| *** cppforlife_ has quit IRC | 00:00 | |
| *** spotz has quit IRC | 00:00 | |
| *** amit213 has quit IRC | 00:01 | |
| *** zacksh has quit IRC | 00:02 | |
| *** nikhil|afk has quit IRC | 00:02 | |
| *** nikhil|afk has joined #openstack-swift | 00:03 | |
| *** amit213 has joined #openstack-swift | 00:06 | |
| *** zacksh has joined #openstack-swift | 00:09 | |
| *** spotz has joined #openstack-swift | 00:09 | |
| clayg | tdasilva: timburke: kota_: any ideas? https://bugs.launchpad.net/pyeclib/+bug/1668447 | 00:09 |
|---|---|---|
| openstack | Launchpad bug 1668447 in PyECLib "python3 tests fail" [Undecided,New] | 00:09 |
| *** sams-gleb has joined #openstack-swift | 00:13 | |
| *** dja has quit IRC | 00:15 | |
| *** darrenc has quit IRC | 00:16 | |
| *** jamielennox|away is now known as jamielennox | 00:16 | |
| *** darrenc has joined #openstack-swift | 00:17 | |
| *** sams-gleb has quit IRC | 00:17 | |
| *** dja has joined #openstack-swift | 00:18 | |
| *** rledisez has quit IRC | 00:18 | |
| *** rledisez has joined #openstack-swift | 00:18 | |
| timburke | clayg: your py27 run created a pyeclib_c.so in the working tree, yeah? try deleting it then running py34 (or w/e) again | 00:31 |
| timburke | that's what seemed to do it for me... master still passes on my machine | 00:32 |
| clayg | timburke: brilliant! | 00:33 |
| *** gatuus has quit IRC | 00:51 | |
| clayg | timburke: so just a simple `tox -re py34` isn't enough because I guess the .so get's built not in the tox env? | 00:58 |
| clayg | timburke: is something funky going on with our paths? .tox/py34/lib/python3.4/site-packages/pyeclib_c.cpython-34m.so | 01:01 |
| clayg | timburke: well you were right, deleting the .so made the tox -e py34 pass | 01:05 |
| *** zhugaoxiao has quit IRC | 01:06 | |
| *** Administrator_ has quit IRC | 01:06 | |
| * tdasilva is back | 01:06 | |
| clayg | gotta be directory prefernces kinda thing? | 01:06 |
| kota_ | godd morning | 01:14 |
| kota_ | clayg, timburke: i also suppose kevin and tushar had py3 compatible work in those days | 01:15 |
| clayg | kota_: timburke: tdasilva: this is so opaque - I have no idea if this is even close? https://review.openstack.org/#/c/438756/ | 01:16 |
| kota_ | clayg, timburke: and nice to have py3 gate job to avoid regression, let me look at what's going on the test | 01:16 |
| patchbot_ | patch 438756 - openstack-infra/project-config - Add experimental python34 job to pyeclib | 01:16 |
| kota_ | clayg: what's "experimental"? is that for non-voting? | 01:17 |
| clayg | kota_: I was just reading https://docs.openstack.org/infra/manual/creators.html#zuul-best-practices and following tdasilva's lead | 01:18 |
| clayg | it's definately non-voting | 01:18 |
| kota_ | ok, got it | 01:18 |
| clayg | i see absolutely NO reason to add a job to voting until we can see if it works? | 01:18 |
| clayg | but I think maybe I have to land something before I can see if it works... i have no idea | 01:18 |
| clayg | i just thought it was good first step to get my feet wet with project-config | 01:18 |
| tdasilva | clayg: need to update one more file IIRC | 01:20 |
| clayg | tdasilva: i was hoping using the pre-existing job made that not needed? | 01:21 |
| tdasilva | clayg: yeah, i think you are right: - python-jobs should take care of it in projects.yaml ??? | 01:22 |
| tdasilva | i agree it is very opaque | 01:22 |
| tdasilva | lots of trial and error | 01:22 |
| openstackgerrit | Thiago da Silva proposed openstack/swift master: use correct py3 rpm package in bindep https://review.openstack.org/438758 | 01:25 |
| tdasilva | clayg: patch 436642 failed because of above ^^^ | 01:25 |
| patchbot_ | https://review.openstack.org/#/c/436642/ - swift - WIP: centos func test | 01:25 |
| clayg | tdasilva: i don't really understand who uses bindep and to what ends? | 01:25 |
| clayg | 2017-02-27 23:29:02.283980 | No package python3-devel available. | 01:26 |
| tdasilva | right, it is called python34-devel | 01:26 |
| clayg | so apparently something in the gate/node magic somewhere expects bindep to do something... | 01:26 |
| tdasilva | now here is the fun part: in centos liberasurecode-devel is packaged with rdo packages | 01:26 |
| tdasilva | so we need to install rdo repo :( | 01:26 |
| clayg | ^ zaitcev sound good to me! | 01:27 |
| clayg | tdasilva: so can you do the "experimental" on patch 438758 and see if there's any other issues? | 01:28 |
| patchbot_ | https://review.openstack.org/#/c/438758/ - swift - use correct py3 rpm package in bindep | 01:28 |
| tdasilva | the problem is that rdo repos are per release. eg: centos-release-openstack-newton | 01:28 |
| tdasilva | yeah, i think i'm going to update it first | 01:28 |
| clayg | tdasilva: even for dependencies? | 01:28 |
| mattoliverau | kota_: morning | 01:29 |
| tdasilva | clayg: i believe so | 01:29 |
| kota_ | mattoliverau: o/ | 01:29 |
| kota_ | mattoliverau: is everything fine? | 01:30 |
| * kota_ is thinking mattoliverau had a long trip from Atlanta | 01:30 | |
| clayg | tdasilva: so I'm not sure the py34 tox env will still run in the gate? | 01:33 |
| clayg | # Restrict python34 testing to <= Mitaka. | 01:33 |
| clayg | - name: ^gate-.*-python34(-db)?(-nv)?$ | 01:33 |
| clayg | branch: ^stable/(liberty|mitaka) | 01:33 |
| clayg | # Restrict python35 testing to >= Newton. | 01:33 |
| clayg | - name: ^gate-.*-python35(-db)?(-nv)?$ | 01:33 |
| clayg | branch: ^(?!stable/(liberty|mitaka)).*$ | 01:33 |
| tdasilva | lol | 01:33 |
| tdasilva | great! | 01:33 |
| clayg | py34 was so last year | 01:33 |
| tdasilva | probably should remove the line then??? | 01:34 |
| tdasilva | and isn't that a problem for the whole TC Goals thing? | 01:35 |
| tdasilva | this is so weird | 01:35 |
| kota_ | that means, we need py35 gate and bindep should follow that? | 01:36 |
| kota_ | i mean patch 438758 | 01:36 |
| patchbot_ | https://review.openstack.org/#/c/438758/ - swift - use correct py3 rpm package in bindep | 01:36 |
| clayg | who knows? | 01:36 |
| kota_ | idk | 01:36 |
| tdasilva | i'm trying to figure out if anyone is running py3 tests on centos | 01:37 |
| clayg | why does swift have a bindep on python3<anything>? | 01:37 |
| clayg | tdasilva: but you're not *trying* to run py3 tests? you're just trying to run on centos? | 01:37 |
| tdasilva | clayg: because we want to test again py3 right? don't we have unit tests already in py3? | 01:37 |
| tdasilva | clayg: correct! but i think running bindep is part of the gates setup script ??? | 01:37 |
| mattoliverau | kota_: yeah it was long, but back at work now.. now just a little jet lagged the other way round :P | 01:37 |
| mattoliverau | you? | 01:38 |
| * mattoliverau stops for some lunch | 01:38 | |
| kota_ | clayg, tdasilva, mattoliverau: btw, could you guys review https://review.openstack.org/#/q/Ib90adcc2f059adaf203fba1c95b2154561ea7487,n,z? | 01:38 |
| kota_ | if you have time to look at | 01:38 |
| tdasilva | kota_: starred, but probably won't be able to look at it today :( | 01:39 |
| kota_ | that is needed to close... | 01:39 |
| kota_ | timburke: thx | 01:39 |
| clayg | tdasilva: yeah you're right - we have gate-swift-python35 | 01:39 |
| kota_ | tdasilva: ah, i ping timburke incorrectly, thanks for starring. | 01:41 |
| tdasilva | does anyone know what is the file we are supposed to update talking about py3 goals | 01:43 |
| tdasilva | would like to have a look to see if I can learn if there's a project doing py3 on centos testing | 01:44 |
| *** sams-gleb has joined #openstack-swift | 01:45 | |
| *** sams-gleb has quit IRC | 01:50 | |
| clayg | tdasilva: https://governance.openstack.org/tc/goals/pike/python35.html ? | 01:51 |
| *** cppforlife_ has joined #openstack-swift | 02:00 | |
| *** disaster has joined #openstack-swift | 02:01 | |
| disaster | Hi! Is there a way in openstack swift to set an user as sendonly? With ACL or something | 02:02 |
| clayg | disaster: you should be able to drop the --read-acl from this example: https://docs.openstack.org/developer/swift/overview_acl.html#example-shared-writable-container | 02:03 |
| clayg | and get a write-only container | 02:03 |
| clayg | I think most folks try to leverage PUT tempurls for that use-case, YMMV | 02:04 |
| tdasilva | clayg: given that paste you shared earlier, should you have used py35 in patch 438756 ? | 02:05 |
| patchbot_ | https://review.openstack.org/#/c/438756/ - openstack-infra/project-config - Add experimental python34 job to pyeclib | 02:05 |
| clayg | I think I need to add a py35 env to pyeclib tox.ini but yes | 02:05 |
| disaster | clayg: and write wont allow delete? | 02:05 |
| clayg | disaster: PUT tempurl would be better then | 02:06 |
| clayg | disaster: PUT can always destory data - you can over-write? | 02:06 |
| openstackgerrit | Thiago da Silva proposed openstack/swift master: add rdo repo to install liberasurecode-devel https://review.openstack.org/438758 | 02:07 |
| clayg | tdasilva: ^ brilliant | 02:08 |
| tdasilva | clayg: really ugly IMO :/ | 02:08 |
| clayg | well see if it works first | 02:08 |
| tdasilva | clayg: sad if we have to keep updating this line every release: centos-release-openstack-newton | 02:09 |
| disaster | clayg: arf my use case is cronjob backup but in case of hack i would like the server don't store the full access to the object storage | 02:09 |
| clayg | disaster: container level tempurls are great for that use-case | 02:10 |
| disaster | clayg: ok but how my cronjob can have the tempurl? | 02:10 |
| clayg | create a container server-xyz-backups and put a tempurl on it - then from the cronjob just `swift tempurl` with the right bits and upload with curl | 02:11 |
| clayg | maybe tricker if you also need segmented uploads | 02:11 |
| clayg | disaster: are you using keystone? | 02:11 |
| disaster | clayg: in fact it's on the ovh object storage | 02:12 |
| disaster | clayg: so i don't really know | 02:12 |
| *** klrmn has quit IRC | 02:12 | |
| tdasilva | ovh is using keystone | 02:13 |
| clayg | disaster: ok, well if it's public cloud you might ask their support what's supported - a non-admin account with a write-acl container is probably near the best solution | 02:14 |
| disaster | clayg: ok and the account use from the server will not be able to delete files? | 02:16 |
| clayg | disaster: you need to ask you're service provier what they support - a write-acl even without listing could still allow for delete/overwrite - but limited to the container | 02:17 |
| disaster | clayg: ok thank you | 02:19 |
| tdasilva | disaster: just note what clayg said about over-writing objects | 02:19 |
| openstackgerrit | Clay Gerrard proposed openstack/pyeclib master: Add more isolation to tests https://review.openstack.org/438770 | 02:24 |
| disaster | tdasilva: yes i will ask the support, my only other option is to take a server for backup on chroot ssl and rotate the chroot directory | 02:25 |
| disaster | tdasilva: or maybe you have a third way? | 02:25 |
| tdasilva | disaster: what i mean to say, is that clayg pointed out that even if you are able to find a way to get acls to only write data, and not delete the data, your server could still overwrite the same object with 0 bytes, effectively deleting it | 02:26 |
| tdasilva | disaster: unless i'm missing something | 02:26 |
| clayg | tdasilva: I think he mostly just doesn't want to put his keys-to-the-kingdom on a server just for backups | 02:27 |
| clayg | the use-case may be more restrictive than that - but I'd shoot for solving that risk model first and going from there | 02:27 |
| *** bkopilov has quit IRC | 02:27 | |
| tdasilva | oic | 02:28 |
| openstackgerrit | Clay Gerrard proposed openstack/pyeclib master: Don't let nose play with our PYTHONPATH. https://review.openstack.org/438770 | 02:30 |
| tdasilva | now i'm getting this: /tmp/04-cd800e142a044f52b054d7a1af3acebd.sh: line 5: mkfs.xfs: command not found | 02:31 |
| tdasilva | but i thought i'd get mkfs.xfs with xfsprogs | 02:31 |
| disaster | tdasilva: clayg: thank you for your help | 02:34 |
| *** disaster has quit IRC | 02:34 | |
| tdasilva | clayg: where is your mkfs.xfs installed ? /usr/sbin ? | 03:02 |
| *** rledisez has quit IRC | 03:06 | |
| *** rledisez has joined #openstack-swift | 03:08 | |
| *** adriant has joined #openstack-swift | 03:16 | |
| *** tqtran has quit IRC | 03:19 | |
| tdasilva | clayg: just got a link for https://trunk.rdoproject.org/centos7-master-head/report.html | 03:20 |
| tdasilva | still trying to understand what that is.... | 03:21 |
| tdasilva | just FYI... | 03:21 |
| openstackgerrit | Thiago da Silva proposed openstack/swift master: fix bindep file to enable func test in centos https://review.openstack.org/438758 | 03:23 |
| adriant | Just digging through the swiftclient code and trying to confirm something... Does the swiftclient handle expired tokens in any capacity? | 03:26 |
| adriant | As in, does it catch the error, and fetch a new token. | 03:26 |
| adriant | It looks like it sort of does in the _retrying function, but only once | 03:27 |
| adriant | What I'm trying to handle here is a case where a duplicity backup with Swift is taking over 12 hrs, and the token expiry is 12 hours. | 03:30 |
| adriant | The token is running out, and the backup falls over. | 03:30 |
| tdasilva | clayg: yay!! http://logs.openstack.org/58/438758/3/experimental/gate-swift-tox-xfs-tmp-func-centos-7/cd2467a/console.html | 03:31 |
| adriant | Most of the other clients handle this fine by using keystoneauth which handles fetching new tokens for you, but Swift doesn't seem to... | 03:31 |
| *** klrmn has joined #openstack-swift | 03:45 | |
| *** sams-gleb has joined #openstack-swift | 03:47 | |
| *** bkopilov has joined #openstack-swift | 03:48 | |
| *** cshastri has joined #openstack-swift | 03:50 | |
| *** sams-gleb has quit IRC | 03:51 | |
| mattoliverau | adriant: swiftclient now supports using a keystone auth context, and I also thought it retried. But timburke or joeljwright would be the best to answer this. If they or noone else can't chime in I can take a look at the code.. but am about to step out for an errand so might be gone for a bit. | 04:01 |
| adriant | mattoliverau, the weird part is that Swift seems to has responded with a 502, but on looking through the HAproxy logs I've matched that to a 401 from Keystone, so I'm not sure if the token expired and the error swiftclient got was silly... | 04:03 |
| openstackgerrit | Clay Gerrard proposed openstack/pyeclib master: Don't let nose play with our PYTHONPATH. https://review.openstack.org/438770 | 04:20 |
| *** psachin has joined #openstack-swift | 04:23 | |
| zaitcev | clayg: https://review.openstack.org/438793 - but I can almost guarantee you that this won't build, because it needs Source0 automatically found somehow. | 04:25 |
| patchbot_ | patch 438793 - rpm-packaging - Add PyECLib and liberasurecode | 04:25 |
| clayg | zaitcev: AWESOME! | 04:30 |
| zaitcev | I tried to cargo-cult what I could from pre-existing specs, mostly the {{ }} patterns. | 04:31 |
| clayg | zaitcev: *totally* - AFAIK all packages are cargo-cult's - srly let the gods please give us blessings from the sky! | 04:32 |
| *** klrmn has quit IRC | 04:32 | |
| clayg | why does gate-rpm-packaging-tox-lint-ubuntu-xenial sound weird to me? | 04:34 |
| zaitcev | Yeah, I don't see how lint(1) would help | 04:34 |
| zaitcev | In fact, it probably does not exist on Linux. | 04:35 |
| clayg | liberasurecode.spec.j2.suse should not use '%setup'. Please use '%autosetup' instead. | 04:35 |
| zaitcev | ok | 04:36 |
| zaitcev | What about the Source0 question | 04:36 |
| *** jrichli has quit IRC | 04:36 | |
| *** jrichli has joined #openstack-swift | 04:36 | |
| zaitcev | Also, autosetup comes from openstack_macros, right? | 04:36 |
| clayg | gate jobs say the darndest things | 04:36 |
| zaitcev | Oh. I thought you said it | 04:37 |
| clayg | zaitcev: there's got to be some magic somewhere that untars the upstream source tar balls in the right place to make something work | 04:37 |
| zaitcev | master-rpm-packaging-build-centos7 #2995: FAILURE // Reported result: SUCCESS | 04:38 |
| zaitcev | They do say darnest things! | 04:38 |
| zaitcev | I think Source0 has to be an http:// and then it all works. But which one? | 04:38 |
| *** ChanServ sets mode: +v jrichli | 04:39 | |
| clayg | i say we call this a huge success for the day - and tomorrow we'll try to get dirk to give us some advice | 04:39 |
| zaitcev | I have a bunch of meetings starting 07:30 tomorrow | 04:41 |
| clayg | well i hope the first one is where a bunch of people come in to say how awesome zaitcev is! I think that's what'll be my first meeting tomorrow. | 04:41 |
| clayg | g'night pete! | 04:42 |
| zaitcev | Later | 04:42 |
| kota_ | how awesome zaitcev is! good night! | 04:46 |
| zaitcev | oh no | 04:46 |
| zaitcev | You are the only one who managed to put your patch in, in the aftermath of PTG. | 04:47 |
| zaitcev | :-) | 04:47 |
| kota_ | :-) | 04:48 |
| *** Jeffrey4l__ has quit IRC | 04:49 | |
| openstackgerrit | Clay Gerrard proposed openstack/pyeclib master: Add py35 tox environment https://review.openstack.org/438799 | 04:51 |
| *** Jeffrey4l__ has joined #openstack-swift | 04:52 | |
| clayg | notmyname: i'm worried if i try to make https://wiki.openstack.org/wiki/Meetings/Swift setup for next meeting I'll ruin everything | 05:03 |
| *** rickflare has joined #openstack-swift | 05:07 | |
| *** adriant has quit IRC | 05:30 | |
| clayg | i'm still having trouble with that cannot switch to different thread someimes... even after patch 425495 | 05:39 |
| patchbot_ | https://review.openstack.org/#/c/425495/ - swift - Give spawn a cycle (MERGED) | 05:39 |
| *** sams-gleb has joined #openstack-swift | 05:49 | |
| *** zaitcev has quit IRC | 05:49 | |
| *** sams-gleb has quit IRC | 05:54 | |
| *** dmorita has quit IRC | 06:03 | |
| *** Jeffrey4l__ is now known as Jeffrey4l | 06:10 | |
| *** dja has quit IRC | 06:22 | |
| *** gk-1wm-su has joined #openstack-swift | 06:35 | |
| *** gk-1wm-su has left #openstack-swift | 06:35 | |
| *** sams-gleb has joined #openstack-swift | 07:02 | |
| *** tqtran has joined #openstack-swift | 07:03 | |
| *** tqtran has quit IRC | 07:07 | |
| *** rcernin has joined #openstack-swift | 07:11 | |
| *** ChubYann has quit IRC | 07:15 | |
| *** geaaru has joined #openstack-swift | 07:17 | |
| *** tesseract has joined #openstack-swift | 07:34 | |
| *** hseipp has joined #openstack-swift | 07:40 | |
| *** circ-user-6O5Su has joined #openstack-swift | 07:47 | |
| *** dja has joined #openstack-swift | 07:57 | |
| circ-user-6O5Su | I have a query, I have done a setup of openstack cloud with swift as a backend for glance in mitaka version.. From command line it is working fine, but in dashboard, when i click object-store→ container, it shows blank page.. Is there anything to be enabled in horizon for mitaka version?? My proxy-server.conf - http://paste.openstack.org/show/600718/ In swift log everything is normal except this, http://paste.openstack.org/show/60071 | 08:00 |
| circ-user-6O5Su | pls do guide | 08:00 |
| *** dmorita has joined #openstack-swift | 08:03 | |
| *** dmorita has quit IRC | 08:07 | |
| *** bob_cheesey has joined #openstack-swift | 08:11 | |
| *** gk-1wm-su has joined #openstack-swift | 08:14 | |
| *** gk-1wm-su has left #openstack-swift | 08:14 | |
| *** pcaruana has joined #openstack-swift | 08:17 | |
| *** klrmn has joined #openstack-swift | 08:38 | |
| *** jordanP has joined #openstack-swift | 08:39 | |
| *** klrmn has quit IRC | 08:49 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Small homework for ec duplciation https://review.openstack.org/438862 | 08:58 |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Small homework for ec duplciation https://review.openstack.org/438862 | 08:59 |
| *** Jeffrey4l has quit IRC | 09:00 | |
| *** Jeffrey4l has joined #openstack-swift | 09:00 | |
| *** csmart has quit IRC | 09:09 | |
| *** cbartz has joined #openstack-swift | 09:12 | |
| *** circ-user-6O5Su has quit IRC | 09:32 | |
| *** surf has joined #openstack-swift | 09:37 | |
| *** surf has quit IRC | 09:56 | |
| *** edausq has joined #openstack-swift | 09:57 | |
| *** bob_cheesey has quit IRC | 10:08 | |
| *** bob_cheesey has joined #openstack-swift | 10:09 | |
| *** csmart has joined #openstack-swift | 10:36 | |
| *** dfflanders has quit IRC | 10:37 | |
| *** jith has joined #openstack-swift | 10:39 | |
| *** jith has left #openstack-swift | 10:39 | |
| *** jith has joined #openstack-swift | 10:40 | |
| *** mvk has quit IRC | 10:42 | |
| jith | hi all, I have a query, I have done a setup of openstack cloud with swift as a backend for glance in mitaka version.. From command line it is working fine, but in dashboard, when i click object-store→ container, it shows blank page.. Is there anything to be enabled in horizon for mitaka version?? | 10:44 |
| jith | My proxy-server.conf - http://paste.openstack.org/show/600718/In swift log everything is normal except this, http://paste.openstack.org/show/600719/ | 10:44 |
| *** dja has quit IRC | 10:45 | |
| *** mvk has joined #openstack-swift | 11:14 | |
| *** sams-gleb has quit IRC | 11:31 | |
| *** sams-gleb has joined #openstack-swift | 11:31 | |
| *** sams-gleb has quit IRC | 11:36 | |
| *** sams-gleb has joined #openstack-swift | 11:51 | |
| *** cppforlife_ has quit IRC | 12:02 | |
| *** bkopilov has quit IRC | 12:07 | |
| *** fungi has quit IRC | 12:11 | |
| *** fungi has joined #openstack-swift | 12:18 | |
| *** kei_yama has quit IRC | 12:30 | |
| *** csmart has quit IRC | 12:39 | |
| *** csmart has joined #openstack-swift | 12:40 | |
| jith | hi all.. I have a query, I have done a setup of openstack cloud with swift as a backend for glance in mitaka version.. From command line it is working fine, but in dashboard, when i click object-store→ container, it shows blank page.. Is there anything to be enabled in horizon for mitaka version?? | 12:48 |
| jith | My proxy-server.conf - http://paste.openstack.org/show/600718/In swift log everything is normal except this, http://paste.openstack.org/show/600719/ | 12:48 |
| jith | pls do guide | 12:48 |
| *** cshastri has quit IRC | 13:03 | |
| -openstackstatus- NOTICE: restarting gerrit to address performance problems | 13:08 | |
| *** ChanServ changes topic to "restarting gerrit to address performance problems" | 13:09 | |
| *** klamath has joined #openstack-swift | 13:26 | |
| *** klamath has quit IRC | 13:26 | |
| *** klamath has joined #openstack-swift | 13:26 | |
| *** _JZ_ has joined #openstack-swift | 13:28 | |
| *** amoralej is now known as amoralej|lunch | 13:32 | |
| -openstackstatus- NOTICE: ok gerrit is back to normal | 13:38 | |
| *** ChanServ changes topic to "ok gerrit is back to normal" | 13:38 | |
| tdasilva | good morning | 13:43 |
| *** ChanServ changes topic to "Topic: Let's talk, we're nice. | PTG etherpad: https://etherpad.openstack.org/p/swift-ptg-pike | Logs: http://eavesdrop.openstack.org/irclogs/%23openstack-swift/ | Meetings: https://wiki.openstack.org/wiki/Meetings/Swift | Priority Reviews: https://wiki.openstack.org/wiki/Swift/PriorityReviews" | 13:44 | |
| -openstackstatus- NOTICE: gerrit is back to normal and I don't know how to use the openstackstaus bot | 13:44 | |
| asettle | Jesse is incredibly type A and very driven. Him and I both are in different ways. JP sits between us and I do feel sorry for him sometimes. | 13:45 |
| asettle | OH FFS | 13:45 |
| asettle | Guess who changed the channel and pressed enter | 13:45 |
| asettle | This idiot | 13:45 |
| openstackgerrit | Gábor Antal proposed openstack/swift master: Use more specific asserts in test/unit/common https://review.openstack.org/342781 | 13:49 |
| *** silor has joined #openstack-swift | 13:49 | |
| *** silor1 has joined #openstack-swift | 13:53 | |
| *** silor has quit IRC | 13:55 | |
| *** silor1 is now known as silor | 13:55 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Small homework for ec duplciation https://review.openstack.org/438862 | 13:57 |
| *** openstackgerrit has quit IRC | 14:03 | |
| *** bkopilov has joined #openstack-swift | 14:03 | |
| *** cppforlife_ has joined #openstack-swift | 14:09 | |
| *** chlong has joined #openstack-swift | 14:30 | |
| *** zaitcev has joined #openstack-swift | 14:35 | |
| *** ChanServ sets mode: +v zaitcev | 14:35 | |
| *** amoralej|lunch is now known as amoralej | 14:38 | |
| *** foutatoro has joined #openstack-swift | 14:45 | |
| *** sams-gleb has quit IRC | 15:01 | |
| *** sams-gleb has joined #openstack-swift | 15:02 | |
| *** dmorita has joined #openstack-swift | 15:04 | |
| *** sams-gleb has quit IRC | 15:06 | |
| *** dmorita has quit IRC | 15:08 | |
| *** chlong has quit IRC | 15:11 | |
| *** sams-gleb has joined #openstack-swift | 15:16 | |
| *** hseipp has quit IRC | 15:39 | |
| *** psachin has quit IRC | 15:41 | |
| *** chlong has joined #openstack-swift | 15:43 | |
| *** ediardo has quit IRC | 15:44 | |
| *** ediardo has joined #openstack-swift | 15:46 | |
| *** tonanhngo has joined #openstack-swift | 15:49 | |
| notmyname | good morning | 16:00 |
| *** bob_cheesey has quit IRC | 16:04 | |
| *** rcernin has quit IRC | 16:08 | |
| timburke | good morning | 16:08 |
| *** foutatoro has quit IRC | 16:09 | |
| *** jcaron has quit IRC | 16:15 | |
| *** jcaron has joined #openstack-swift | 16:16 | |
| *** jcaron1 has joined #openstack-swift | 16:26 | |
| *** jcaron has quit IRC | 16:26 | |
| *** pcaruana has quit IRC | 16:34 | |
| *** chsc has joined #openstack-swift | 16:37 | |
| *** chsc has joined #openstack-swift | 16:37 | |
| *** Jeffrey4l has quit IRC | 16:41 | |
| *** csmart has quit IRC | 16:42 | |
| *** Jeffrey4l has joined #openstack-swift | 16:42 | |
| *** gatuus has joined #openstack-swift | 16:44 | |
| *** openstackgerrit has joined #openstack-swift | 16:46 | |
| openstackgerrit | Clay Gerrard proposed openstack/pyeclib master: Add py35 tox environment https://review.openstack.org/438799 | 16:46 |
| openstackgerrit | Clay Gerrard proposed openstack/pyeclib master: Use py35 tox environment by default https://review.openstack.org/438799 | 16:47 |
| *** cppforlife_ has quit IRC | 16:47 | |
| *** cbartz has quit IRC | 16:49 | |
| *** csmart has joined #openstack-swift | 16:51 | |
| notmyname | I'm going through Boston summit presentations this morning. some of them stand out just because of "here's this cool stuff in swift!" but I've never seen any of it upstream | 16:52 |
| tdasilva | notmyname: yeah, i noticed that too :/ | 16:52 |
| notmyname | which isn't to say that everything has to go upstream, but on the other hand, stuff that's modifying swift should probably have some upstream collaboration | 16:52 |
| notmyname | so, I'm like "cool!" but also "who are you?" | 16:53 |
| *** geaaru has quit IRC | 16:55 | |
| openstackgerrit | Thiago da Silva proposed openstack/swift master: refactor some common code from crypto https://review.openstack.org/439058 | 16:58 |
| *** tqtran has joined #openstack-swift | 16:59 | |
| *** tqtran has quit IRC | 16:59 | |
| *** mvk has quit IRC | 17:02 | |
| *** tesseract has quit IRC | 17:09 | |
| *** csmart has quit IRC | 17:11 | |
| *** jordanP has quit IRC | 17:19 | |
| clayg | tdasilva: so you think that we don't really *need* to merge something like this https://review.openstack.org/#/c/438793 | 17:21 |
| patchbot_ | patch 438793 - rpm-packaging - Add PyECLib and liberasurecode | 17:21 |
| clayg | tdasilva: in order to get infra/upstream packages in the gate? | 17:22 |
| clayg | tdasilva: because you can just use existing rdo packages on centos nodes? | 17:22 |
| tdasilva | clayg: i actually am not sure. I need to figure that out...I was given a link yesterday that seems to list packages that are built for every commit, which maybe could be kind of cool??? idk | 17:25 |
| tdasilva | OTOH, does it solve our problem of specifying a release tag? | 17:25 |
| clayg | notmyname: it reminds me of the "never had only 3 contributors" problem - rather we have years of experience keeping clasically-closed-source-big-corp from messing up the projects while learning how to upstream | 17:26 |
| tdasilva | clayg: because even if with that we can build again top of master, we still really want to be able to build again a tagged release that has not been made available yet by downstream packagers | 17:26 |
| clayg | tdasilva: yeah ideally we would have a cent7-downstream and cent7-upstream job - they would both say "I need liberasurecode-devel" one would get what you or I would get on our cent7 vm - the cent7-upstream would get whatever was latest the tagged release of liberasurecode - because those packages are built automatically and stuffed into the infra repository | 17:27 |
| tdasilva | right | 17:27 |
| clayg | tdasilva: ah right, and the rdo packages might have "newer than stock downstream" - but they won't have "constantly built from tags during the development of the next release" | 17:27 |
| clayg | tdasilva: ok, so we should try to address Dirk's comments? And try to better understand what's the next steps after we have liberasurecode in the rpm-pacaging | 17:28 |
| tdasilva | clayg: looking at Dirk's comments | 17:29 |
| tdasilva | clayg: a problem we might have is that liberasurecode does not get a tarball in http://tarballs.openstack.org/ | 17:30 |
| tdasilva | not sure if that's what Dirk is asking for on hist third comment there | 17:31 |
| clayg | tdasilva: ah, maybe so - i kind of remember writing that down somewhere | 17:31 |
| clayg | that's good tho!? | 17:31 |
| clayg | ... except the deb-liberasurecode job doesn't need it apparently? | 17:31 |
| tdasilva | isn't that because the deb-liberasurecode actually copies the whole source code to their repo? | 17:32 |
| notmyname | timburke just pointed this ML post out to me. http://lists.openstack.org/pipermail/openstack-dev/2017-February/113029.html tl;dr horizon moving to a 1 +2 to land model (with a goal of landing small patches sooner) | 17:34 |
| robcresswell | notmyname: My ears are burning | 17:34 |
| *** mmotiani_ has joined #openstack-swift | 17:34 | |
| notmyname | robcresswell: it's really interesting. something that we've talked about before (and chosen not to make "official" guidelines about) | 17:35 |
| robcresswell | notmyname: I dont know what Swifts situation is, but Horizon now has 2 active cores, and a couple very part time. | 17:35 |
| notmyname | robcresswell: for us, the only rule is "use your best judgement" with a bias towards 2 cores. | 17:35 |
| robcresswell | So we really dont haver a choice :) | 17:35 |
| notmyname | yeah, that's a really important point | 17:36 |
| robcresswell | have a * | 17:36 |
| tdasilva | we have a hard time landing patches with 10+ cores, can't imagine with 2... | 17:36 |
| robcresswell | Yeah. The idea here is that we can bump small patches that people feel comfortable with; minor improvements, docs, etc, without holding them up on dual approval. | 17:38 |
| robcresswell | higher risk of course, but thats the decision I'm going with for now :) | 17:38 |
| notmyname | again, openstack never learned how to be successful as or with small projects. we're only learning that now | 17:38 |
| notmyname | robcresswell: yeah, we've always had a standing "if you think it's low impact and an ok patch, go ahead and land it" rule | 17:39 |
| notmyname | robcresswell: so I think it's a totally reasonable decision to make, especially given the limited cores you have. | 17:39 |
| robcresswell | notmyname: Thats good to hear, thanks | 17:40 |
| notmyname | robcresswell: I hope it results in shorter patch review time overall. however, I suspect the cost might be increased "stability" testing or integration work. but only in the sense that more eyes up front is likely to account for more edge cases before it lands instead of after | 17:42 |
| *** gatuus has quit IRC | 17:42 | |
| notmyname | robcresswell: regardless of how many reviewers you have, we'll always land bugs, so the basic "try to make it better and solve user problems, fixing bugs as we know about them" is still exactly the same | 17:43 |
| robcresswell | notmyname: It'll definitely increase the risk of faulty/incomplete patches landing. | 17:43 |
| robcresswell | My hope is that maintaining some pace will encourage new contributors who may become core | 17:43 |
| robcresswell | the other option is to make the merge rate painfully slow, which wont add to the cores long term | 17:43 |
| notmyname | I've floated the idea that we should just give everyone commit rights. (I was told that is a bad idea) ;-) | 17:44 |
| zaitcev | I routinely eschew +A even on 2x+2 if the patch is something tricky like the part++ patch. | 17:45 |
| * robcresswell shudders at the thought | 17:45 | |
| robcresswell | Absolutely, hence the note in the email to ask for a second opinion for complex code. The important part is that cores feel they *can* approve code with a single +2, not that they *must* | 17:46 |
| notmyname | that's the key, to me. core, or having a commit bit, is about trust. so if a core says something should or shouldn't land, they should be able to do that on their own. and I trust them to make the right call regardless of how many other reviews there are | 17:48 |
| *** gatuus has joined #openstack-swift | 17:48 | |
| *** dmorita has joined #openstack-swift | 17:50 | |
| tdasilva | clayg: i'm trying to organize my thoughts on how to make progress on gate jobs tasks. What are you focusing on, what can I help you with or maybe find others to help? | 17:51 |
| robcresswell | notmyname: Agreed | 17:51 |
| *** csmart has joined #openstack-swift | 17:51 | |
| tdasilva | clayg: would it help to share a etherpad, trello, whatever to compile a list of task items, goals? | 17:51 |
| *** dmorita has quit IRC | 17:52 | |
| *** dmorita has joined #openstack-swift | 17:52 | |
| *** mmotiani_ has quit IRC | 18:05 | |
| *** chlong has quit IRC | 18:06 | |
| *** klrmn has joined #openstack-swift | 18:11 | |
| tdasilva | another major s3 outage :/ | 18:26 |
| zaitcev | all regions or just the default Virginia center? | 18:39 |
| *** thumpba has quit IRC | 18:40 | |
| notmyname | us-east-1 | 18:42 |
| *** mmotiani_ has joined #openstack-swift | 18:43 | |
| *** cppforlife_ has joined #openstack-swift | 18:45 | |
| notmyname | "But wait. Isn't S3 "the cloud". Everyone promised the cloud would never go down, ever. It has infinite uptime and reliability. Well good thing I have my backups on [some service that happens to also use S3 as a backend]." | 18:46 |
| clayg | tdasilva: i'm just sort of chipping at the eduges | 18:48 |
| *** bkopilov has quit IRC | 18:49 | |
| clayg | tdasilva: if I'm entirely honest about the epic, it's really - know enough about openstack-infra/devstack/gate so that when someone says "but golang!? you can't!" I can say "we already do" | 18:50 |
| notmyname | that's the ticket! ^ | 18:50 |
| clayg | all this libec stuff is mostly a means to an end - if it has independent value along the way it's kind of only a bonus | 18:50 |
| *** ChubYann has joined #openstack-swift | 18:51 | |
| clayg | it also help me get infra to help with stuff that is already a problem so they don't see me as making new golang problems | 18:51 |
| clayg | by the time I start to ask for changes to be merged to infra projects "because golang" I should already be able to make a reasoned argument as to why the changes is obviously the correct path forward because point to how we handle other non-python services/depencies | 18:53 |
| clayg | tdasilva: so i'm ok with a trello board for "swift merges golang code" - and it having libec/pyeclib stories in it because that's a great way to get started | 18:55 |
| *** dmorita has quit IRC | 18:59 | |
| *** dmorita has joined #openstack-swift | 19:01 | |
| tdasilva | clayg: ack | 19:03 |
| *** bkopilov has joined #openstack-swift | 19:05 | |
| *** thumpba has joined #openstack-swift | 19:10 | |
| *** thumpba has quit IRC | 19:10 | |
| tdasilva | clayg: when you have a chance, can you take a look at patch 438758 | 19:11 |
| patchbot_ | https://review.openstack.org/#/c/438758/ - swift - fix bindep file to enable func test in centos | 19:11 |
| timburke | oh yeah, what *is* the deal with that? can we get rid of python-devel, too? what about the deb -dev packages? | 19:12 |
| *** patchbot_ is now known as patchbot | 19:12 | |
| *** amoralej is now known as amoralej|off | 19:13 | |
| clayg | tdasilva: I'm with timburke that doesn't look obviously correct at all | 19:16 |
| clayg | tdasilva: "fix bindep/centos" is not enough context to judge the change | 19:16 |
| clayg | tdasilva: what's broken? why does that change help? | 19:17 |
| tdasilva | clayg: sorry, should have provided better patch comments | 19:18 |
| tdasilva | it's currently broken because there's no such thing as python3-devel | 19:18 |
| tdasilva | rpm package | 19:18 |
| tdasilva | ideally we would substitue for the right thing | 19:18 |
| timburke | ...and we can't because...? | 19:18 |
| tdasilva | but in centos there's only py34, which i don't think we are interested in at the moment | 19:19 |
| tdasilva | so for now....i'm just not adding a py3 job for centos nodes | 19:19 |
| tdasilva | but would like to add other jobs | 19:19 |
| tdasilva | then i'll look into adding py3 jobs in centos nodes | 19:19 |
| openstackgerrit | Jim Cheung proposed openstack/liberasurecode master: Add Phazr.IO libphazr backend to liberasurecode https://review.openstack.org/424353 | 19:19 |
| tdasilva | py35 jobs that is | 19:19 |
| tdasilva | timburke, clayg: does that make sense? if it does, i'll add as a comment to the patch in case we need to look back into it | 19:21 |
| tdasilva | i mean commit message | 19:21 |
| openstackgerrit | Merged openstack/python-swiftclient master: Expose --prefix as an option for st_delete https://review.openstack.org/437332 | 19:22 |
| clayg | tdasilva: that makes *more* sense | 19:24 |
| tdasilva | but? | 19:25 |
| *** chlong has joined #openstack-swift | 19:25 | |
| timburke | is py34 available for centos in the gate currently? | 19:25 |
| tdasilva | timburke: yeah, the rpm package i believe is available, so I could add that in place if you all think it will be helpful | 19:26 |
| clayg | tdasilva: so I guess the part that would complete the picture for me is a failing gate-swift-tox-xfs-tmp-py35-centos-7 job since I guess ubuntu-xenial's python3-dev installs a py35 version? | 19:26 |
| tdasilva | and we can focus on py35 later?? | 19:26 |
| clayg | yeah for me python34-devel makes more sense than pulling it out | 19:27 |
| clayg | even if no jobs use it | 19:27 |
| timburke | i'm just struggling to see why the py34 -devel package doesn't already satisfy python3-devel | 19:27 |
| tdasilva | works for me | 19:27 |
| * clayg goes to find a cent7 node | 19:28 | |
| tdasilva | timburke: i was just going by a paste clayg provided yesterday that py34 was for <= Mitaka | 19:28 |
| tdasilva | so my understanding is that we are now targeting py35 | 19:28 |
| tdasilva | but that's ok, let's add py34 now, this should not be a big deal... | 19:28 |
| timburke | that's certainly true, at least for ubuntu. no idea about what -infra's plans are wrt centos, though | 19:29 |
| tdasilva | timburke: me neither :/ | 19:30 |
| clayg | timburke: I think infra thinks ubuntu by default - I would have thought sometime around xenial they started thinking py35 instead of py34 because that's what their default downstream ships? | 19:31 |
| clayg | also py35 is better than py34 in a bunch of cool async def ways ;) | 19:31 |
| timburke | clayg: yeah, that's my impression, too | 19:31 |
| timburke | but py36 is *even better* | 19:31 |
| timburke | (i assume. it's 0.1 bigger!) | 19:31 |
| clayg | 10% is a big deal | 19:32 |
| notmyname | at least I know I've made *some* impact on the swift community over the years. "bigger version == better" ;-) | 19:35 |
| clayg | is swift-get-nodes -p known broken for non-legacy policies? | 19:38 |
| clayg | ssh 127.0.0.1 "ls -lah ${DEVICE:-/srv/node*}/sdb14/objects-/673" | 19:38 |
| notmyname | does it use the right ring and just mess up the output there? | 19:39 |
| clayg | looks like the right devices for that part on that ring | 19:40 |
| clayg | notmyname: output is correct if you use -P option instead of ring | 19:41 |
| clayg | i don't think we have a bug for it | 19:41 |
| notmyname | I'm working on a PTG summary for the ML. anyone want something specific to be said? | 19:41 |
| clayg | notmyname: it was awesome | 19:44 |
| clayg | notmyname: put it something about the food | 19:44 |
| notmyname | clayg: clearly the most important part of the week ;-) | 19:44 |
| redbo | And someone should make a costco run for snacks next time. | 19:45 |
| *** Renich has joined #openstack-swift | 19:46 | |
| timburke | hmmm.... now i wonder how swift-get-nodes behaves mid-part-power-increase.... | 19:47 |
| clayg | timburke: which part? | 19:48 |
| notmyname | clayg: all of the parts! *ba-dum-dish* | 19:48 |
| timburke | well, both parts, anyway | 19:48 |
| timburke | at any rate, it might be worth noting that an increase is in progress... | 19:49 |
| *** dja has joined #openstack-swift | 19:55 | |
| *** gatuus has quit IRC | 20:07 | |
| *** mmotiani_ has quit IRC | 20:10 | |
| *** mmotiani__ has joined #openstack-swift | 20:10 | |
| *** dja has quit IRC | 20:11 | |
| *** mmotiani_ has joined #openstack-swift | 20:15 | |
| *** mmotiani__ has quit IRC | 20:15 | |
| *** mmotiani_ has quit IRC | 20:23 | |
| *** silor has quit IRC | 20:36 | |
| *** mvk has joined #openstack-swift | 20:37 | |
| notmyname | my thoughts on the PTG http://lists.openstack.org/pipermail/openstack-dev/2017-February/113062.html | 20:39 |
| notmyname | (I didnt' mention food) | 20:39 |
| *** dja has joined #openstack-swift | 20:39 | |
| clayg | I didn't mention food is basically like saying "you don't need to read this there's nothing important in it" | 20:41 |
| *** tonanhngo has quit IRC | 20:42 | |
| *** dja has quit IRC | 20:44 | |
| tdasilva | lol | 20:46 |
| clayg | tdasilva: I've been making jokes all week about how everytime someone talks about "how was the PTG" food comes up - my attitude is basically "did we eat; not sure I remember that.." | 20:47 |
| tdasilva | i just remember bbq! that was good | 20:48 |
| tdasilva | but yeah, sandwiches got old really quickly and no snacks :/ | 20:49 |
| *** dja has joined #openstack-swift | 20:54 | |
| openstackgerrit | Thiago da Silva proposed openstack/swift master: fix bindep file to enable func test in centos https://review.openstack.org/438758 | 20:56 |
| *** dja has quit IRC | 21:00 | |
| jrichli | notmyname: good write-up. I am glad you mentioned that scheduling and awareness has issues to work-out. I didn't see that in a different PTG summary I saw on Twitter. | 21:07 |
| clayg | rledisez: onovy: do we have a bug for the thing that says "we shouldn't force a recalc when we do_listdir?" | 21:23 |
| *** dfflanders has joined #openstack-swift | 21:24 | |
| *** chlong has quit IRC | 21:27 | |
| timburke | tdasilva: so it seems kinda strange that there wouldn't be a python3-devel meta-package that would grab 3.4 or 3.5 or whatever -- how does that work for the non-devel package? do they have to install python34? is that just baked into the base image? | 21:31 |
| tdasilva | timburke: i think that's just what's available in centos epel | 21:33 |
| tdasilva | i'm not sure why they didn't provide a python3-devel meta-package | 21:33 |
| tdasilva | strange, i agree | 21:33 |
| tdasilva | at least that's my understanding... | 21:33 |
| timburke | yep, seems to check out looking at https://dl.fedoraproject.org/pub/epel/7/x86_64/p/ ... | 21:34 |
| *** Jeffrey4l has quit IRC | 21:35 | |
| *** dja has joined #openstack-swift | 21:36 | |
| *** chlong has joined #openstack-swift | 21:39 | |
| clayg | tdasilva: i added check experimental - should work | 21:40 |
| tdasilva | clayg: thanks! also proposed this: https://review.openstack.org/#/c/439143/ | 21:41 |
| patchbot | patch 439143 - openstack-infra/project-config - add centos-7 jobs to libec, pyeclib, swift | 21:41 |
| tdasilva | not sure i got it all right, but maybe -infra guys can provide some guidance | 21:41 |
| *** Renich has quit IRC | 21:42 | |
| *** Jeffrey4l has joined #openstack-swift | 21:48 | |
| *** chlong has quit IRC | 21:58 | |
| timburke | tdasilva: where was the change for pyeclib? i think i missed it... | 22:01 |
| *** tonanhngo has joined #openstack-swift | 22:01 | |
| timburke | clayg: it's no fair nerd-sniping me with a bug and then not reviewing the patch -- https://review.openstack.org/#/c/438720/ | 22:08 |
| patchbot | patch 438720 - swift - Stop using client headers for cross-middleware com... | 22:08 |
| timburke | bah, i mean https://review.openstack.org/#/c/437236/ | 22:08 |
| patchbot | patch 437236 - swift - Log correct status code for conditional requests | 22:08 |
| mattoliverau | morning | 22:13 |
| *** tonanhngo has quit IRC | 22:13 | |
| *** adriant has joined #openstack-swift | 22:21 | |
| timburke | oh oh! mattoliverau! you can go land patch 331369 now that you understand how os.environ['TZ'] is bass-ackwards! | 22:23 |
| patchbot | https://review.openstack.org/#/c/331369/ - swift - Always set swift processes to use UTC | 22:23 |
| notmyname | mattoliverau: hello! I wanted to ask you a couple of questions about your PR to runway | 22:33 |
| notmyname | mattoliverau: but I just got back from the kids' school. give me a bit to get my mind back in work mode ;-) | 22:33 |
| mattoliverau | timburke: kk | 22:34 |
| mattoliverau | notmyname: kk | 22:34 |
| timburke | thanks mattoliverau :-) | 22:42 |
| mattoliverau | timburke: nps :) | 22:43 |
| mattoliverau | I had tested it alot, which is what lead to my initial confustion, thanks for clearing it up at the PTG | 22:43 |
| notmyname | mattoliverau: is the TZ sign different because you're in the southern hemisphere? | 22:47 |
| *** sams-gleb has quit IRC | 22:48 | |
| timburke | sounds right | 22:48 |
| notmyname | openstack summit going back to Vancouver https://www.openstack.org/blog/2017/02/the-openstack-summit-is-returning-to-vancouver-in-may-2018/ | 22:54 |
| mattoliverau | yay, I liked vancouver! | 22:55 |
| mattoliverau | notmyname: sounds a good a reason as any :P | 22:55 |
| zaitcev | But... there's no developer-oriented content at the Summit anymore, right? | 23:19 |
| klrmn | that doesn't mean notmyname gets a pass =) | 23:21 |
| notmyname | klrmn: I'll probably get a free pass to go ;-) | 23:22 |
| notmyname | zaitcev: I need to look in to that. I'll do that this week | 23:23 |
| notmyname | mattoliverau: how does this work? https://github.com/swiftstack/runway/pull/1/files#diff-9339dccc15a12a00ff8ca4ba686a9417R50 | 23:24 |
| notmyname | mattoliverau: what's the source dir for that? | 23:24 |
| notmyname | I'd expect a chdir or something in there | 23:24 |
| klrmn | notmyname: i kinda meant pass *from* going, not pass *to* go…but only realized the ambiguity after hitting send. | 23:24 |
| notmyname | klrmn: I know :-) | 23:25 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Container sharding docs https://review.openstack.org/438322 | 23:25 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient master: Updated from global requirements https://review.openstack.org/89250 | 23:26 |
| *** kei_yama has joined #openstack-swift | 23:26 | |
| *** klamath has quit IRC | 23:31 | |
| mattoliverau | notmyname: roles/swift/files/ | 23:31 |
| mattoliverau | notmyname: anything relative comes from a roles files direcory. So it easier to move around. | 23:32 |
| mattoliverau | notmyname: so https://github.com/swiftstack/runway/pull/1/files#diff-1c6427c722c534363f630ecdb9339558 | 23:32 |
| mattoliverau | notmyname: that didn't link properly, so the 'ansible/make_rings.sh → ansible/roles/swift/files/remakerings' | 23:33 |
| *** chsc has quit IRC | 23:39 | |
| kota_ | good morning | 23:44 |
| timburke | good morning kota_! | 23:45 |
| kota_ | good morning timburke! | 23:45 |
| kota_ | timburke: did the fix for auth land? | 23:46 |
| timburke | not yet. notmyname's been distracted, i guess? he's the one that needs to push the buttons so the swift3 fix can land... | 23:46 |
| kota_ | i see | 23:47 |
| timburke | yeah, i may have pulled the trigger a bit early on submitting those patches :-/ | 23:48 |
| notmyname | timburke: what did I (not) do? | 23:50 |
| notmyname | mattoliverau: hmm...ok | 23:51 |
| mattoliverau | kota_: morning | 23:51 |
| notmyname | mattoliverau: ah! I see | 23:51 |
| timburke | merge the tempauth fix (& backport) so we can close https://bugs.launchpad.net/swift3/+bug/1561199 | 23:51 |
| openstack | Launchpad bug 1561199 in OpenStack Security Advisory "Client-accessible headers are used to send authentication information to other middlewares" [Undecided,Incomplete] | 23:51 |
| timburke | notmyname: i'm tempted to go +2/+A https://review.openstack.org/#/c/438720/ but you're the one that needs convincing on the backport | 23:52 |
| patchbot | patch 438720 - swift - Stop using client headers for cross-middleware com... | 23:52 |
| notmyname | timburke: IIRC the only reason to do a backport is to enable gate checking of stable branches (ie to backport the real patch) | 23:53 |
| timburke | kota_: you should take a look at ps2 on https://review.openstack.org/#/c/438719/ -- i think i covered most of your concerns? | 23:54 |
| patchbot | patch 438719 - swift3 - Stop using client headers for cross-middleware com... | 23:54 |
| timburke | notmyname: swift3 (master) gates off of ocata swift | 23:55 |
| notmyname | clayg: check out mordred's ML post about zuul v3. looks like actual job config can eventually be moved to our repo (I think that's awesome) | 23:55 |
| timburke | (used to be even older, so at least we've got that going for us) | 23:55 |
| *** dja has quit IRC | 23:57 | |
| *** dja has joined #openstack-swift | 23:58 | |
| kota_ | timburke: yup, i just read the subscription from gerrit that notify me it was updadted. | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!