| *** alenavolk has quit IRC | 00:23 | |
| *** alenavolk has joined #openstack-swift | 00:25 | |
| *** kiennt has joined #openstack-swift | 00:38 | |
| kota_ | good morning | 00:41 |
|---|---|---|
| *** tovin07_ has joined #openstack-swift | 00:54 | |
| *** abhitechie has joined #openstack-swift | 00:56 | |
| *** mat128 has joined #openstack-swift | 00:57 | |
| *** alenavolk has quit IRC | 01:09 | |
| *** baojg_ has quit IRC | 01:10 | |
| *** klrmn has quit IRC | 01:13 | |
| *** two_tired has joined #openstack-swift | 01:58 | |
| *** CowboyPr1de has quit IRC | 02:03 | |
| *** alenavolk has joined #openstack-swift | 02:13 | |
| openstackgerrit | Merged openstack/python-swiftclient master: Add support for versionless endpoints https://review.openstack.org/473740 | 02:15 |
| *** klrmn has joined #openstack-swift | 02:20 | |
| *** itlinux has joined #openstack-swift | 03:12 | |
| *** abhinavtechie has joined #openstack-swift | 03:13 | |
| *** baojg has joined #openstack-swift | 03:14 | |
| *** abhitechie has quit IRC | 03:15 | |
| *** abhinavtechie has quit IRC | 03:19 | |
| *** baojg has quit IRC | 03:19 | |
| *** kei_yama has quit IRC | 03:36 | |
| *** gkadam_ has joined #openstack-swift | 03:41 | |
| *** kei_yama has joined #openstack-swift | 03:42 | |
| *** abhitechie has joined #openstack-swift | 03:56 | |
| *** two_tired has quit IRC | 04:00 | |
| *** links has joined #openstack-swift | 04:03 | |
| *** gyee has quit IRC | 04:09 | |
| *** hoonetorg has quit IRC | 04:11 | |
| *** itlinux has quit IRC | 04:19 | |
| *** hoonetorg has joined #openstack-swift | 04:23 | |
| *** abhitechie has quit IRC | 04:38 | |
| *** abhitechie has joined #openstack-swift | 04:38 | |
| *** alenavolk has quit IRC | 04:55 | |
| *** alenavolk has joined #openstack-swift | 04:55 | |
| *** alenavolk has quit IRC | 04:56 | |
| *** alenavolk has joined #openstack-swift | 04:56 | |
| *** alenavolk has quit IRC | 04:56 | |
| *** alenavolk has joined #openstack-swift | 04:57 | |
| *** alenavolk has quit IRC | 04:57 | |
| *** alenavolk has joined #openstack-swift | 04:58 | |
| *** alenavolk has quit IRC | 04:58 | |
| *** alenavolk has joined #openstack-swift | 05:26 | |
| *** alenavolk has quit IRC | 05:32 | |
| *** alenavolk has joined #openstack-swift | 05:32 | |
| *** alenavolk has quit IRC | 05:32 | |
| *** alenavolk has joined #openstack-swift | 05:33 | |
| *** alenavolk has joined #openstack-swift | 05:34 | |
| *** alenavolk has quit IRC | 05:34 | |
| *** tovin07_ has quit IRC | 05:41 | |
| *** tovin07_ has joined #openstack-swift | 05:42 | |
| *** abhitechie has quit IRC | 05:57 | |
| *** cschwede has joined #openstack-swift | 06:07 | |
| *** ChanServ sets mode: +v cschwede | 06:07 | |
| *** abhitechie has joined #openstack-swift | 06:12 | |
| *** mat128 has quit IRC | 06:16 | |
| *** hseipp has joined #openstack-swift | 06:21 | |
| *** hseipp has quit IRC | 06:32 | |
| *** pcaruana has joined #openstack-swift | 06:39 | |
| *** rcernin has joined #openstack-swift | 07:10 | |
| *** bkopilov has quit IRC | 07:21 | |
| *** silor has joined #openstack-swift | 07:21 | |
| *** oshritf has joined #openstack-swift | 07:23 | |
| *** tesseract has joined #openstack-swift | 07:25 | |
| *** bkopilov has joined #openstack-swift | 07:26 | |
| *** geaaru has joined #openstack-swift | 07:31 | |
| *** __david_ has quit IRC | 07:33 | |
| acoles | good morning | 07:42 |
| kota_ | good morning acoles | 07:47 |
| acoles | kota_: hi | 07:47 |
| rledisez | hi | 08:02 |
| rledisez | i was looking at the statsd counter object-server.async_pendings. it is a bit disturbing (even if doc is correct), because it is incremented when a container update fails (or timeouts) AND when an object with X-Delete-At meta is created/deleted | 08:03 |
| rledisez | not sure it's a bug because doc is right, but it is no help when you want to monitor an issue (container update failure) | 08:03 |
| rledisez | should we add a new counter for this specific case or stop incrementing when it's about x-delete-at update | 08:04 |
| rledisez | ? | 08:04 |
| acoles | rledisez: it is incremented when either of those ops fail, correct? (container update, x-delete-at-update) | 08:13 |
| acoles | rledisez: hmm, but also for delete_at_update for a DELETE? (I'm just reading the code) | 08:14 |
| rledisez | acoles: yes, when there is a delete_at_udpate, there is no request sent, it's a async_pendings directly (to avoid more latency i guess) | 08:15 |
| mattoliverau | Does the updater itself push the same metrics? /Me isn't looking at code cause he's at hospital with wife (nothing serious, wife just had day surgery) | 08:19 |
| mattoliverau | Cause there could be a case for a different one there, so you can see the different types of failures and track potential probs | 08:20 |
| *** joeljwright has joined #openstack-swift | 08:22 | |
| *** ChanServ sets mode: +v joeljwright | 08:22 | |
| acoles | async_pendings seems to be incremented whenever an update is pickled which as rledisez says is either because the sync update failed or it's a delete_at_update | 08:24 |
| rledisez | mattoliverau: updater only seems to track successes/errors when treating the async-update fules | 08:24 |
| rledisez | so, no way to tell the difference between a "container update error" or a "x-delete-at job" | 08:25 |
| acoles | The delete_at_update may be because obj server doesn't know the destination for update so defers to the updater which has the container ring loaded??? | 08:25 |
| acoles | rledisez: yeah I can see that it could be usefult o differentiate between the two | 08:26 |
| acoles | useful to* | 08:26 |
| acoles | mattoliverau: o/ hope all goes well there for you | 08:26 |
| mattoliverau | acoles: thanks | 08:28 |
| acoles | rledisez: or at least add another so the async_pendings due to timeouts vs delete_at_updates can be inferred | 08:29 |
| mattoliverau | So something like async_pendings.timeout and async_pendings | 08:37 |
| rledisez | i might propose to move the increment out of the diskfile to have it in container_update() and delete_at_update(). or a new param for the name of the counter? | 08:49 |
| *** hseipp has joined #openstack-swift | 09:15 | |
| acoles | rledisez: might be worth considering adding another metric for one subset and leaving async_pendings as it is - IDK, I am just wondering if changing async_pendings could affect monitoring for someone - at the moment async_pendings does reflect how many updates are pending, just not *why* | 09:32 |
| rledisez | acoles: sure, makes sense | 09:37 |
| *** abhitechie has quit IRC | 09:49 | |
| *** silor has quit IRC | 09:58 | |
| *** tovin07_ has quit IRC | 10:06 | |
| *** sanchitmalhotra has joined #openstack-swift | 10:20 | |
| *** sanchitmalhotra has quit IRC | 10:21 | |
| *** sanchitmalhotra has joined #openstack-swift | 10:27 | |
| *** kei_yama has quit IRC | 10:29 | |
| *** joeljwright has quit IRC | 10:47 | |
| *** kiennt has quit IRC | 11:28 | |
| *** baojg has joined #openstack-swift | 11:52 | |
| *** abhitechie has joined #openstack-swift | 12:19 | |
| *** MVenesio has joined #openstack-swift | 12:20 | |
| *** MVenesio has quit IRC | 12:20 | |
| *** bkopilov has quit IRC | 12:25 | |
| *** bkopilov has joined #openstack-swift | 12:25 | |
| *** gkadam_ has quit IRC | 12:39 | |
| *** lucasxu has joined #openstack-swift | 12:42 | |
| *** mat128 has joined #openstack-swift | 12:47 | |
| *** joeljwright has joined #openstack-swift | 12:47 | |
| *** ChanServ sets mode: +v joeljwright | 12:47 | |
| *** zhurong has joined #openstack-swift | 12:50 | |
| *** gabor_antal_ has joined #openstack-swift | 12:59 | |
| *** hoonetorg has quit IRC | 13:08 | |
| *** hoonetorg has joined #openstack-swift | 13:09 | |
| *** links has quit IRC | 13:18 | |
| *** catintheroof has joined #openstack-swift | 13:19 | |
| *** abhinavtechie has joined #openstack-swift | 13:49 | |
| *** abhitechie has quit IRC | 13:51 | |
| *** zhurong has quit IRC | 14:03 | |
| *** baojg has quit IRC | 14:25 | |
| *** baojg has joined #openstack-swift | 14:26 | |
| *** vint_bra has joined #openstack-swift | 14:34 | |
| *** baojg has quit IRC | 14:37 | |
| *** baojg has joined #openstack-swift | 14:39 | |
| *** baojg has quit IRC | 14:40 | |
| *** baojg has joined #openstack-swift | 14:40 | |
| *** baojg has quit IRC | 14:40 | |
| *** chlong has joined #openstack-swift | 14:41 | |
| *** baojg has joined #openstack-swift | 14:42 | |
| *** baojg has quit IRC | 14:42 | |
| *** itlinux has joined #openstack-swift | 14:49 | |
| timburke | good morning | 14:49 |
| *** SkyRocknRoll has joined #openstack-swift | 15:10 | |
| *** alenavolk has joined #openstack-swift | 15:15 | |
| *** itlinux has quit IRC | 15:15 | |
| *** gyee has joined #openstack-swift | 15:21 | |
| *** tovin07 has joined #openstack-swift | 15:30 | |
| *** saint_ has joined #openstack-swift | 15:44 | |
| *** SkyRocknRoll has quit IRC | 15:55 | |
| *** SkyRocknRoll has joined #openstack-swift | 16:08 | |
| clayg | I'm confused - do x-delete-at meta updates *always* go to async pending? i would expect the update at least attempt to hit the container db inline... | 16:08 |
| rledisez | clayg: as far as i understand the code, it always go through async pendings | 16:11 |
| rledisez | https://github.com/openstack/swift/blob/master/swift/obj/server.py#L429-L434 | 16:11 |
| rledisez | s/always/only/ | 16:11 |
| rledisez | ooops, forget it, i'm doubting now | 16:11 |
| * clayg shrugs | 16:13 | |
| *** tesseract has quit IRC | 16:15 | |
| timburke | that makes sense to me -- otherwise we've got as many as three container update attempts before we send back our 2xx | 16:17 |
| *** saint_ has quit IRC | 16:19 | |
| *** SkyRocknRoll has quit IRC | 16:22 | |
| *** itlinux has joined #openstack-swift | 16:23 | |
| *** SkyRocknRoll has joined #openstack-swift | 16:24 | |
| *** hseipp has quit IRC | 16:26 | |
| *** SkyRocknRoll_ has joined #openstack-swift | 16:33 | |
| *** SkyRocknRoll has quit IRC | 16:34 | |
| *** gkadam_ has joined #openstack-swift | 16:37 | |
| *** rcernin has quit IRC | 16:46 | |
| notmyname | hello, world | 16:47 |
| *** itlinux has quit IRC | 16:48 | |
| *** pcaruana has quit IRC | 16:48 | |
| *** itlinux has joined #openstack-swift | 16:49 | |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient master: Add Python 3.6 classifier & default tox env https://review.openstack.org/497980 | 16:50 |
| openstackgerrit | Tim Burke proposed openstack/python-swiftclient master: Add pypy-devel for RPM-based systems https://review.openstack.org/498869 | 16:50 |
| clayg | @timburke: I really want subtest... I think we should depend on unittest2. | 16:52 |
| clayg | timburke: modernizing tests to use py3 patterns would be an ok thing? What's your opinion? | 16:53 |
| *** links has joined #openstack-swift | 16:54 | |
| timburke | i'm not opposed... though the behavior will depend on the test runner, it seems | 16:59 |
| timburke | http://paste.openstack.org/show/619805/ | 17:03 |
| *** geaaru has quit IRC | 17:07 | |
| *** silor has joined #openstack-swift | 17:12 | |
| *** gkadam_ has quit IRC | 17:15 | |
| *** tovin07 has quit IRC | 17:17 | |
| timburke | still, it degrades gracefully | 17:19 |
| clayg | timburke: that's bullshit! | 17:32 |
| clayg | nose2? | 17:34 |
| *** links has quit IRC | 17:34 | |
| clayg | some recent toy project on ned batchelder (author of nose) is using https://github.com/HypothesisWorks/hypothesis-python#hypothesis | 17:39 |
| *** aagrawal has joined #openstack-swift | 17:40 | |
| openstackgerrit | Merged openstack/swift master: Make X-Backend-Replication consistent for HEAD https://review.openstack.org/498107 | 17:42 |
| *** abhinavtechie has quit IRC | 17:44 | |
| *** catintheroof has quit IRC | 17:49 | |
| *** lucasxu has quit IRC | 18:01 | |
| *** SkyRocknRoll_ has quit IRC | 18:02 | |
| timburke | interesting... i've played around with parameterized tests a bit before... where'd i put that... | 18:18 |
| timburke | https://gist.github.com/tipabu/1a06eafe0a43b29b82f1d806488cdb8a | 18:21 |
| *** catintheroof has joined #openstack-swift | 18:23 | |
| *** cschwede has quit IRC | 18:29 | |
| tdasilva | I remember watching a presentation on hypothesis a while back here at redhat...looked pretty neat | 18:30 |
| *** alenavolk has quit IRC | 18:33 | |
| *** alenavolk has joined #openstack-swift | 18:34 | |
| *** alenavolk has quit IRC | 18:34 | |
| *** alenavolk has joined #openstack-swift | 18:34 | |
| *** alenavolk has quit IRC | 18:35 | |
| *** early has quit IRC | 18:59 | |
| *** early has joined #openstack-swift | 19:02 | |
| *** hseipp has joined #openstack-swift | 19:05 | |
| *** alenavolk has joined #openstack-swift | 19:11 | |
| *** hseipp has quit IRC | 19:17 | |
| *** catintheroof has quit IRC | 19:30 | |
| *** abhinavtechie has joined #openstack-swift | 19:33 | |
| *** catintheroof has joined #openstack-swift | 19:36 | |
| *** aagrawal has quit IRC | 19:36 | |
| *** alenavolk has quit IRC | 19:38 | |
| *** alenavolk has joined #openstack-swift | 19:41 | |
| *** silor has quit IRC | 20:01 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Move listing formatting out to proxy middleware https://review.openstack.org/449394 | 20:45 |
| *** alenavolk has quit IRC | 21:00 | |
| *** joeljwright_ has joined #openstack-swift | 21:05 | |
| *** mat128 has quit IRC | 21:08 | |
| *** joeljwright has quit IRC | 21:08 | |
| *** alenavolk has joined #openstack-swift | 21:24 | |
| *** catintheroof has quit IRC | 21:34 | |
| tdasilva | notmyname: https://www.backblaze.com/blog/hard-drive-failure-stats-q2-2017/ | 21:48 |
| notmyname | tdasilva: thanks. yeah, I read over it this morning | 21:48 |
| notmyname | seems seagate is having a bad time of it lateley :/ | 21:48 |
| tdasilva | interesting bit: "During Q2 we migrated nearly 30 Petabytes of data."...to a remote location | 21:48 |
| openstackgerrit | Joel Wright proposed openstack/swift master: Add Preamble and Postamble to SLO and SegmentedIterable https://review.openstack.org/365371 | 22:05 |
| *** itlinux has quit IRC | 22:14 | |
| *** alenavolk has quit IRC | 22:19 | |
| *** alenavolk has joined #openstack-swift | 22:19 | |
| *** alenavolk has quit IRC | 22:21 | |
| *** vint_bra has quit IRC | 22:23 | |
| *** okdas has quit IRC | 22:35 | |
| *** okdas has joined #openstack-swift | 22:36 | |
| *** okdas has joined #openstack-swift | 22:36 | |
| *** abhinavtechie has quit IRC | 23:01 | |
| *** joeljwright_ has quit IRC | 23:06 | |
| mattoliverau | mornign | 23:12 |
| mattoliverau | *morning | 23:12 |
| *** ianychoi has quit IRC | 23:17 | |
| *** ianychoi has joined #openstack-swift | 23:17 | |
| *** kei_yama has joined #openstack-swift | 23:34 | |
| *** abhitechie has joined #openstack-swift | 23:41 | |
| *** __david_ has joined #openstack-swift | 23:47 | |
| *** nottrobin has quit IRC | 23:59 | |
| *** nottrobin has joined #openstack-swift | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!