| *** vint_bra has quit IRC | 00:21 | |
| openstackgerrit | Merged openstack/swift master: api-ref: update docs links https://review.openstack.org/499293 | 00:22 |
|---|---|---|
| *** baojg_ has quit IRC | 00:37 | |
| *** tovin07_ has joined #openstack-swift | 00:37 | |
| *** alenavolk has joined #openstack-swift | 00:41 | |
| *** kiennt has joined #openstack-swift | 00:45 | |
| *** alenavolk has quit IRC | 01:00 | |
| *** alenavolk has joined #openstack-swift | 01:01 | |
| *** alenavolk has quit IRC | 01:01 | |
| *** alenavolk has joined #openstack-swift | 01:01 | |
| *** alenavolk has joined #openstack-swift | 01:02 | |
| *** alenavolk has quit IRC | 01:02 | |
| *** alenavolk has joined #openstack-swift | 01:03 | |
| *** mat128 has joined #openstack-swift | 01:04 | |
| *** alenavolk has quit IRC | 01:08 | |
| *** alenavolk has joined #openstack-swift | 01:08 | |
| *** alenavolk has quit IRC | 01:08 | |
| *** alenavolk has joined #openstack-swift | 01:09 | |
| *** alenavolk has quit IRC | 01:09 | |
| *** alenavolk has joined #openstack-swift | 01:10 | |
| *** chsc has quit IRC | 01:11 | |
| tdasilva | notmyname: is the idea of https://bugs.launchpad.net/swift/+bug/1581146 that one day maybe we could instead return the sha of the object? | 01:12 |
| openstack | Launchpad bug 1581146 in OpenStack Object Storage (swift) "add content-md5 header" [Wishlist,New] | 01:12 |
| notmyname | tdasilva: not quite | 01:13 |
| notmyname | tdasilva: and standards, being what they are, have moved on | 01:13 |
| notmyname | tdasilva: unfortunately, the http standard has dropped the content-md5 header | 01:13 |
| notmyname | but, thats' not really what it's about | 01:13 |
| notmyname | what it's really about is to separate a checksum and an etag | 01:14 |
| notmyname | so we can have an etag that we can define (and change) as needed. IIRC the http definition is simply "something unique to the content so that same content == same etag" | 01:14 |
| notmyname | and we can do the bit flip checks in a different header. which, yes, could eventually be something other than md5 | 01:15 |
| tdasilva | just noticed google computes both md5 and crc32: https://cloud.google.com/storage/docs/hashes-etags | 01:16 |
| notmyname | that seems pretty smart | 01:18 |
| *** catintheroof has quit IRC | 02:00 | |
| *** catintheroof has joined #openstack-swift | 02:01 | |
| *** itlinux has joined #openstack-swift | 02:12 | |
| *** thegreenhundred has joined #openstack-swift | 02:31 | |
| kota_ | hello | 02:44 |
| *** abhitechie has quit IRC | 02:51 | |
| *** alenavolk has joined #openstack-swift | 02:59 | |
| *** alenavolk has quit IRC | 03:02 | |
| *** two_tired has joined #openstack-swift | 03:14 | |
| *** links has joined #openstack-swift | 03:32 | |
| *** abhitechie has joined #openstack-swift | 03:40 | |
| *** psachin has joined #openstack-swift | 03:41 | |
| mattoliverau | afternoon all | 03:41 |
| *** gkadam_ has joined #openstack-swift | 03:41 | |
| *** vint_bra has joined #openstack-swift | 03:42 | |
| mahatic | good morning | 03:59 |
| kota_ | mattoliverau, mahatic: o/ | 04:17 |
| mattoliverau | mahatic: morning | 04:17 |
| mahatic | mattoliverau: kota_ hello | 04:21 |
| *** two_tired has quit IRC | 04:21 | |
| *** gyee has quit IRC | 04:39 | |
| *** vint_bra has quit IRC | 04:39 | |
| *** catintheroof has quit IRC | 04:41 | |
| *** klrmn has quit IRC | 04:44 | |
| *** abhinavtechie has joined #openstack-swift | 04:55 | |
| *** abhitechie has quit IRC | 04:59 | |
| *** hoonetorg has quit IRC | 05:05 | |
| *** mat128 has quit IRC | 05:07 | |
| *** hoonetorg has joined #openstack-swift | 05:22 | |
| *** chsc has joined #openstack-swift | 05:24 | |
| *** skudlik has joined #openstack-swift | 05:34 | |
| *** chsc has quit IRC | 05:37 | |
| *** aagrawal has joined #openstack-swift | 05:39 | |
| *** abhinavtechie has quit IRC | 05:43 | |
| *** links has quit IRC | 05:50 | |
| *** silor has joined #openstack-swift | 05:59 | |
| *** links has joined #openstack-swift | 06:01 | |
| *** pcaruana has joined #openstack-swift | 06:17 | |
| *** rcernin has joined #openstack-swift | 06:33 | |
| *** cbartz has joined #openstack-swift | 06:40 | |
| *** hseipp has joined #openstack-swift | 06:52 | |
| *** abhinavtechie has joined #openstack-swift | 07:09 | |
| *** tesseract has joined #openstack-swift | 07:12 | |
| *** aagrawal has quit IRC | 07:12 | |
| *** d0ugal has quit IRC | 07:20 | |
| *** cschwede has joined #openstack-swift | 07:26 | |
| *** ChanServ sets mode: +v cschwede | 07:26 | |
| *** kei_yama has joined #openstack-swift | 07:39 | |
| *** lifeless_ is now known as lifeless | 07:39 | |
| *** geaaru has joined #openstack-swift | 07:41 | |
| *** itlinux has quit IRC | 07:42 | |
| *** kiennt has quit IRC | 07:50 | |
| *** kiennt has joined #openstack-swift | 07:51 | |
| *** hseipp has quit IRC | 08:06 | |
| *** joeljwright has joined #openstack-swift | 08:36 | |
| *** ChanServ sets mode: +v joeljwright | 08:36 | |
| *** joeljwright has quit IRC | 08:39 | |
| *** joeljwright has joined #openstack-swift | 08:56 | |
| *** ChanServ sets mode: +v joeljwright | 08:56 | |
| *** tovin07_ has quit IRC | 08:59 | |
| *** psachin has quit IRC | 09:00 | |
| *** kei_yama has quit IRC | 09:01 | |
| *** abhinavtechie has quit IRC | 10:17 | |
| *** d0ugal has joined #openstack-swift | 10:18 | |
| *** abhitechie has joined #openstack-swift | 10:24 | |
| *** abhitechie has quit IRC | 10:34 | |
| *** psachin has joined #openstack-swift | 10:45 | |
| *** mat128 has joined #openstack-swift | 11:35 | |
| *** kiennt has quit IRC | 11:43 | |
| *** abhitechie has joined #openstack-swift | 11:58 | |
| *** catintheroof has joined #openstack-swift | 12:17 | |
| *** gkadam_ has quit IRC | 12:27 | |
| *** catintheroof has quit IRC | 12:34 | |
| *** catintheroof has joined #openstack-swift | 12:34 | |
| *** psachin has quit IRC | 12:46 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 12:48 |
| *** oshritf has joined #openstack-swift | 12:56 | |
| *** oshritf has quit IRC | 13:01 | |
| *** thegreenhundred has quit IRC | 13:02 | |
| *** abhinavtechie has joined #openstack-swift | 13:10 | |
| *** abhitechie has quit IRC | 13:13 | |
| *** catintheroof has quit IRC | 13:16 | |
| *** catintheroof has joined #openstack-swift | 13:16 | |
| *** links has quit IRC | 13:44 | |
| *** joeljwright has quit IRC | 13:59 | |
| *** baojg has joined #openstack-swift | 14:04 | |
| *** catintheroof has quit IRC | 14:04 | |
| *** catintheroof has joined #openstack-swift | 14:05 | |
| *** catintheroof has quit IRC | 14:15 | |
| *** catintheroof has joined #openstack-swift | 14:15 | |
| *** joeljwright has joined #openstack-swift | 14:16 | |
| *** ChanServ sets mode: +v joeljwright | 14:16 | |
| *** gyee has joined #openstack-swift | 14:24 | |
| *** catintheroof has quit IRC | 14:25 | |
| *** catintheroof has joined #openstack-swift | 14:25 | |
| *** catintheroof has quit IRC | 14:35 | |
| *** catintheroof has joined #openstack-swift | 14:36 | |
| *** catintheroof has quit IRC | 14:37 | |
| *** catintheroof has joined #openstack-swift | 14:37 | |
| *** vint_bra has joined #openstack-swift | 14:39 | |
| openstackgerrit | Alistair Coles proposed openstack/swift master: Cleanup test tempdirs in tearDown https://review.openstack.org/500078 | 14:43 |
| *** catintheroof has quit IRC | 14:52 | |
| *** catintheroof has joined #openstack-swift | 14:52 | |
| *** abhinavtechie has quit IRC | 14:55 | |
| *** abhinavtechie has joined #openstack-swift | 14:55 | |
| *** catintheroof has quit IRC | 15:00 | |
| *** catintheroof has joined #openstack-swift | 15:00 | |
| *** EmilienM has quit IRC | 15:12 | |
| *** EmilienM has joined #openstack-swift | 15:12 | |
| *** links has joined #openstack-swift | 15:18 | |
| *** aagrawal has joined #openstack-swift | 15:19 | |
| *** skudlik has quit IRC | 15:20 | |
| *** thegreenhundred has joined #openstack-swift | 15:22 | |
| *** abhinavtechie has quit IRC | 15:23 | |
| *** chsc has joined #openstack-swift | 15:24 | |
| *** chsc has joined #openstack-swift | 15:24 | |
| *** rcernin has quit IRC | 15:29 | |
| *** pcaruana has quit IRC | 15:30 | |
| timburke | good morning | 15:46 |
| notmyname | good morning | 15:56 |
| *** itlinux has joined #openstack-swift | 16:05 | |
| *** joeljwright has quit IRC | 16:23 | |
| *** links has quit IRC | 16:31 | |
| *** catintheroof has quit IRC | 16:36 | |
| *** catintheroof has joined #openstack-swift | 16:36 | |
| clayg | tdasilva: I'm not logged in to gerrit (on the train) can you +A https://review.openstack.org/#/c/500078/ for me? | 16:49 |
| tdasilva | clayg: done | 16:49 |
| patchbot | patch 500078 - swift - Cleanup test tempdirs in tearDown | 16:49 |
| clayg | Thanks! | 16:49 |
| tdasilva | np :) | 16:49 |
| tdasilva | now looking at p 449394 | 16:50 |
| patchbot | https://review.openstack.org/#/c/449394/ - swift - Move listing formatting out to proxy middleware | 16:50 |
| *** catintheroof has quit IRC | 16:50 | |
| *** oshritf has joined #openstack-swift | 16:51 | |
| timburke | oh, cool! thanks tdasilva! i think it's a decent idea? feels like it should clean up some things for encryption/symlinks/sharding (and whatever other crazy thing we think of next) | 16:54 |
| tdasilva | timburke: yeah, i like the idea of removing all that xml,json handling from the different middlewares | 16:55 |
| tdasilva | plus it standardizes proxy <-> backend communications a bit | 16:56 |
| openstackgerrit | Merged openstack/swift master: Cleanup test tempdirs in tearDown https://review.openstack.org/500078 | 16:56 |
| timburke | i was a little sad that it didn't result in more lines deleted... *shrug* | 16:56 |
| *** oshritf has quit IRC | 16:59 | |
| tdasilva | timburke: yeah, i was surprised by that too, but it still cleans up the code nicely. The part i'm not a big fan of is the pipeline change. I wish we just auto-added all the required middlewares without actually requiring the pipeline change | 16:59 |
| timburke | it's got auto-insertion in https://review.openstack.org/#/c/449394/8/swift/proxy/server.py (otherwise there isn't a good upgrade path) -- or did you mean something else? | 17:01 |
| patchbot | patch 449394 - swift - Move listing formatting out to proxy middleware | 17:01 |
| tdasilva | 'catch_errors, gatekeeper, healthcheck proxy-loggin cache listing_formats' should always be there, no? | 17:01 |
| tdasilva | timburke: something else, i understand we auto-insert, but i mean we should auto-insert like it already is, plus actually remove it from the pipeline config. it would help simplify the pipeline IMO | 17:02 |
| timburke | mmm. yeah, set some restrictions for people -- "you can have custom middleware, sure, but it has to be somewhere here in the middle, around the auth-y bits" | 17:04 |
| tdasilva | yeah, we've had different conversations before about simplifying the pipeline, but there's no easy solution... | 17:05 |
| *** klrmn has joined #openstack-swift | 17:12 | |
| *** abhinavtechie has joined #openstack-swift | 17:21 | |
| *** thegreenhundred has quit IRC | 17:24 | |
| *** aagrawal has quit IRC | 17:25 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Always require device dir for containers https://review.openstack.org/458070 | 17:33 |
| *** geaaru has quit IRC | 17:34 | |
| notmyname | I added a new section to the top of our PTG etherpad https://etherpad.openstack.org/p/swift-ptg-queens | 17:43 |
| notmyname | it's for stuff that other teams are doing that we should know about and participate in | 17:43 |
| notmyname | this will be especially helpful for monday and tuesday | 17:43 |
| notmyname | and just this morning I heard about something the interop working group (the old "defcore") group is doing on monday that we should participate in | 17:44 |
| notmyname | something about setting up official testing programs (for trademarks) for stand-alone openstack services | 17:44 |
| *** cbartz has quit IRC | 17:59 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Use check_drive consistently https://review.openstack.org/500158 | 18:15 |
| clayg | timburke: ^ now that was a yak! | 18:15 |
| notmyname | https://skeptics.stackexchange.com/questions/28217/do-yaks-get-shaved | 18:17 |
| timburke | way to shear, clayg! | 18:21 |
| notmyname | clayg: is this one for testing purposes? https://review.openstack.org/#/c/500158/1/swift/common/base_storage_server.py | 18:21 |
| patchbot | patch 500158 - swift - Use check_drive consistently | 18:21 |
| clayg | I think I really do *finally* understand the upper-left hand side of gerrit - I think sometimes it changes the tab based on some context dependent situation - it's funny tho because it like made me not trust my reality | 18:21 |
| clayg | that should... probably fail pep8? | 18:21 |
| notmyname | I don't remember if we have "unused imports" turned on | 18:22 |
| timburke | pretty sure we do | 18:22 |
| notmyname | what is this? golang? | 18:22 |
| clayg | *my* bad - flake8 | 18:22 |
| timburke | pycodestyle? | 18:24 |
| timburke | "the tool formerly known as pep8" | 18:24 |
| notmyname | timburke: seriously? | 18:24 |
| clayg | "renamed to pycodestyle to reduce confusion" | 18:24 |
| clarkb | I think there were enough people complaining that their rules don't all follow pep8 that they changed. But also peps are peps and programs aren't | 18:25 |
| notmyname | I will now call everyone "steve", to reduce (my) confusion | 18:25 |
| *** itlinux has quit IRC | 18:30 | |
| clayg | it makes sense - but we're super old and crotchety - we self identify as perpetually annoyed neckbeards - my hope is that most folks can come to forgive the fleeting/knee-jerk "stop moving the cheese" reaction when they realize we don't *mean* anything by it... | 18:30 |
| clayg | I guess my rebase workflow has a gap for flake8's - the first patch in the change is ok... which is good - the follow-up is just to answer the question "shouldn't we use check_device *everywhere*" "yes, but it's a bunch of annoying tedius work of dubious value" | 18:32 |
| *** mat128 has quit IRC | 20:06 | |
| *** mat128 has joined #openstack-swift | 20:06 | |
| *** mat128 has quit IRC | 20:10 | |
| *** silor has quit IRC | 20:13 | |
| *** rcernin has joined #openstack-swift | 20:31 | |
| *** rcernin has quit IRC | 20:55 | |
| *** catintheroof has joined #openstack-swift | 20:57 | |
| *** catinthe_ has joined #openstack-swift | 20:58 | |
| *** catintheroof has quit IRC | 21:01 | |
| zaitcev | Guys... I have a working cluster that constantly says Timeout getting a connection to memcached: 192.168.50.2:11211 | 21:01 |
| notmyname | zaitcev: could be a connection limit? | 21:02 |
| clayg | get a faster memcached? | 21:02 |
| zaitcev | Looking at netstat, it connects there fine, most of the time. Is there a magic number of workers that I can bump? | 21:02 |
| zaitcev | But it's already at 1024 | 21:02 |
| notmyname | something more esoteric like too many client connections in the proxy causing eventlet hub starvation? | 21:04 |
| clayg | any crazy middleware might be doing some blocking business? I think our memcache connect timeout is only like 300ms - you can't make too many LDAP calls into C-land in the main thread before you are cursing "cooperative" yields | 21:05 |
| clayg | maybe step zero is... does it go away if you make the connection timeout 10s? | 21:05 |
| clayg | sometimes slow but working is better than not working... sometimes not working is better than working but slow... YMMV | 21:06 |
| zaitcev | Well, I have 2, so it's not like tokens get rejected. Or at least I don't see any 401s | 21:08 |
| zaitcev | For some reason local memcached always, 100% works on any given proxy node. Only times out with remote ones. | 21:08 |
| clayg | ugh - that *does* sound familiar! | 21:11 |
| clayg | https://www.youtube.com/watch?v=iqEpoi00_Ws&feature=youtu.be&t=128 | 21:14 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Use check_drive consistently https://review.openstack.org/500158 | 21:15 |
| zaitcev | It appears that setting the connection_timeout to 1.2 cured the problem. I don't know why 0.3 is small on a 10GE pipe but maybe it's an overloaded switch or kernel is just being dumb with CPU scheduling. | 21:22 |
| clayg | bandwidth isn't the same as latency - networking can be weird... | 21:23 |
| *** aagrawal has joined #openstack-swift | 21:24 | |
| *** chsc has quit IRC | 21:24 | |
| *** abhinavtechie has quit IRC | 21:27 | |
| *** gyee has quit IRC | 21:46 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Fix memcached exception out of range stacktrace https://review.openstack.org/219963 | 21:58 |
| openstackgerrit | Tim Burke proposed openstack/swift master: Fix memcached exception out of range stacktrace https://review.openstack.org/219963 | 22:00 |
| *** tesseract has quit IRC | 22:09 | |
| *** catinthe_ has quit IRC | 22:54 | |
| *** catintheroof has joined #openstack-swift | 23:03 | |
| *** thegreenhundred has joined #openstack-swift | 23:16 | |
| *** vint_bra has quit IRC | 23:17 | |
| *** thegreenhundred has quit IRC | 23:24 | |
| *** gyee has joined #openstack-swift | 23:29 | |
| *** catintheroof has quit IRC | 23:51 | |
| *** gyee has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!