joined #openstack-swift | 11:50 | |
| *** proteusguy has quit IRC | 11:53 | |
| *** proteusguy has joined #openstack-swift | 11:54 | |
| *** dtantsur has joined #openstack-swift | 11:55 | |
| dtantsur | hi swift gurus :) quick question: is it possible to upload object using one Keystone user+tenenat, so that it's visible in another user+tenant? | 11:56 |
| *** proteusguy has quit IRC | 11:57 | |
| *** proteusguy has joined #openstack-swift | 11:57 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 11:59 | |
| *** NM has quit IRC | 12:01 | |
| *** trown has joined #openstack-swift | 12:07 | |
| *** NM has joined #openstack-swift | 12:09 | |
| *** proteusguy has quit IRC | 12:09 | |
| *** proteusguy has joined #openstack-swift | 12:10 | |
| *** acoles has quit IRC | 12:10 | |
| *** acoles has joined #openstack-swift | 12:12 | |
| *** ChanServ sets mode: +v acoles | 12:12 | |
| *** proteusguy has quit IRC | 12:12 | |
| *** proteusguy has joined #openstack-swift | 12:13 | |
| *** proteusguy has quit IRC | 12:16 | |
| *** bill_az has joined #openstack-swift | 12:17 | |
| *** proteusguy has joined #openstack-swift | 12:17 | |
| *** joeljwright has joined #openstack-swift | 12:21 | |
| *** ChanServ sets mode: +v joeljwright | 12:21 | |
| *** joeljwright has quit IRC | 12:30 | |
| *** benny__ has joined #openstack-swift | 12:30 | |
| *** CaioBrentano has quit IRC | 12:32 | |
| *** CaioBrentano has joined #openstack-swift | 12:33 | |
| *** annegentle has joined #openstack-swift | 12:34 | |
| *** CaioBren_ has joined #openstack-swift | 12:35 | |
| *** vinsh_ has quit IRC | 12:36 | |
| *** CaioBrentano has quit IRC | 12:37 | |
| *** jasondotstar has joined #openstack-swift | 12:37 | |
| *** joeljwright has joined #openstack-swift | 12:44 | |
| *** ChanServ sets mode: +v joeljwright | 12:44 | |
| *** NM has quit IRC | 12:44 | |
| *** amoturi has quit IRC | 12:53 | |
| *** amoturi has joined #openstack-swift | 12:53 | |
| *** NM has joined #openstack-swift | 12:56 | |
| *** hrou has joined #openstack-swift | 13:00 | |
| *** mutoulbj has quit IRC | 13:04 | |
| *** petertr7_away is now known as petertr7 | 13:06 | |
| *** SkyRocknRoll_ has quit IRC | 13:07 | |
| *** vinsh has joined #openstack-swift | 13:08 | |
| *** jkugel has joined #openstack-swift | 13:16 | |
| *** thumpba has joined #openstack-swift | 13:25 | |
| *** lastops has joined #openstack-swift | 13:26 | |
| *** km has quit IRC | 13:27 | |
| *** vinsh has quit IRC | 13:32 | |
| *** kei_yama has quit IRC | 13:33 | |
| *** ppai has quit IRC | 13:33 | |
| *** mutoulbj has joined #openstack-swift | 13:35 | |
| *** petertr7 is now known as petertr7_away | 13:47 | |
| *** petertr7_away is now known as petertr7 | 13:53 | |
| *** fifieldt has joined #openstack-swift | 13:53 | |
| *** vinsh has joined #openstack-swift | 13:57 | |
| *** vinsh_ has joined #openstack-swift | 13:58 | |
| *** amoturi_ has joined #openstack-swift | 14:00 | |
| *** vinsh has quit IRC | 14:02 | |
| *** amoturi has quit IRC | 14:04 | |
| *** annegentle has quit IRC | 14:04 | |
| *** lcurtis has joined #openstack-swift | 14:04 | |
| *** blmartin has joined #openstack-swift | 14:08 | |
| *** NM has quit IRC | 14:10 | |
| blmartin | good morning | 14:12 |
| *** acampbel11 has joined #openstack-swift | 14:12 | |
| *** adiantum has joined #openstack-swift | 14:18 | |
| *** NM has joined #openstack-swift | 14:19 | |
| *** ChanServ sets mode: +v tdasilva | 14:19 | |
| *** NM has quit IRC | 14:19 | |
| *** aix has quit IRC | 14:21 | |
| *** jrichli has joined #openstack-swift | 14:24 | |
| *** NM has joined #openstack-swift | 14:29 | |
| *** CaioBren_ has quit IRC | 14:32 | |
| *** CaioBrentano has joined #openstack-swift | 14:32 | |
| openstackgerrit | Bogun Dmitriy proposed openstack/swift: FIX http header size constraint https://review.openstack.org/197764 | 14:33 |
| *** dtantsur has left #openstack-swift | 14:40 | |
| *** annegentle has joined #openstack-swift | 14:41 | |
| *** NM has quit IRC | 14:42 | |
| *** NM has joined #openstack-swift | 14:45 | |
| *** joeljwright has quit IRC | 14:49 | |
| *** mutoulbj has quit IRC | 14:51 | |
| *** mutoulbj has joined #openstack-swift | 14:52 | |
| *** petertr7 is now known as petertr7_away | 14:52 | |
| *** petertr7_away is now known as petertr7 | 14:58 | |
| lcurtis | hello all...i am using separate replication network for swift within a single region. I am planning out new region...does inter-region replication take place over the 'replication' network? | 15:06 |
| *** amoturi_ has quit IRC | 15:14 | |
| openstackgerrit | Samuel Merritt proposed openstack/swift: Stop moving partitions unnecessarily when overload is on. https://review.openstack.org/197396 | 15:16 |
| openstackgerrit | Samuel Merritt proposed openstack/swift: Add ring-builder analyzer. https://review.openstack.org/197395 | 15:16 |
| *** benny__ has quit IRC | 15:16 | |
| *** pbourke has left #openstack-swift | 15:18 | |
| *** DericHorn-HP has joined #openstack-swift | 15:18 | |
| *** nadeem has joined #openstack-swift | 15:18 | |
| *** foexle has quit IRC | 15:20 | |
| *** chlong has quit IRC | 15:23 | |
| *** NM has quit IRC | 15:25 | |
| *** hrou has quit IRC | 15:26 | |
| *** nadeem has quit IRC | 15:31 | |
| *** nadeem has joined #openstack-swift | 15:31 | |
| *** benny__ has joined #openstack-swift | 15:32 | |
| *** nexusz99 has quit IRC | 15:34 | |
| *** janonymous has joined #openstack-swift | 15:35 | |
| *** proteusguy has quit IRC | 15:45 | |
| *** proteusguy has joined #openstack-swift | 15:45 | |
| *** proteusguy has quit IRC | 15:52 | |
| *** proteusguy has joined #openstack-swift | 15:53 | |
| *** proteusguy has quit IRC | 15:55 | |
| *** SkyRocknRoll has quit IRC | 15:55 | |
| *** jistr has quit IRC | 16:02 | |
| *** joeljwright has joined #openstack-swift | 16:05 | |
| *** ChanServ sets mode: +v joeljwright | 16:05 | |
| *** proteusguy has joined #openstack-swift | 16:08 | |
| *** benny__ has quit IRC | 16:11 | |
| *** proteusguy has quit IRC | 16:11 | |
| *** proteusguy has joined #openstack-swift | 16:12 | |
| *** proteusguy has quit IRC | 16:14 | |
| *** proteusguy has joined #openstack-swift | 16:15 | |
| *** bill_az has quit IRC | 16:15 | |
| *** proteusguy has quit IRC | 16:17 | |
| *** proteusguy has joined #openstack-swift | 16:17 | |
| *** jordanP has quit IRC | 16:19 | |
| *** silor has quit IRC | 16:22 | |
| *** benny__ has joined #openstack-swift | 16:24 | |
| *** petertr7 is now known as petertr7_away | 16:25 | |
| *** SkyRocknRoll has joined #openstack-swift | 16:26 | |
| *** thurloat_isgone is now known as thurloat | 16:27 | |
| *** rledisez has quit IRC | 16:30 | |
| *** zhill has joined #openstack-swift | 16:30 | |
| *** proteusguy has quit IRC | 16:33 | |
| notmyname | lcurtis: yes | 16:33 |
| *** proteusguy has joined #openstack-swift | 16:34 | |
| lcurtis | hey notmyname! Thanks a million | 16:34 |
| *** trown has left #openstack-swift | 16:35 | |
| *** janonymous_ has joined #openstack-swift | 16:45 | |
| *** cazino has left #openstack-swift | 16:49 | |
| *** nadeem has quit IRC | 16:49 | |
| *** proteusguy has quit IRC | 16:49 | |
| *** proteusguy has joined #openstack-swift | 16:50 | |
| openstackgerrit | John Dickinson proposed openstack/swift: update AUTHORS file https://review.openstack.org/198064 | 16:54 |
| *** joeljwright has quit IRC | 16:57 | |
| janonymous_ | +2 :P :) | 16:57 |
| notmyname | I figured it had been a while since I updated it :-) | 16:59 |
| notmyname | janonymous_: is that email ok for you? I can use the other one if you want? | 16:59 |
| *** gyee has joined #openstack-swift | 17:01 | |
| janonymous_ | Yeah it's ok too :) Thanks . | 17:01 |
| *** blmartin has quit IRC | 17:03 | |
| *** blmartin has joined #openstack-swift | 17:03 | |
| *** proteusguy has quit IRC | 17:06 | |
| *** proteusguy has joined #openstack-swift | 17:08 | |
| *** petertr7_away is now known as petertr7 | 17:17 | |
| *** proteusguy has quit IRC | 17:18 | |
| *** jasondotstar is now known as jasondotstar|afk | 17:19 | |
| *** proteusguy has joined #openstack-swift | 17:21 | |
| *** hseipp has quit IRC | 17:25 | |
| *** petertr7 is now known as petertr7_away | 17:27 | |
| *** proteusguy has quit IRC | 17:27 | |
| *** petertr7_away is now known as petertr7 | 17:27 | |
| *** NM has joined #openstack-swift | 17:28 | |
| *** acampbel11 has quit IRC | 17:36 | |
| *** jasondotstar|afk has quit IRC | 17:45 | |
| openstackgerrit | John Dickinson proposed openstack/swift: update AUTHORS file https://review.openstack.org/198064 | 17:45 |
| *** hrou has joined #openstack-swift | 17:53 | |
| *** acampbel11 has joined #openstack-swift | 17:55 | |
| *** thumpba has quit IRC | 17:58 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: Centralize header parsing https://review.openstack.org/184336 | 17:58 |
| timburke | notmyname: yay! i feel special now :) | 18:00 |
| *** logan2 has quit IRC | 18:03 | |
| *** acoles is now known as acoles_away | 18:07 | |
| *** nadeem has joined #openstack-swift | 18:13 | |
| *** petertr7 is now known as petertr7_away | 18:17 | |
| *** petertr7_away is now known as petertr7 | 18:23 | |
| *** mutoulbj has quit IRC | 18:24 | |
| *** wuhg has quit IRC | 18:34 | |
| *** geaaru has quit IRC | 18:35 | |
| clayg | redbo is right - if this jury selection process keeps up much longer I *am* going to have to jump up and whiteboard it for these people - LOOK - it's not that hard | 18:40 |
| clayg | bah zaitcev isn't in here - but I disagree about it being inherintly racy - the mount_check was not there to prevent writing to root - a LBYL *is* always racy - but that's not what it is about | 18:44 |
| openstackgerrit | janonymous proposed openstack/swift: Fix Python 3 issues https://review.openstack.org/198088 | 18:45 |
| clayg | we could just not create a dir at /srv/node/<device> if it doesn't exist and all would be fine | 18:45 |
| clayg | s/if it doesn't exist// - I mean just don't do it, if it's not there then os.mkdir('/srv/node/<device>/objects[-N] would fail enoent - and that's when we stop | 18:47 |
| *** nadeem has quit IRC | 18:47 | |
| *** logan2 has joined #openstack-swift | 18:50 | |
| *** nadeem has joined #openstack-swift | 18:51 | |
| *** SkyRocknRoll has quit IRC | 18:53 | |
| *** blmartin has quit IRC | 18:54 | |
| *** minwoob has quit IRC | 18:55 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient: Updated from global requirements https://review.openstack.org/89250 | 18:57 |
| *** SkyRocknRoll has joined #openstack-swift | 19:07 | |
| *** marcusvrn_ has joined #openstack-swift | 19:14 | |
| *** marcusvrn_ is now known as marcusvrn | 19:17 | |
| *** petertr7 is now known as petertr7_away | 19:17 | |
| *** petertr7_away is now known as petertr7 | 19:20 | |
| clayg | nm, i take it all back - torgomatic's right - rsync will create device dirs - so you can prevent it in our code - needs to be enforced by the os | 19:22 |
| clayg | or rather letting the os enforce it is simplist - doing some sort of shim about the rsync glibc calls - or maybe moving to per device rysnc modules - or ssync - would all be other things you could do | 19:23 |
| *** rbrooker has joined #openstack-swift | 19:25 | |
| *** janonymous_ has quit IRC | 19:30 | |
| openstackgerrit | janonymous proposed openstack/swift: Replace xrange() with six.moves.range() https://review.openstack.org/198102 | 19:31 |
| *** delatte has joined #openstack-swift | 19:34 | |
| *** janonymous_ has joined #openstack-swift | 19:35 | |
| *** delattec has quit IRC | 19:37 | |
| *** silor has joined #openstack-swift | 19:38 | |
| *** CaioBren_ has joined #openstack-swift | 19:43 | |
| *** CaioBrentano has quit IRC | 19:47 | |
| *** rvasilets_ has joined #openstack-swift | 19:47 | |
| *** igor___ has joined #openstack-swift | 19:48 | |
| *** jasondotstar|afk has joined #openstack-swift | 19:49 | |
| *** DericHorn-HP has quit IRC | 19:53 | |
| *** silor has quit IRC | 20:00 | |
| *** lastops has quit IRC | 20:14 | |
| *** acoles_away has quit IRC | 20:15 | |
| *** acoles_away has joined #openstack-swift | 20:20 | |
| *** acoles_away is now known as acoles | 20:20 | |
| *** ChanServ sets mode: +v acoles | 20:20 | |
| *** jasondotstar|afk has quit IRC | 20:22 | |
| *** acampbel11 has quit IRC | 20:34 | |
| *** jasondotstar|afk has joined #openstack-swift | 20:34 | |
| *** petertr7 is now known as petertr7_away | 20:47 | |
| *** petertr7_away is now known as petertr7 | 20:52 | |
| *** DericHorn-HP has joined #openstack-swift | 20:56 | |
| *** petertr7 is now known as petertr7_away | 21:08 | |
| *** DericHorn-HP has quit IRC | 21:08 | |
| *** annegentle has quit IRC | 21:08 | |
| *** jasondotstar|afk has quit IRC | 21:17 | |
| *** adiantum has quit IRC | 21:18 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: Prompt for missing username and/or password https://review.openstack.org/189387 | 21:26 |
| *** jrichli has quit IRC | 21:29 | |
| *** logan2 has quit IRC | 21:31 | |
| *** DericHorn-HP has joined #openstack-swift | 21:33 | |
| *** lpabon has quit IRC | 21:36 | |
| *** CaioBren_ has quit IRC | 21:38 | |
| *** CaioBrentano has joined #openstack-swift | 21:38 | |
| *** DericHorn-HP has quit IRC | 21:40 | |
| *** CaioBrentano has quit IRC | 21:44 | |
| *** lastops has joined #openstack-swift | 21:44 | |
| *** CaioBrentano has joined #openstack-swift | 21:44 | |
| *** NM has quit IRC | 21:47 | |
| *** CaioBrentano has quit IRC | 21:47 | |
| *** CaioBrentano has joined #openstack-swift | 21:47 | |
| *** lastops has quit IRC | 21:48 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient: Add bulkdelete command https://review.openstack.org/190887 | 21:49 |
| *** thurloat is now known as thurloat_isgone | 21:52 | |
| *** rvasilets_ has quit IRC | 21:54 | |
| *** logan2 has joined #openstack-swift | 21:56 | |
| *** jkugel has left #openstack-swift | 21:58 | |
| *** marzif_ has joined #openstack-swift | 21:58 | |
| *** ianbrown has joined #openstack-swift | 22:01 | |
| *** annegentle has joined #openstack-swift | 22:07 | |
| *** marzif_ has quit IRC | 22:10 | |
| openstackgerrit | Victor Stinner proposed openstack/swift: Get StringIO and cStringIO from six.moves https://review.openstack.org/185457 | 22:13 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Replace StringIO with BytesIO in ssync https://review.openstack.org/186073 | 22:13 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Replace StringIO with BytesIO for file https://review.openstack.org/186072 | 22:13 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Replace StringIO with BytesIO for WSGI input https://review.openstack.org/186071 | 22:13 |
| *** chlong has joined #openstack-swift | 22:18 | |
| *** DericHorn-HP has joined #openstack-swift | 22:24 | |
| *** annegentle has quit IRC | 22:24 | |
| *** annegentle has joined #openstack-swift | 22:24 | |
| mattoliverau | Morning | 22:28 |
| *** annegentle has quit IRC | 22:30 | |
| notmyname | hello mattoliverau | 22:30 |
| mattoliverau | Looks like its been a quiet night in channel | 22:31 |
| tdasilva | hi matt | 22:32 |
| notmyname | mattoliverau: yeah, not too much being talked about in here | 22:32 |
| notmyname | in my continuing quest to measure stuff in the community, I've derived a new tool | 22:32 |
| notmyname | I can now find people who have contributed to swift more than once, and done so in the last 6 months, but haven't in the last 30 days | 22:33 |
| notmyname | gives me a lit of people that might be at risk of stop contributing to the project | 22:33 |
| mattoliverau | tdasilva: hey man | 22:33 |
| notmyname | (obviously, the exact numbers are configurable) | 22:33 |
| tdasilva | notmyname: that sounds cool | 22:34 |
| tdasilva | notmyname: i get the impression that there's a lot of new folks since the summit, which is great | 22:34 |
| notmyname | for example, both tdasilva and mattoliverau haven't had a patch land in the last 30 days ;-) | 22:34 |
| tdasilva | really? i thought i had one last week | 22:35 |
| mattoliverau | notmyname: I've had a specs one land | 22:35 |
| mattoliverau | And a co-authored one | 22:35 |
| notmyname | I can't really track co-authoreds yet | 22:35 |
| hrou | notmyname: Fairly new to the game but I hope that soon enough I'll on that list (well actually not on it given the criteria ; - ) | 22:36 |
| notmyname | and this is just in the swift repo, not swift-specs. later I'd love to aggregate data across repos | 22:36 |
| notmyname | tdasilva: last one one was on 5/13 | 22:36 |
| notmyname | hrou: heh :-) | 22:36 |
| tdasilva | notmyname: ok, so you are tracking when they were submitted | 22:36 |
| tdasilva | not merged | 22:36 |
| *** briancurtin has quit IRC | 22:36 | |
| notmyname | the date that's on the commit | 22:37 |
| notmyname | basically, I'm just parsing the git log (which means that this tool isn't swift-specific) | 22:37 |
| tdasilva | patch 182826 | 22:37 |
| patchbot | tdasilva: https://review.openstack.org/#/c/182826/ | 22:37 |
| *** DericHorn-HP has quit IRC | 22:37 | |
| tdasilva | notmyname: got it | 22:37 |
| tdasilva | notmyname: are you still making the graph of total contributors in the last so many days? | 22:38 |
| notmyname | yeah. this is all the same tool | 22:38 |
| tdasilva | neat | 22:38 |
| notmyname | actually, that one that you mention is the very next one I'm going to try to improve | 22:38 |
| *** briancurtin has joined #openstack-swift | 22:39 | |
| mattoliverau | Well hopefully one day most my dev effort will be in swifts main repo not in my sharding github repo. Maybe that can be once I have finished gathering numbers to see if the current sharding implementation is worthwhile :) | 22:40 |
| *** amit213 has quit IRC | 22:42 | |
| *** SkyRocknRoll has quit IRC | 22:43 | |
| *** tdasilva has quit IRC | 22:43 | |
| *** lcurtis has quit IRC | 22:44 | |
| *** SkyRocknRoll has joined #openstack-swift | 22:45 | |
| *** hrou has quit IRC | 22:47 | |
| *** tdasilva has joined #openstack-swift | 22:47 | |
| notmyname | tdasilva: here it is, now with a 90-day rolling average) http://d.not.mn/active_contribs.png | 22:47 |
| tdasilva | rolling average??? man, pretty soon you are going to have 10 and 20 day EMA there :) | 22:48 |
| tdasilva | notmyname: that's pretty cool | 22:49 |
| notmyname | http://d.not.mn/contrib_activity.png is my current favorite. the bars are the entirety of time someone has contributed no matter what happens in the middle | 22:50 |
| tdasilva | notmyname: how are you looking for the tools to help you besides giving you the data? what actionable item can you take? | 22:50 |
| notmyname | great question :-) | 22:51 |
| notmyname | so the first thing I have is a picture of where we are today | 22:51 |
| notmyname | the next thing is the list I mentioned. I'll start email the people who haven't landed a patch recently but have done so in the past | 22:51 |
| notmyname | and there are a few other things that I don't know yet. eg theres some interesting inflection points on the active contribs graph, and I'd like to find what caused them | 22:52 |
| notmyname | overall, my goal is to grow the community of contributors and keep it healthy | 22:53 |
| notmyname | if you've got ideas on what to look for or what to measure, I'd love to hear it :-) | 22:53 |
| tdasilva | notmyname: funny you mention the inflection points cause I was just looking at them | 22:54 |
| tdasilva | are they related to the releases? | 22:54 |
| tdasilva | like there's a big spike up and down every so often | 22:55 |
| notmyname | on the active contribs graph? | 22:55 |
| tdasilva | it seems like the average # of contribs have been pretty constant in the last 1000 | 22:56 |
| tdasilva | yeah | 22:56 |
| tdasilva | notmyname: eg. in the last 500 days there were two spikes to almost 20 | 22:56 |
| tdasilva | but then it goes way down | 22:57 |
| notmyname | http://d.not.mn/active_contribs_30.png <-- a little smoother. this one has an "active window" of 30 days. the other has it at 14 days | 22:57 |
| notmyname | ok, let me show you one more and then answer that (i think I know the answer) | 22:57 |
| *** km has joined #openstack-swift | 22:58 | |
| openstackgerrit | Thiago da Silva proposed openstack/swift: versioned writes middleware https://review.openstack.org/134347 | 22:59 |
| notmyname | http://d.not.mn/contrib_deltas_30.png | 23:00 |
| notmyname | that's deltas in active contribs and total contribs for the past year | 23:00 |
| notmyname | note the spike up about 90 days ago and the spike down abot 50 days ago | 23:00 |
| notmyname | that was the EC branch merge | 23:00 |
| notmyname | I suspect the other spike you saw on the long-erterm graph was the storage policy merge | 23:01 |
| *** ndk has quit IRC | 23:01 | |
| notmyname | tdasilva: thanks for the rebase | 23:01 |
| *** portante has quit IRC | 23:01 | |
| tdasilva | notmyname: yw | 23:01 |
| * tdasilva is confused about the spike down about 50 days ago | 23:03 | |
| *** portante has joined #openstack-swift | 23:03 | |
| *** ndk has joined #openstack-swift | 23:04 | |
| notmyname | I suspect that's people who were marked as active with the EC merge patch and then 30 days later we're marked as inactive because they hadn't landed a commit in that time | 23:04 |
| notmyname | looks like the spike up may have been ~80 days ago (not 90 | 23:04 |
| tdasilva | ok | 23:05 |
| tdasilva | makes sense | 23:05 |
| *** benny__ has quit IRC | 23:09 | |
| notmyname | yup. I added grid markers (locally). it was about 84 days ago | 23:09 |
| tdasilva | april 9th, that sounds about right! | 23:14 |
| tdasilva | i went on leave on the 13th | 23:14 |
| tdasilva | notmyname: gotta run! have a great weekend! | 23:16 |
| *** sorrison_laptop has joined #openstack-swift | 23:19 | |
| notmyname | you too | 23:20 |
| mattoliverau | tdasilva: enjoy your weekend o/ | 23:21 |
| *** hrou has joined #openstack-swift | 23:23 | |
| *** kei_yama has joined #openstack-swift | 23:29 | |
| *** zaitcev has joined #openstack-swift | 23:30 | |
| *** ChanServ sets mode: +v zaitcev | 23:30 | |
| zaitcev | Guys, did anyone see gate failing in test_PUT_ssync_multi_frag with "got 201 != 409 for frag_index=3 node_index=3"? | 23:32 |
| * notmyname did not | 23:33 | |
| *** StevenK has quit IRC | 23:35 | |
| *** chlong has quit IRC | 23:36 | |
| *** nadeem has quit IRC | 23:39 | |
| *** nadeem has joined #openstack-swift | 23:40 | |
| *** ho has joined #openstack-swift | 23:45 | |
| ho | good morning! | 23:50 |
| *** zhill has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
