| *** peluse_ has joined #openstack-swift | 00:00 | |
| *** peluse has quit IRC | 00:01 | |
| ho | good morning! | 00:06 |
|---|---|---|
| *** NM has joined #openstack-swift | 00:06 | |
| *** jasondotstar has quit IRC | 00:11 | |
| *** thurloat is now known as thurloat_isgone | 00:11 | |
| *** kota_ has joined #openstack-swift | 00:15 | |
| *** ChanServ sets mode: +v kota_ | 00:15 | |
| kota_ | hello | 00:16 |
| *** jasondotstar has joined #openstack-swift | 00:17 | |
| *** esker has quit IRC | 00:23 | |
| *** david-lyle has quit IRC | 00:24 | |
| *** ujjain has quit IRC | 00:24 | |
| *** ujjain has joined #openstack-swift | 00:26 | |
| *** ujjain has joined #openstack-swift | 00:26 | |
| mattoliverau | ho, kota_: morning | 00:27 |
| *** changbl has joined #openstack-swift | 00:34 | |
| *** zhill has quit IRC | 00:39 | |
| ho | kota_, mattoliverau: morning! | 00:39 |
| *** ccavanna has joined #openstack-swift | 00:46 | |
| *** janonymous_ has quit IRC | 00:46 | |
| *** nakagawamsa has joined #openstack-swift | 00:57 | |
| *** haypo has left #openstack-swift | 00:57 | |
| *** kota_ has quit IRC | 01:02 | |
| *** thurloat_isgone has quit IRC | 01:03 | |
| *** thurloat_isgone has joined #openstack-swift | 01:06 | |
| *** thurloat_isgone is now known as thurloat | 01:06 | |
| *** jasondotstar has quit IRC | 01:12 | |
| *** jasondotstar has joined #openstack-swift | 01:17 | |
| *** jasondotstar has quit IRC | 01:22 | |
| *** breitz has quit IRC | 01:22 | |
| *** thurloat is now known as thurloat_isgone | 01:27 | |
| *** janonymous has quit IRC | 01:36 | |
| *** jrichli has joined #openstack-swift | 01:37 | |
| *** breitz has joined #openstack-swift | 01:37 | |
| *** nexusz99 has joined #openstack-swift | 01:43 | |
| *** NM has quit IRC | 01:53 | |
| *** breitz has quit IRC | 01:57 | |
| *** wbhuber has joined #openstack-swift | 02:08 | |
| *** david-lyle has joined #openstack-swift | 02:19 | |
| *** haomaiwang has quit IRC | 02:21 | |
| *** haomaiwa_ has joined #openstack-swift | 02:22 | |
| *** david-lyle has quit IRC | 02:23 | |
| *** haomaiwa_ has quit IRC | 02:36 | |
| *** haomaiwang has joined #openstack-swift | 02:36 | |
| *** nakagawamsa has quit IRC | 02:36 | |
| *** nakagawamsa has joined #openstack-swift | 02:49 | |
| *** wbhuber has quit IRC | 02:51 | |
| *** sanchitmalhotra has joined #openstack-swift | 03:00 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 03:05 | |
| *** sanchitmalhotra has quit IRC | 03:07 | |
| *** sanchitmalhotra has joined #openstack-swift | 03:10 | |
| notmyname | good evening, world | 03:10 |
| lifeless | o/ | 03:11 |
| *** mragupat has joined #openstack-swift | 03:12 | |
| *** sanchitmalhotra1 has quit IRC | 03:12 | |
| notmyname | lifeless: hi! I don't normally see you around these parts. welcome :-) | 03:12 |
| *** mragupat_ has joined #openstack-swift | 03:13 | |
| *** haomaiwang has quit IRC | 03:14 | |
| lifeless | notmyname: I realised I wasn't lurking here, so fixed that :) | 03:15 |
| notmyname | :-) | 03:15 |
| lifeless | can't represent folk if I'm not seeing the pan,just for starters | 03:16 |
| notmyname | clayg: kota_: ho: acoles_away: peluse_: redbo: dfg_: torgomatic: in tomorrow's meeting I want to talk about heypo's patch to change flake8 from a whitelist to a blacklist https://review.openstack.org/#/c/205977/ | 03:16 |
| *** mragupat has quit IRC | 03:16 | |
| *** dmorita has joined #openstack-swift | 03:23 | |
| mattoliverau | notmyname: o/ | 03:28 |
| *** NM has joined #openstack-swift | 03:32 | |
| *** david-lyle has joined #openstack-swift | 03:34 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 03:44 |
| *** jrichli has quit IRC | 03:46 | |
| openstackgerrit | Carlos Cavanna proposed openstack/swift: Proposed improvement to Swift documentation aimed at first-time contributors. https://review.openstack.org/206861 | 03:59 |
| notmyname | mattoliverau: hey! I saw in the buffer playback that you're deathly ill | 04:00 |
| *** SkyRocknRoll has joined #openstack-swift | 04:02 | |
| notmyname | mattoliverau: I sincerely hope that you do not die | 04:02 |
| notmyname | ;-) | 04:02 |
| mattoliverau | notmyname: yeah, I have had a bad winter this year.. might make sure I get a flu shot next year. | 04:02 |
| *** NM has quit IRC | 04:03 | |
| mattoliverau | notmyname: thanks, but you know man flu is the biggest killer :P | 04:03 |
| notmyname | mattoliverau: not in your country | 04:04 |
| *** SkyRocknRoll has quit IRC | 04:07 | |
| *** proteusguy has quit IRC | 04:07 | |
| openstackgerrit | Merged openstack/swift: Filter Etag key from ssync replication-headers https://review.openstack.org/173973 | 04:07 |
| notmyname | ccavanna: that's a great patch! thank you! | 04:09 |
| *** ppai has joined #openstack-swift | 04:09 | |
| notmyname | clayg: check this one out https://review.openstack.org/#/c/206861/ | 04:09 |
| ccavanna | notmyname: Sure. I need to ammend one thing that didn't quite come as I expected :-) | 04:09 |
| *** mragupat_ has quit IRC | 04:09 | |
| ccavanna | notmayname: next, I will look for some reviews. | 04:10 |
| notmyname | ccavanna: I've got some notes that I email to new contributors, so I'll probably add them to your base document after it lands | 04:11 |
| openstackgerrit | Carlos Cavanna proposed openstack/swift: Proposed improvement to Swift documentation aimed at first-time contributors. Ammended link. https://review.openstack.org/206861 | 04:15 |
| lifeless | notmyname: I've commented on https://review.openstack.org/#/c/88736/ | 04:16 |
| notmyname | ccavanna: I think you also need to fix the py26 reference. also, and you could mention this in the doc too, (and this is a very small nit that wouldn't block landing) the commit message itself should have a short first line followed by longer paragraphs as needed | 04:16 |
| lifeless | notmyname: I don't know what you mean by 'resolver for openstack' - if you mean pip resolver, I have a working prototype in pip itself, but I don't understand the relevance to merging/not merging the requirements updates | 04:17 |
| notmyname | lifeless: that's mostly a note to mordred referencing something we talked about a long time ago. note that comment was in early 2014. there are current other patches to update hacking in swift | 04:18 |
| lifeless | notmyname: yep; I mentioned other bits in my comment there. | 04:18 |
| ccavanna | notmyname: I saw it after my ammed :-) Will fix it now. I really didn't have much to say in the commit message. | 04:18 |
| notmyname | (which BTW is annoying to me because the new versions bring in new flake8 which have IMO a bug but it's required because newer pbr is neede for pbr and newer pbr uses newer hacking) | 04:19 |
| lifeless | notmyname: but - since I'm the one driving the resolver/ecosystem work, unless I understand what you're actually looking for, all I can be sure of is that I'm unlikely to deliver what you need :) | 04:19 |
| lifeless | notmyname: hmm, newer pbr doesn't use newer hacking | 04:19 |
| *** proteusguy has joined #openstack-swift | 04:19 | |
| lifeless | notmyname: the dep is the other way around: old hacking has 'pbr<1.0' as a dep | 04:19 |
| notmyname | ah ok | 04:19 |
| lifeless | notmyname: and that holds pbr back | 04:19 |
| notmyname | I misremembered | 04:19 |
| lifeless | it's a result of the overly tight deps that folk tried during K | 04:20 |
| lifeless | and which I'm hoping we'll get fully unwound this cycle | 04:20 |
| notmyname | py requires newer pbr which requires newer hacking which requires newer flake8 | 04:21 |
| ccavanna | notmyname: I will change the text to just "There's no current support for Python 3" | 04:21 |
| notmyname | *py3 | 04:21 |
| lifeless | notmyname: py3 compatible requirements requires newer pbr which is blocked by old hacking | 04:22 |
| lifeless | notmyname: and newer hacking is undesirable due to flake8 ? | 04:22 |
| *** SkyRocknRoll has joined #openstack-swift | 04:22 | |
| lifeless | notmyname: or is it some specific tests in flake8 that concern you? | 04:23 |
| notmyname | ccavanna: hmm...that's something that may soon change (one hopes). but yeah, that would be good | 04:23 |
| notmyname | lifeless: newer flake8 introduces new tests that we want to ignore. but it doesn't support both ignore and select directives at the same time | 04:24 |
| notmyname | whereas older flake8 did | 04:24 |
| lifeless | notmyname: ah; seems like pinging sigmavirus about that would be good | 04:24 |
| notmyname | haypo has been running with it on our side | 04:24 |
| lifeless | yep | 04:25 |
| notmyname | peluse_: so when do I get some of this in a lab to run swift on? http://www.bbc.com/news/technology-33675734 ;-) | 04:25 |
| notmyname | lifeless: see https://review.openstack.org/#/c/205977/ | 04:26 |
| lifeless | notmyname: the suggestion that xpoint will be a cpu adjunct makes me think its pricey | 04:26 |
| lifeless | notmyname: [to make, not just to buy :)] | 04:26 |
| notmyname | which is why I ask for free samples ;-) | 04:26 |
| lifeless | notmyname: anyhow | 04:27 |
| lifeless | notmyname: what resolver thing are you speaking of ? | 04:27 |
| lifeless | and how will it make merging proposal bot updates more ok ? | 04:28 |
| openstackgerrit | Carlos Cavanna proposed openstack/swift: Swift documentation for first-time contributors. https://review.openstack.org/206861 | 04:28 |
| notmyname | the idea we talked about was something that would look at simple dependency version directives to determine if what was required in one file is a superset of what's required in another | 04:29 |
| notmyname | eg if global-requirements says foo>2.3 and swift says foo>2.0, then pass | 04:29 |
| notmyname | and by "simple" we were talking about a limited grammar. maybe not worrying about the places that had exclusions. depends on how complicated it would end up being | 04:30 |
| notmyname | but the goal was to be able to do better than a simple string match on a project's local requirements vs global requirements. | 04:30 |
| lifeless | ok, just forcing the projects to be a superset | 04:30 |
| notmyname | right | 04:30 |
| notmyname | my understanding is that there is something in the works | 04:30 |
| lifeless | nope | 04:30 |
| lifeless | nothing | 04:30 |
| notmyname | or similar. | 04:31 |
| lifeless | nothing at all | 04:31 |
| notmyname | somethign about the way requirements are parsed | 04:31 |
| lifeless | no spec, no work queued up, nothing in the backlog | 04:31 |
| lifeless | I need to run, but we should talk about actually putting something in the queue to be done | 04:31 |
| lifeless | and who needs to buy-in to such a thing | 04:31 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 04:32 |
| lifeless | the parsing has been overhauled already by me during this cycle for the constraints work | 04:32 |
| *** bwall has joined #openstack-swift | 04:49 | |
| bwall | notmyname: is it cool if i dpm you? | 04:50 |
| bwall | *pm | 04:50 |
| notmyname | sure | 04:50 |
| *** tamizh_g_ has joined #openstack-swift | 05:00 | |
| *** tamizh_geek has joined #openstack-swift | 05:03 | |
| *** tamizh_g_ has quit IRC | 05:03 | |
| tamizh_geek | Hello people, my patch is waiting for a final +2 here : https://review.openstack.org/#/c/204196/ Requesting core devs to please take a look. | 05:11 |
| *** sanchitmalhotra1 has joined #openstack-swift | 05:13 | |
| *** sanchitmalhotra has quit IRC | 05:16 | |
| *** sanchitmalhotra has joined #openstack-swift | 05:20 | |
| *** sanchitmalhotra1 has quit IRC | 05:22 | |
| *** proteusguy has quit IRC | 05:24 | |
| lifeless | notmyname: ok, I'm back | 05:27 |
| lifeless | notmyname: so yeah - I don't think 'strict superset' is incompatible with the current things driving policies | 05:28 |
| lifeless | notmyname: but it is wide enough impact that I think we need a cross project spec, + operator and distro checkins | 05:28 |
| *** nakagawamsa has quit IRC | 05:32 | |
| notmyname | lifeless: I'll keep my eyes open for any work on that. if something comes up, I'd appreciate a ping | 05:35 |
| lifeless | notmyname: I'll put a spec together for the tokyo cycle; we're sufficiently slow implementing the tox constraints stuff that I don't want to put more into the pipeline yet | 05:37 |
| *** proteusguy has joined #openstack-swift | 05:40 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 06:15 |
| *** hrou has quit IRC | 06:20 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Imported Translations from Transifex https://review.openstack.org/206899 | 06:36 |
| *** sanchitmalhotra1 has joined #openstack-swift | 06:36 | |
| *** sanchitmalhotra has quit IRC | 06:37 | |
| *** sanchitmalhotra has joined #openstack-swift | 06:58 | |
| *** tongli_ has quit IRC | 07:00 | |
| *** sanchitmalhotra1 has quit IRC | 07:00 | |
| *** rledisez has joined #openstack-swift | 07:02 | |
| *** geaaru has joined #openstack-swift | 07:24 | |
| openstackgerrit | Matthew Oliver proposed openstack/swift: Add handoff region affinity support to replicators https://review.openstack.org/199863 | 07:33 |
| *** ccavanna has quit IRC | 07:52 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 07:58 | |
| *** sanchitmalhotra has quit IRC | 08:00 | |
| *** jordanP has joined #openstack-swift | 08:06 | |
| *** acoles_away is now known as acoles | 08:08 | |
| *** jistr has joined #openstack-swift | 08:08 | |
| *** tamizh_g_ has joined #openstack-swift | 08:16 | |
| *** tsubic has joined #openstack-swift | 08:23 | |
| *** jasondotstar has joined #openstack-swift | 08:25 | |
| *** mahatic has joined #openstack-swift | 08:31 | |
| *** ho has quit IRC | 08:35 | |
| *** kei_yama has quit IRC | 08:39 | |
| *** km has quit IRC | 08:39 | |
| *** mahatic has quit IRC | 09:17 | |
| *** janonymous has joined #openstack-swift | 09:17 | |
| *** sanchitmalhotra has joined #openstack-swift | 09:23 | |
| *** sanchitmalhotra1 has quit IRC | 09:25 | |
| *** kota_ has joined #openstack-swift | 09:26 | |
| *** ChanServ sets mode: +v kota_ | 09:26 | |
| -openstackstatus- NOTICE: Currently our CI system is broken, jobs are not getting processed at all. | 09:30 | |
| *** ChanServ changes topic to "Currently our CI system is broken, jobs are not getting processed at all." | 09:30 | |
| *** sanchitmalhotra1 has joined #openstack-swift | 09:43 | |
| *** haypo has joined #openstack-swift | 09:44 | |
| *** sanchitmalhotra has quit IRC | 09:45 | |
| *** nexusz99 has quit IRC | 09:52 | |
| *** jasondotstar has quit IRC | 09:57 | |
| *** kota_ has quit IRC | 10:00 | |
| *** Kennan has quit IRC | 10:00 | |
| *** mitz has quit IRC | 10:00 | |
| *** sc68cal has quit IRC | 10:00 | |
| *** mitz has joined #openstack-swift | 10:00 | |
| *** Kennan has joined #openstack-swift | 10:01 | |
| *** kota_ has joined #openstack-swift | 10:01 | |
| *** ChanServ sets mode: +v kota_ | 10:01 | |
| *** sc68cal has joined #openstack-swift | 10:02 | |
| *** tdasilva has quit IRC | 10:05 | |
| *** blair has quit IRC | 10:05 | |
| *** tsubic has quit IRC | 10:05 | |
| *** acoles has quit IRC | 10:05 | |
| *** jasondotstar has joined #openstack-swift | 10:09 | |
| *** tsubic has joined #openstack-swift | 10:17 | |
| *** geaaru has quit IRC | 10:24 | |
| *** kota_ has quit IRC | 10:37 | |
| *** aix has quit IRC | 10:38 | |
| *** geaaru has joined #openstack-swift | 10:41 | |
| *** blair has joined #openstack-swift | 10:43 | |
| *** NM has joined #openstack-swift | 10:44 | |
| *** acoles_away has joined #openstack-swift | 11:00 | |
| *** acoles_away is now known as acoles | 11:00 | |
| *** ChanServ sets mode: +v acoles | 11:00 | |
| *** tdasilva has joined #openstack-swift | 11:01 | |
| *** dmorita has quit IRC | 11:04 | |
| *** geaaru has quit IRC | 11:10 | |
| *** geaaru has joined #openstack-swift | 11:13 | |
| *** jasondot_ has joined #openstack-swift | 11:23 | |
| *** jasondotstar has quit IRC | 11:25 | |
| *** cdelatte has joined #openstack-swift | 11:33 | |
| *** delattec has joined #openstack-swift | 11:33 | |
| *** sanchitmalhotra1 has quit IRC | 11:35 | |
| *** delatte has joined #openstack-swift | 11:40 | |
| *** delattec has quit IRC | 11:41 | |
| *** cdelatte has quit IRC | 11:42 | |
| *** tamizh_g_ has quit IRC | 11:47 | |
| *** delattec has joined #openstack-swift | 11:48 | |
| *** delatte has quit IRC | 11:51 | |
| *** ppai has quit IRC | 11:54 | |
| *** ChanServ sets mode: +v tdasilva | 11:57 | |
| *** ppai has joined #openstack-swift | 12:01 | |
| *** NM has quit IRC | 12:02 | |
| *** NM has joined #openstack-swift | 12:04 | |
| *** aix has joined #openstack-swift | 12:07 | |
| *** NM has quit IRC | 12:08 | |
| *** marcusvrn has joined #openstack-swift | 12:16 | |
| *** ccavanna has joined #openstack-swift | 12:17 | |
| *** ccavanna has quit IRC | 12:31 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Remove unused code in swift/obj/server.py https://review.openstack.org/207024 | 12:38 |
| *** ppai has quit IRC | 12:40 | |
| tdasilva | good morning | 12:43 |
| *** ChanServ changes topic to "Review Dashboard: http://goo.gl/8IUcKl | Summary Dashboard: http://goo.gl/qHus5v | Hackathon topics: https://etherpad.openstack.org/p/swift-midcycle-aug-2015 | https://etherpad.openstack.org/p/swift_encryption_issues | Logs: http://eavesdrop.openstack.org/irclogs/%23openstack-swift/" | 12:51 | |
| -openstackstatus- NOTICE: zuul's disks were at capacity. Space has been freed up and jobs are being re-queued. | 12:51 | |
| *** ppai has joined #openstack-swift | 12:53 | |
| *** annegentle has joined #openstack-swift | 12:57 | |
| *** mahatic has joined #openstack-swift | 13:02 | |
| *** tsubic has quit IRC | 13:04 | |
| *** hrou has joined #openstack-swift | 13:05 | |
| *** jrichli has joined #openstack-swift | 13:07 | |
| *** ccavanna has joined #openstack-swift | 13:15 | |
| *** tsubic has joined #openstack-swift | 13:17 | |
| *** NM has joined #openstack-swift | 13:21 | |
| *** jkugel has joined #openstack-swift | 13:22 | |
| -openstackstatus- NOTICE: zuul jobs after about 07:00 UTC may need a 'recheck' to enter the queue. Look if your change is in http://status.openstack.org/zuul/ and recheck if not. | 13:27 | |
| *** ppai has quit IRC | 13:32 | |
| *** annegentle has quit IRC | 13:36 | |
| *** NM has quit IRC | 13:43 | |
| *** jasondotstar has joined #openstack-swift | 13:49 | |
| *** jasondot_ has quit IRC | 13:52 | |
| *** mragupat has joined #openstack-swift | 13:59 | |
| *** mragupat has quit IRC | 13:59 | |
| *** wbhuber has joined #openstack-swift | 13:59 | |
| *** breitz has joined #openstack-swift | 13:59 | |
| *** mragupat has joined #openstack-swift | 14:00 | |
| openstackgerrit | Thiago da Silva proposed openstack/swift: fix rdo link https://review.openstack.org/207059 | 14:03 |
| *** fbo has quit IRC | 14:04 | |
| *** EmilienM has quit IRC | 14:05 | |
| *** haypo has quit IRC | 14:06 | |
| *** haypo has joined #openstack-swift | 14:06 | |
| *** haypo has quit IRC | 14:06 | |
| *** haypo has joined #openstack-swift | 14:06 | |
| *** EmilienM has joined #openstack-swift | 14:09 | |
| *** fbo has joined #openstack-swift | 14:09 | |
| *** annegentle has joined #openstack-swift | 14:14 | |
| *** NM has joined #openstack-swift | 14:15 | |
| *** NM1 has joined #openstack-swift | 14:15 | |
| *** NM has quit IRC | 14:19 | |
| *** silor has joined #openstack-swift | 14:24 | |
| *** SkyRocknRoll has quit IRC | 14:32 | |
| *** mahatic has quit IRC | 14:38 | |
| *** janonymous_ has joined #openstack-swift | 14:47 | |
| *** jlhinson has joined #openstack-swift | 14:53 | |
| *** esker has joined #openstack-swift | 14:54 | |
| *** minwoob has joined #openstack-swift | 14:55 | |
| *** nadeem has joined #openstack-swift | 14:55 | |
| *** nadeem has joined #openstack-swift | 14:56 | |
| *** jasondotstar has quit IRC | 14:59 | |
| *** jistr has quit IRC | 15:00 | |
| *** haomaiwang has joined #openstack-swift | 15:01 | |
| *** mragupat has quit IRC | 15:01 | |
| *** jistr has joined #openstack-swift | 15:02 | |
| *** haomaiwang has quit IRC | 15:03 | |
| *** haomaiwang has joined #openstack-swift | 15:03 | |
| *** zaitcev has joined #openstack-swift | 15:07 | |
| *** ChanServ sets mode: +v zaitcev | 15:07 | |
| *** mahatic has joined #openstack-swift | 15:10 | |
| *** SkyRocknRoll has joined #openstack-swift | 15:17 | |
| *** hunius has joined #openstack-swift | 15:19 | |
| hunius | Hi, I wan't to know, what's the best place to extend swift by a Meteor connection. I wan't to insert data into the meteor database from within a custom middleware in swift. My problem is, where to establish the ddp connection to the meteor server? This should happen on startup, but I don't know exactly the best way to do it in swift. Any would be great. Best Daniel | 15:19 |
| notmyname | good morning | 15:22 |
| notmyname | hunius: in wsgi middleware, the __init__ is called on startup and the __call__ is called for each request. so you should probably look into a threadpool of connections created in the __init__ that is consumed in the __call__ | 15:23 |
| notmyname | err..maybe not "probably". that's just where I'd start :-) | 15:23 |
| hunius | sounds good. I will try this. I | 15:24 |
| hunius | sounds good. I will try this. I am not so familiar with python, so I will start researching how I can implement a threadpool | 15:25 |
| notmyname | hunius: if you haven't seen it, here's a good place to start with swift middleware http://docs.openstack.org/developer/swift/development_middleware.html#creating-your-own-middleware | 15:26 |
| notmyname | hunius: ah, ok. well a pool of connections might be step 2 then. to start with, just to get the framework of the idea down, just deal with a single connection to meteor and one request at a time. then scale | 15:27 |
| hunius | Thanks a very lot for your help. Yes, I know the examples. I have also bought a swift book, but all the wsgi stuff is absolutely new to me:) | 15:27 |
| hunius | yes. My plan is to extract metadata from the uploaded file and write this into the meteor database. I think this has to be done also within a new thread, so the client gets the success answer when storing was successfull. The extracted metadata should be passed then by meteor automagically:) | 15:29 |
| hurricanerix | hunius: If you are new to Python, I think this book is pretty good: http://www.effectivepython.com/ | 15:31 |
| *** mahatic has quit IRC | 15:33 | |
| hunius | hurricanerix: thanks, I definetively need to extend my knowledge. I think it will be a bit more work on the swift side as initially assumed... | 15:34 |
| notmyname | hurricanerix: did I see in email that you're doing the defcore thing in austin today? | 15:34 |
| hurricanerix | notmyname: yeah, I am here now. | 15:36 |
| hurricanerix | There are currently networking talks going on. So I am just trying to learn what I can. :) | 15:38 |
| *** esker has quit IRC | 15:39 | |
| notmyname | hurricanerix: fun ;-) | 15:41 |
| hurricanerix | notmyname: This is interesting, but I am more excited about the swift midcycle. :) | 15:42 |
| *** esker has joined #openstack-swift | 15:50 | |
| *** esker has quit IRC | 15:50 | |
| *** esker has joined #openstack-swift | 15:54 | |
| *** mahatic has joined #openstack-swift | 15:56 | |
| *** mahatic has quit IRC | 15:56 | |
| *** tamizh_g_ has joined #openstack-swift | 15:57 | |
| *** haomaiwang has quit IRC | 16:01 | |
| *** esker has joined #openstack-swift | 16:01 | |
| *** haomaiwang has joined #openstack-swift | 16:02 | |
| *** jistr has quit IRC | 16:08 | |
| *** briancurtin has quit IRC | 16:13 | |
| *** zhiyan has quit IRC | 16:13 | |
| *** nottrobin has quit IRC | 16:13 | |
| *** serverascode has quit IRC | 16:13 | |
| peluse_ | acoles or jrichli - which patch is the tail of crypto again? | 16:16 |
| clayg | peluse_: it's on the wiki! | 16:17 |
| jrichli | https://review.openstack.org/#/c/203454/ | 16:17 |
| clayg | peluse_: p 203454 | 16:18 |
| jrichli | and yes, its on the wiki :-) | 16:18 |
| peluse_ | wiki wiki wiki | 16:18 |
| jrichli | wack | 16:18 |
| clayg | like Fozzy the bear? | 16:18 |
| *** david-lyle has quit IRC | 16:18 | |
| *** silor1 has joined #openstack-swift | 16:34 | |
| *** rledisez has quit IRC | 16:35 | |
| *** silor has quit IRC | 16:36 | |
| *** silor1 is now known as silor | 16:36 | |
| *** nottrobin has joined #openstack-swift | 16:37 | |
| *** jordanP has quit IRC | 16:39 | |
| *** david-lyle has joined #openstack-swift | 16:42 | |
| *** serverascode has joined #openstack-swift | 16:43 | |
| *** zhiyan has joined #openstack-swift | 16:46 | |
| acoles | peluse_: i saw some func test failures with patch 201283 - is that just badness with my SAIO? (I pasted failures on review) | 16:47 |
| patchbot | acoles: https://review.openstack.org/#/c/201283/ | 16:47 |
| *** tamizh_g_ has quit IRC | 16:49 | |
| peluse_ | thanks, I don't see any. I'm stuck in an all day meeting so have limited time to get anything real done but will take a look at your comments when I can - thanks! | 16:49 |
| wbhuber | clayg: wocka wocka wocka :) | 16:49 |
| peluse_ | acoles, yeah, those are strange errors for this patch :) I don't see them | 16:51 |
| peluse_ | acoles, and wrt the refactor diskfile patch, no I didn't see that until I had this other one working | 16:52 |
| peluse_ | acoles, do you think we should land that one and then work on this one? If so I'll review and push it through if it doesn't before I get a chance (tomorrow) and then have fun rebasing | 16:52 |
| acoles | peluse_: ok thats weird i don't see them on master. i'll dig some more. | 16:53 |
| *** wbhuber has quit IRC | 16:54 | |
| peluse_ | I'll rerun as well | 16:55 |
| acoles | peluse_: well that would be nice but i am biased ;P i'll try the rebase and see how painful it will be | 16:56 |
| peluse_ | :) | 16:57 |
| *** briancurtin has joined #openstack-swift | 16:57 | |
| peluse_ | acoles, bah you're right on the func tests. I didn't have EC as my default. You'd think I've never done this before... | 16:59 |
| acoles | peluse_: what! you of all people aren't running EC default! :D | 17:00 |
| acoles | peluse_: you need to hand back your t-shirt man :P | 17:00 |
| peluse_ | heh, I didn't get one! | 17:00 |
| acoles | awww | 17:01 |
| *** haomaiwang has quit IRC | 17:01 | |
| *** haomaiwang has joined #openstack-swift | 17:02 | |
| acoles | peluse_: we really ought to get a jenkins job running against EC. now there's a nice task for someone. | 17:03 |
| peluse_ | for sure | 17:05 |
| * peluse_ has no idea how to go about doing that | 17:05 | |
| *** _hrou_ has joined #openstack-swift | 17:05 | |
| *** hrou has quit IRC | 17:06 | |
| notmyname | acoles: peluse_: I can do that pretty easily in the swiftstack community qa cluster. or, more specifically, I can ask charz to set it up :-) | 17:06 |
| peluse_ | delegation... nice | 17:07 |
| acoles | notmyname: great idea | 17:08 |
| notmyname | it's 1am for him now, so I sent an email. | 17:09 |
| openstackgerrit | John Dickinson proposed openstack/swift: Fixes for mock 1.1 https://review.openstack.org/207131 | 17:09 |
| openstackgerrit | John Dickinson proposed openstack/swift: Fixes for mock 1.1 https://review.openstack.org/207132 | 17:10 |
| notmyname | ^ those are backports for the stuff that mock>1.0 broke. both stable branches have mock>=1.0, so any new patches on those branches (any other backports) won't pass the gate until those are landed first | 17:11 |
| notmyname | they are cherry-picked from haypo's patch: https://review.openstack.org/#/q/Id179513c6010d827cbcbdda7692a920e29213bcb,n,z | 17:11 |
| charz | notmyname: acoles What's ec test against swiftstack community cluster? | 17:14 |
| peluse_ | just func tests using an EC policy as default | 17:14 |
| notmyname | charz: setting up an EC policy as a default policy in swift.conf and running func tests | 17:14 |
| zaitcev | goodie grief, mock hurt swift too? | 17:15 |
| notmyname | zaitcev: it got more strict (see the patch on master). turns out we were calling some methods that didn't exist and some methods incorrectly | 17:16 |
| notmyname | so it's our fault, not mock's ;-) | 17:16 |
| charz | peluse_: notmyname oh, got it, I already have ec policy and 3-replica on the cluster, so I need put functional test for ec policy. | 17:17 |
| notmyname | charz: right | 17:17 |
| notmyname | just so that we have a test run against both | 17:17 |
| charz | notmyname: yes, we'll. | 17:17 |
| notmyname | thanks | 17:18 |
| charz | notmyname: np | 17:18 |
| zaitcev | notmyname: I do not understand. The mock 1.1 simply does not support py26. How are we using it in Kilo? | 17:21 |
| notmyname | that is a good question. I just ran py27 and it worked. might need to think about it more, then. or rather, might need to cap the version there | 17:22 |
| zaitcev | I hit that problem with git-review and I took the patch from swiftclient, which does | 17:23 |
| zaitcev | mock==1.0.1;python_version=='2.6' | 17:23 |
| zaitcev | mock>=1.0;python_version!='2.6' | 17:23 |
| notmyname | oh yeah. that requires newer pbr/pip too, though | 17:24 |
| timburke | zaitcev: notmyname: fwiw, mock went back to supporting 2.6 in 1.1.4 (and continues to through the latest release of 1.3.0) | 17:25 |
| *** hunius has quit IRC | 17:25 | |
| *** aix has quit IRC | 17:30 | |
| notmyname | timburke: zaitcev: I'm open to suggestions on what to do there | 17:33 |
| zaitcev | notmyname: if the infrastructure team moves to 1.1.4, I'll just review what you posted and we're done. Might need to undo the changes in swiftclient too. And abandon my fix for git-review (it's not merged yet). | 17:34 |
| notmyname | lifeless: mordred: ^ | 17:36 |
| *** janonymous_ has quit IRC | 17:36 | |
| lifeless | notmyname: ? econtex | 17:37 |
| zaitcev | lifeless: is it true that timburke said, that 1.1.4 is okay with py26, and it yes, are you going to have that in the gate? We'll drop "mock==1.0.1;python_version=='2.6'" then. | 17:38 |
| lifeless | zaitcev: I don't know what you mean by 'in the gate' | 17:38 |
| *** SkyRocknRoll has quit IRC | 17:38 | |
| notmyname | global test requirements is mock!=1.1.4,>=1.1;python_version!='2.6' | 17:38 |
| notmyname | timburke says mock >1.1.4 works with py26. | 17:39 |
| lifeless | zaitcev: no its not: http://git.openstack.org/cgit/openstack/requirements/tree/global-requirements.txt#n250 | 17:39 |
| notmyname | and this matters because I just proposed some mock fixes as a result of newer mock to a backport-branch that still supports py26 | 17:39 |
| zaitcev | lifeless: currently, if a project supports Python 2.6, it cannot be pushed through the gate at all. Like COMPLETELY FAILS. The blow-up looks like this: https://bugs.launchpad.net/openstack-gate/+bug/1473286 | 17:39 |
| openstack | Launchpad bug 1473286 in OpenStack-Gate "mock fails with SyntaxError in py26" [Undecided,Invalid] | 17:39 |
| lifeless | >=1.2 is the global requirements | 17:40 |
| lifeless | zaitcev: whichever project you're lookin at is probably not synced with global-requirements for some reason | 17:41 |
| notmyname | ah ok | 17:41 |
| lifeless | and yes, 1.2 and 1.3 support 2.6 | 17:41 |
| notmyname | so it seems we're good | 17:42 |
| notmyname | we need to make sure the mock dependency is updated, but that's easy. and the proposed backport will work and the branch is still supportable py26_ | 17:43 |
| notmyname | s/_/+ | 17:43 |
| openstackgerrit | John Dickinson proposed openstack/swift: update mock in test-requirements https://review.openstack.org/207143 | 17:49 |
| openstackgerrit | John Dickinson proposed openstack/swift: update mock in test-requirements https://review.openstack.org/207145 | 17:49 |
| notmyname | zaitcev: ^ there. mock>=1.2 for juno and kilo | 17:50 |
| openstackgerrit | John Dickinson proposed openstack/swift: Update test-requirements to match global requirements. https://review.openstack.org/204179 | 17:50 |
| *** SkyRocknRoll has joined #openstack-swift | 17:52 | |
| notmyname | and with that my laptop battery is getting very low. I'll be offline until I find more power. definitely by the meeting at 2100utc | 17:54 |
| *** jasondotstar has joined #openstack-swift | 17:54 | |
| lifeless | so the history here is this | 17:59 |
| lifeless | mock was vey stale | 17:59 |
| lifeless | with dev moved to cpython | 17:59 |
| lifeless | and I volunteered to do the rolling backport of it | 18:00 |
| lifeless | when I asked around noone volunteered to help with 2.6, so I took the route of 'if folk want 2.6 support pin < 1.1' | 18:00 |
| lifeless | turns out there were some folk willing to volunteer :). So we reintroduced 2.6 support | 18:00 |
| *** haomaiwang has quit IRC | 18:01 | |
| lifeless | now, there was a bad change in python 3.4, which we bckported, then reverted, redid and reapplyied | 18:01 |
| lifeless | which was the 1.1.x -> 1.2 -> 1.3 transitions | 18:01 |
| *** haomaiwang has joined #openstack-swift | 18:02 | |
| *** lcurtis has joined #openstack-swift | 18:19 | |
| clayg | history lessons are so much fun | 18:21 |
| *** jasondotstar has quit IRC | 18:23 | |
| openstackgerrit | Carlos Cavanna proposed openstack/swift: Swift documentation for first-time contributors. https://review.openstack.org/206861 | 18:23 |
| *** annegentle has quit IRC | 18:26 | |
| *** annegentle has joined #openstack-swift | 18:26 | |
| acoles | peluse_: i rebased your review+my suggested patch on top of patch 198429. It took some work to get things back into the right place but i have it working. | 18:31 |
| patchbot | acoles: https://review.openstack.org/#/c/198429/ | 18:31 |
| acoles | peluse_: the result is also less delta because more diskfile code is shared :) | 18:31 |
| *** openstackgerrit has quit IRC | 18:31 | |
| *** openstackgerrit has joined #openstack-swift | 18:32 | |
| acoles | peluse_: but i dunno what to do with my branch! i don't want to force your hand, either on the diskfile refactor review or on my suggested changes to yours. | 18:32 |
| peluse_ | acoles, that's great, thanks! | 18:33 |
| acoles | peluse_: so for want of better idea i will push to gerrit as a new review just so it is somewhere other than on my laptop | 18:33 |
| acoles | peluse_: then if you like you can slurp it back into your review | 18:33 |
| peluse_ | or you cqan push over mine if you want | 18:34 |
| * acoles isn't sure `git slurp` is a real command | 18:34 | |
| peluse_ | I'm debugging the strange failures... I have a screwy range thing going on. Shouldn't be too bad | 18:34 |
| tdasilva | acoles: I started reviewing the diskfile refactor a while back and noticed that diskfily.py has around 12 different classes now, have we considered moving manager classes to another file? | 18:35 |
| acoles | peluse_: well i could but maybe you want to look at the refactor first | 18:35 |
| acoles | tdasilva: yeah it grew 4 Base* classes | 18:36 |
| peluse_ | I can review as part of working on my patch is what I was thinking. either way works | 18:36 |
| acoles | tdasilva: idk how it would best be split out | 18:36 |
| acoles | tdasilva: would it make sense for the managers to be separated, or for the policy type class families to be separated? | 18:37 |
| acoles | tdasilva: i *think* clayg wanted to eliminate some of the classes altogether (the Reader/Writer perhaps) | 18:38 |
| tdasilva | acoles: yeah...I was strugling with the same questions...it was just that I noticed the file becoming huge and difficult to 'navigate' | 18:39 |
| tdasilva | maybe diskfile_base, diskfile_replication, diskfile_ec ??? not really sure... | 18:40 |
| acoles | tdasilva: i only made it 19 lines longer and I bet those were really useful comments :D :D | 18:40 |
| acoles | tdasilva: but yeah its a pain to navigate | 18:40 |
| acoles | tdasilva: if i get chance i will try what you said - diskfile_base, diskfile_replication, diskfile_ec | 18:41 |
| tdasilva | acoles: +1, didn't mean to offload on you, would be glad to pick it up on a subsequent patch too... | 18:43 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Refactor diskfile https://review.openstack.org/198429 | 18:43 |
| openstackgerrit | Alistair Coles proposed openstack/swift: WIP Alternate Add ability for GET path to see/select alternate frag archs https://review.openstack.org/207165 | 18:43 |
| peluse_ | acoles, cool, thanks | 18:44 |
| acoles | peluse_: np i gotta go so took the quick option :) will be around for mtg later | 18:45 |
| peluse_ | K | 18:45 |
| acoles | tdasilva: np follow on might be good since i have clayg's +2 and i wanted to minimise the grok'ing for reviewers | 18:45 |
| *** nadeem has quit IRC | 18:46 | |
| acoles | peluse_: the 3 failures are still there i'm afraid ;) | 18:46 |
| *** nadeem has joined #openstack-swift | 18:46 | |
| peluse_ | acoles, new ones or same func test as before? | 18:46 |
| acoles | peluse_: same ones | 18:47 |
| peluse_ | phew :) | 18:47 |
| acoles | heh | 18:47 |
| *** acoles is now known as acoles_away | 18:49 | |
| clayg | peluse_: heck yeah let's just make patch 198429 the new master - it'll merge eventually | 18:49 |
| patchbot | clayg: https://review.openstack.org/#/c/198429/ | 18:49 |
| peluse_ | yup, that's the plan. have to fix a few func tests first where I've got an off-by-one on range requests somewhere | 18:50 |
| *** acoles_away is now known as acoles | 18:50 | |
| peluse_ | but I'm stuck in a meeting all day so only sorta half-ass debugging | 18:51 |
| *** acoles is now known as acoles_away | 18:51 | |
| *** geaaru has quit IRC | 18:56 | |
| openstackgerrit | Carlos Cavanna proposed openstack/swift: Swift documentation for first-time contributors. https://review.openstack.org/206861 | 18:59 |
| *** haomaiwang has quit IRC | 19:01 | |
| *** haomaiwa_ has joined #openstack-swift | 19:01 | |
| openstackgerrit | Merged openstack/swift: Imported Translations from Transifex https://review.openstack.org/206899 | 19:05 |
| *** silor has quit IRC | 19:06 | |
| openstackgerrit | Merged openstack/swift: Remove unused code in swift/obj/server.py https://review.openstack.org/207024 | 19:13 |
| *** jasondotstar has joined #openstack-swift | 19:18 | |
| *** joeljwright has joined #openstack-swift | 19:25 | |
| *** ChanServ sets mode: +v joeljwright | 19:25 | |
| openstackgerrit | Carlos Cavanna proposed openstack/swift: Swift documentation for first-time contributors. https://review.openstack.org/206861 | 19:36 |
| *** joeljwright has quit IRC | 19:45 | |
| *** HenryG has quit IRC | 19:49 | |
| *** HenryG has joined #openstack-swift | 19:52 | |
| *** tamizh_g_ has joined #openstack-swift | 19:54 | |
| *** tamizh_g_ has quit IRC | 19:58 | |
| *** haomaiwa_ has quit IRC | 20:01 | |
| *** haomaiwa_ has joined #openstack-swift | 20:02 | |
| *** bill_az has joined #openstack-swift | 20:06 | |
| *** joeljwright has joined #openstack-swift | 20:08 | |
| *** ChanServ sets mode: +v joeljwright | 20:08 | |
| * notmyname back | 20:15 | |
| *** openstackgerrit has quit IRC | 20:16 | |
| *** openstackgerrit has joined #openstack-swift | 20:16 | |
| *** annegentle has quit IRC | 20:17 | |
| *** jasondotstar has quit IRC | 20:19 | |
| notmyname | note that there are 3 patches I'd like to discuss in the meeting (listed on https://wiki.openstack.org/wiki/Meetings/Swift). If you have a chance before the meeting to refresh your memory, that would be good | 20:20 |
| haypo | notmyname: hey. the meeting is in ~40 minutes, right? | 20:24 |
| notmyname | haypo: correct | 20:24 |
| haypo | notmyname: (hi.) should I prepare something for " change flake8 tests from whitelist to blacklist"? | 20:25 |
| notmyname | haypo: no more than you currently are. I'm planning on introducing the issue and I want to make sure people are aware of the implications of that patch and discuss any questions that people have | 20:27 |
| notmyname | haypo: it's not a "beat up on haypo's patch" thing. more of understanding the implications and making sure we are ok with those before we land it | 20:27 |
| haypo | ok | 20:28 |
| haypo | i just plan to answer questions if somone has a question :) | 20:28 |
| lifeless | o/ | 20:29 |
| haypo | lifeless: hey. our pip expert :) | 20:30 |
| lifeless | wow thats a terrifying box to be in :) | 20:31 |
| *** jasondotstar has joined #openstack-swift | 20:32 | |
| *** ahale has quit IRC | 20:32 | |
| *** ahale has joined #openstack-swift | 20:33 | |
| zaitcev | could be a peep expert | 20:40 |
| lifeless | so the problem is, I'm clearly in the state of having forgotten more than I currently know. So I could be whatever and I wouldn't know it :) | 20:42 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 20:47 |
| haypo | lifeless: haha | 20:48 |
| *** blmartin has joined #openstack-swift | 20:52 | |
| *** ho has joined #openstack-swift | 20:57 | |
| notmyname | meeting time in #openstack-meeting | 20:59 |
| *** kota_ has joined #openstack-swift | 21:00 | |
| *** ChanServ sets mode: +v kota_ | 21:00 | |
| openstackgerrit | Michael Barton proposed openstack/swift: go: make replicateDevice more defensive https://review.openstack.org/207217 | 21:00 |
| *** minwoob_ has joined #openstack-swift | 21:00 | |
| *** haomaiwa_ has quit IRC | 21:01 | |
| *** ccavanna has quit IRC | 21:01 | |
| *** 6JTABPDW5 has joined #openstack-swift | 21:02 | |
| *** minwoob_ has quit IRC | 21:02 | |
| *** acoles_away is now known as acoles | 21:03 | |
| *** jordanP has joined #openstack-swift | 21:07 | |
| *** ccavanna has joined #openstack-swift | 21:09 | |
| *** jkugel has quit IRC | 21:15 | |
| *** ekarlso has quit IRC | 21:16 | |
| *** ccavanna_ has joined #openstack-swift | 21:17 | |
| *** ccavanna has quit IRC | 21:20 | |
| *** ccavanna_ has quit IRC | 21:24 | |
| *** joeljwright1 has joined #openstack-swift | 21:32 | |
| *** joeljwright has quit IRC | 21:32 | |
| *** joeljwright1 has quit IRC | 21:33 | |
| *** joeljwright has joined #openstack-swift | 21:34 | |
| *** ChanServ sets mode: +v joeljwright | 21:34 | |
| *** jasondotstar has quit IRC | 21:41 | |
| *** ekarlso has joined #openstack-swift | 21:44 | |
| acoles | clayg: man you are so right about the bigger fish, i need to suppress my ocd :P | 21:47 |
| *** hrou has joined #openstack-swift | 21:48 | |
| *** _hrou_ has quit IRC | 21:48 | |
| blmartin | I dunno, with imagemagick you probably could programmatically generated jpg image with text :P | 21:49 |
| torgomatic | that's less "bigger fish" and more "oh crap, Cthulhu" | 21:51 |
| *** hrou has quit IRC | 21:52 | |
| acoles | mattoliverau: hey wassup? you ill? | 21:59 |
| clayg | phew - that was close | 22:00 |
| peluse_ | clayg, what do you think of https://bugs.launchpad.net/swift/+bug/1469094 ? | 22:00 |
| openstack | Launchpad bug 1469094 in OpenStack Object Storage (swift) "Missing .durable files of FAs can cause rebuildable objects to be inaccessible" [Medium,Confirmed] | 22:00 |
| *** 6JTABPDW5 has quit IRC | 22:01 | |
| joeljwright | notmyname: regards the swiftclient bugs we didn't get to on the agenda | 22:01 |
| clayg | peluse_: FWIW I work with Charles - he's a sharp cat - non native-english | 22:01 |
| clayg | peluse_: I'm not up to speed on that report - do you have a specific question? | 22:01 |
| peluse_ | clayg, yeah I gathered that.... | 22:01 |
| clayg | peluse_: ok sorry :) | 22:02 |
| mattoliverau | acoles: Hey! welcome back! yeah, been hit with some flu/stomache bug all week, it hasn't been pretty and spend the first 3 days out of it in bed. Getting better now tho, so kinda working again | 22:02 |
| *** 16WAAEHWP has joined #openstack-swift | 22:02 | |
| peluse_ | basically only have 2 .durables and relying on their replication via reconstructor worked but it took a while | 22:02 |
| joeljwright | notmyname: combination of deadline and being ill for 2 weeks meant I didn't get to them - will try to get some of them triaged this week | 22:02 |
| peluse_ | so do we just clsoe the bug or is it better to just require all .durables be written | 22:02 |
| peluse_ | is the question | 22:02 |
| acoles | mattoliverau: oh sounds bad, take it easy | 22:02 |
| *** Kennan2 has joined #openstack-swift | 22:02 | |
| clayg | peluse_: ah yes - that's how I read the state of the bug - maybe we could do better? | 22:02 |
| peluse_ | or bump up the "2" to something bigger | 22:02 |
| joeljwright | mattoliverau: if it makes you feel better, you're not the only one who's been sick | 22:02 |
| peluse_ | I mean, we chose "2" kinda randomly | 22:03 |
| clayg | well - it's less about how many durables and more about how many fragments? the durables are just sort of a hint | 22:03 |
| joeljwright | mattoliverau: it's getting boring now, but the combo of cough and screaming headache is one I could live without | 22:03 |
| mattoliverau | joeljwright: :( | 22:03 |
| *** Kennan has quit IRC | 22:03 | |
| acoles | joeljwright: ouch | 22:03 |
| peluse_ | clayg, huh? I think the big report was only about the .durables | 22:03 |
| clayg | peluse_: I think the bug report was about being able to read an object after 200 | 22:04 |
| clayg | er.. 201 | 22:04 |
| *** blmartin has quit IRC | 22:05 | |
| peluse_ | clayg, hmmm, maybe I'm reading it wrong but I'm not sure how that could be | 22:06 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Update hacking to 0.10.0 https://review.openstack.org/205977 | 22:06 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Fix pep8 E warning for hacking 0.10 https://review.openstack.org/207237 | 22:06 |
| haypo | acoles: https://review.openstack.org/207237 fix _most_ E warnings of hacking 0.10 | 22:07 |
| *** kota_ has quit IRC | 22:07 | |
| *** acoles is now known as acoles_away | 22:07 | |
| *** joeljwright has quit IRC | 22:07 | |
| haypo | acoles_away: my hacking 0.10 patch now only ignores E128 & E265 in the E category. they are simply to fix, but they will require larger patches. i prefer to do that later | 22:08 |
| ho | acoles: could you check patch 202411? when you run ./.functinaltests with this patch, you can see the behaivor which you gave me an email. i would like to know whether it's run with expected or not. | 22:08 |
| patchbot | ho: https://review.openstack.org/#/c/202411/ | 22:08 |
| clayg | peluse_: think about it this way, if we have 10 frags of a 10+4 scheme we have enough information to return the object on GET - even with NO .durables - so how many .durables *need* to exist on a GET to return the object to the client? all 10 for all the frags? | 22:08 |
| clayg | note i'm using "need" to mean theoretically - not "need" like what does the code do today. | 22:08 |
| clayg | if *theoretically* at least we could get back with one or two .durables then in practice we'd probably be fine - but I'm not sure that's what charles is observing - i'll try to tease some more understanding from the comments | 22:09 |
| peluse_ | clayg, well in that case none | 22:09 |
| clayg | peluse_: fair point! .durables are only there for dealing with when is it safe to clean up the older version and which versions we should try to replicate! | 22:10 |
| *** nadeem has quit IRC | 22:10 | |
| clayg | if the proxy could some how magically glean it can rebuild an object even without the .durable that'd be neat - but I wouldn't recommend it do so unless it thinks it will also eventually be rebuilt - but even 1 durable should be enough to eventually propogate to all the nodes | 22:11 |
| clayg | so if the proxy can find 10 frags and 1 durable I think it "should" return it | 22:11 |
| clayg | note I'm using "should" there to mean - it's a worthy goal - that would prioritized agains other things that also important | 22:12 |
| peluse_ | clayg, well that's an implementation change that's not so trivial I don't think | 22:12 |
| peluse_ | not for going from beta-->production anyway | 22:12 |
| clayg | peluse_: perhaps - i think this should be done is different from what's blocking taking off a beta tag | 22:13 |
| peluse_ | my vote would be leave it as or change the .durable requirement to something bigger like quorum which would simplify the code slightly | 22:13 |
| peluse_ | yup | 22:13 |
| peluse_ | but either way its not a bug - its doing what we designed it to do right now | 22:14 |
| clayg | peluse_: perhaps, i don't quite understand frags without durables are coming from? | 22:14 |
| clayg | peluse_: maybe in-progress replication/reconstruction? | 22:14 |
| peluse_ | clayg, yeah that's what I was trying to get at with the request for exact steps to repro | 22:14 |
| peluse_ | clayg, or nodes were taken offline at just the right time (middle of 2 phase operation) | 22:15 |
| clayg | peluse_: idk, to that argument you could say https://bugs.launchpad.net/swift/+bug/1457691 is not a bug | 22:15 |
| openstack | Launchpad bug 1457691 in OpenStack Object Storage (swift) "EC GET w/not work with just one PUT during handoff" [High,In progress] - Assigned to Bill Huber (wbhuber) | 22:15 |
| clayg | peluse_: but in that case we descovered what we thought was "good enough" wasn't really "good enough" | 22:15 |
| peluse_ | clayg, wrt the last comment I dont think so. We said "good enough for beta" but always said we wanted to handle that sceanrio for production | 22:15 |
| peluse_ | so maybe not a bug but still a feature gap :) | 22:16 |
| peluse_ | eitehr way we need to fix one and not the other IMHO | 22:16 |
| clayg | peluse_: yeah I don't think charles has tooling to achive that kind of fine grained disablement - he's disabling nodes en masse - and finding GET's don't work when it seems like there should still be enough frags | 22:16 |
| clayg | peluse_: ok, I think we should have both bugs and both should be fixed - they are different priorties | 22:17 |
| peluse_ | clayg, but he does say that there are .data's there and if he touched a .durable things worked, right? So somehow really did have .data without the .druable | 22:17 |
| clayg | peluse_: the durable one I think is acctually an unknown prioirty because I don't understand the failure yet | 22:17 |
| peluse_ | clayg, yeah, OK I can get down w/that | 22:17 |
| clayg | peluse_: yesh his find's show that as well - but he didn't *delete* them or anything - they are just not there - very strange | 22:17 |
| peluse_ | clayg, could be some other bizarre bug deep down in there that's cleaning up a .durable somehow when it shouldn't :) | 22:18 |
| clayg | peluse_: yes, could be - i understand it's not as high a priority as others - i'm glad charz is working on it | 22:18 |
| clayg | peluse_: I think there's something there | 22:19 |
| peluse_ | clayg, is he still? | 22:19 |
| peluse_ | working on it that is... | 22:19 |
| clayg | off-and-on ec testing - yes - on that particuarlar issue he may be waiting for more direction | 22:19 |
| clayg | i hadnt' digested his latest notes | 22:19 |
| peluse_ | OK, I'll go add a note. thanks | 22:19 |
| clayg | peluse_: do you have any idea where the lost durables went? | 22:20 |
| clayg | or a theory that could be tested? | 22:20 |
| peluse_ | into the ether maybe? :) | 22:21 |
| clayg | eah maybe - hard to test :\ | 22:21 |
| peluse_ | alpha particles?? | 22:21 |
| peluse_ | I dunno, but with that lens on the problem its worh another look at some of the code down there.... | 22:22 |
| clayg | ok | 22:22 |
| clayg | good call | 22:22 |
| clayg | timeouts! | 22:22 |
| clayg | the 1x was perfect, but after the 20x run there was missing durables right out the gate | 22:23 |
| clayg | I think the proxy is getting board waiting on the 100 on the fsync of the data file and throwing away the connection - so it never gets the finalization mime part | 22:24 |
| peluse_ | could be... | 22:24 |
| *** ChanServ sets mode: +v peluse_ | 22:24 | |
| *** peluse_ is now known as peluse | 22:25 | |
| *** jasondotstar has joined #openstack-swift | 22:28 | |
| *** jordanP has quit IRC | 22:29 | |
| *** NM1 has quit IRC | 22:30 | |
| openstackgerrit | Victor Stinner proposed openstack/swift: Update hacking to 0.10.0 https://review.openstack.org/205977 | 22:32 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Fix pep8 E warning for hacking 0.10 https://review.openstack.org/207237 | 22:32 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Fix pep8 E265 warning of hacking 0.10 https://review.openstack.org/207242 | 22:32 |
| openstackgerrit | Victor Stinner proposed openstack/swift: Fix warning pep8 E128 warning of hacking 0.10 https://review.openstack.org/207243 | 22:32 |
| *** jrichli has quit IRC | 22:32 | |
| clayg | peluse: we'll see if charz can repeat the test and dig into those missing frags just a little bit more | 22:33 |
| clayg | er... s/missing frags/frags missing durables/ | 22:34 |
| haypo | oooh, here you have my new patch serie for hacking 0.10 --^ | 22:34 |
| clayg | haypo: that's per fact man - thanks - checking out now | 22:40 |
| *** flwang1 has joined #openstack-swift | 22:43 | |
| flwang1 | greetings. anybody know how to create a role to support read only? thanks a lot | 22:44 |
| clayg | x-container-read: acl? | 22:44 |
| flwang1 | clayg: does that mean i have to set the acl for each container? | 22:45 |
| *** esker has quit IRC | 22:46 | |
| clayg | yeah that's how an acl would work; i'm not sure if non admin role is supported with current keystone - probably? if not alsmost definately with ho's policy.json patch | 22:46 |
| flwang1 | seems swift doesn't support the policy.json like most the other components did | 22:46 |
| clayg | flwang1: true! | 22:46 |
| *** fds_ has joined #openstack-swift | 22:47 | |
| flwang1 | clayg: ho's policy.json patch??? in reviewing? | 22:47 |
| clayg | yeah, i was just looking for it :\ | 22:47 |
| flwang1 | clayg: the scenario is we would like to have two roles in keystone, both of them are swift only roles, one can r+w, one only can do read | 22:48 |
| notmyname | https://review.openstack.org/#/c/149930/ | 22:48 |
| *** nadeem has joined #openstack-swift | 22:49 | |
| *** lcurtis has quit IRC | 22:50 | |
| flwang1 | notmyname: cool, cheers | 22:50 |
| flwang1 | so seems for now, i have to use the x-container-read | 22:50 |
| flwang1 | but as for the swift only role, like creating swiftoperator in keystone, I think i can just replace the Member role in proxy_server.conf with operator_roles = admin, swiftoperator | 22:51 |
| flwang1 | then it should work, right? | 22:52 |
| ho | flwang1, clayg: yeah, thanks clayg for the review. now i'm stacked for patch 202411 (presupposition of the policy.json patch) :) | 22:52 |
| patchbot | ho: https://review.openstack.org/#/c/202411/ | 22:52 |
| flwang1 | any swift guy can give me a confirmation ? :) | 22:53 |
| ho | flwang1: i think i can help you | 22:53 |
| flwang1 | ho: (10:51:54 AM) flwang1: but as for the swift only role, like creating swiftoperator in keystone, I think i can just replace the Member role in proxy_server.conf with operator_roles = admin, swiftoperator | 22:53 |
| flwang1 | (10:51:59 AM) flwang1: then it should work, right? | 22:53 |
| *** minwoob has quit IRC | 22:54 | |
| ho | flwang1: you are right! | 22:54 |
| flwang1 | ho: awesome, thanks a lot | 22:54 |
| *** nadeem has quit IRC | 22:54 | |
| flwang1 | ho: you're policy.json patch will be really helpful for my 2nd case | 22:54 |
| flwang1 | looking forward to seeing it in | 22:55 |
| *** nadeem has joined #openstack-swift | 22:55 | |
| ho | flwang1: thanks! i try to do my best! | 22:55 |
| ho | flwang1: 2nd case means read only? in this case using container acl as clayg mentioned is only way now. | 22:57 |
| *** flwang1 has left #openstack-swift | 22:57 | |
| *** flwang1 has joined #openstack-swift | 22:59 | |
| clayg | ho: container acl is the only way!? :'( | 22:59 |
| flwang1 | ho: yep, i see | 22:59 |
| clayg | ho: no wonder you're working so hard on policy changes | 22:59 |
| flwang1 | ho: but after merged the policy.json patch | 22:59 |
| clayg | what does the operator_roles buy you? | 22:59 |
| flwang1 | i think we can just create a new role like swift-readonly | 23:00 |
| flwang1 | and then update the policy.json, then all done | 23:00 |
| *** 16WAAEHWP has quit IRC | 23:01 | |
| flwang1 | clayg: the operator_roles can only make sure a role has permission for swift, but it can't get a more granular control, like read, write, isn't it? | 23:02 |
| *** haomaiwang has joined #openstack-swift | 23:02 | |
| *** km has joined #openstack-swift | 23:05 | |
| *** jlhinson has quit IRC | 23:06 | |
| *** blair has quit IRC | 23:07 | |
| ho | clayg: with current logic i think so. swift has three type of role now, reseller_admin_role, operator_roles and no role. if a role has a right to have write access to a container, the role also can have read access to the container. | 23:07 |
| *** blair has joined #openstack-swift | 23:08 | |
| ho | flwang1: i think your approach will be worked soon.. hopefully. :) | 23:12 |
| *** annegentle has joined #openstack-swift | 23:28 | |
| *** cdelatte has joined #openstack-swift | 23:28 | |
| *** delattec has quit IRC | 23:30 | |
| *** hrou has joined #openstack-swift | 23:31 | |
| *** kei_yama has joined #openstack-swift | 23:32 | |
| *** fds_ has quit IRC | 23:32 | |
| *** jasondotstar has quit IRC | 23:33 | |
| *** jamielennox is now known as jamielennox|away | 23:39 | |
| *** nadeem has quit IRC | 23:43 | |
| *** nadeem has joined #openstack-swift | 23:43 | |
| *** jasondotstar has joined #openstack-swift | 23:46 | |
| *** darrenc is now known as darrenc_afk | 23:47 | |
| *** nadeem has quit IRC | 23:50 | |
| *** Kennan2 is now known as Kennan | 23:53 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!