| mattoliverau | \o/ - maybe we need to document the table a little better, ie that is exists and how to bring it up :) | 00:01 |
|---|---|---|
| *** sams-gleb has joined #openstack-swift | 00:02 | |
| *** MikeG451 has quit IRC | 00:02 | |
| clayg | mattoliverau: I think there's some hints about the dispersion command - but unless you run the it with --help you don't really know that's it's useful | 00:06 |
| clayg | mattoliverau: a hint on "The Rings" page would be cool - even if it just mostly duplicates the --help output for the command? | 00:06 |
| mattoliverau | +1. I'll take a look post meeting and while sharding this crazy large container (thanks swiftstack) | 00:07 |
| *** sams-gleb has quit IRC | 00:09 | |
| clayg | heh, request_node_count on single replica systems probably defaults pretty low :P | 00:16 |
| clayg | torgomatic: i can't ever quite remember the relationship of the PUT path with request_node_count - is it possible for the same proxy configuration to put replicas deeper into the handoff chain that the read path would get into? | 00:17 |
| *** CaioBrentano has quit IRC | 00:19 | |
| *** tqtran has quit IRC | 00:21 | |
| *** klamath has quit IRC | 00:24 | |
| clayg | hrm... i guess it's sort of interesting - error limiting - handoff logging - but all the node iteration should respect request_node_count - in general | 00:25 |
| *** klamath has joined #openstack-swift | 00:25 | |
| kota_ | good morning | 00:30 |
| *** Suyi has quit IRC | 00:31 | |
| *** thumpba has joined #openstack-swift | 00:59 | |
| mattoliverau | kota_: morning | 01:00 |
| *** sams-gleb has joined #openstack-swift | 01:06 | |
| notmyname | so, silly question... `tox -e func` errors on my new SAIO (but ./.functests is fine). what did I forget to set up? | 01:10 |
| notmyname | oh. ha! nm | 01:13 |
| kota_ | mattoliverau: o/ | 01:14 |
| *** sams-gleb has quit IRC | 01:15 | |
| notmyname | so for the first time since I've ever measured it, there are zero unreviewed patches (http://not.mn/swift/swift_community_dashboard.html) | 01:15 |
| clayg | notmyname: sounds like a bug in the dashboard more than a realistic state frankly | 01:18 |
| notmyname | lol | 01:18 |
| notmyname | not that there is nothing to review. just that every patch that is mergable has at least one comment from a reviewer | 01:19 |
| *** sheel has joined #openstack-swift | 01:21 | |
| * clayg puts on skeptical face | 01:21 | |
| mattoliverau | lol | 01:35 |
| mattoliverau | turns out when my biggest rackspace public cloud test swift cluster's private network stops working sharding doesn't work very well. And annoyingly when dealing with a _large_ container on spinning rusty disks it take a while to do anything espically find the problem and reset to start the tests again :P | 01:39 |
| * mattoliverau wonders how hard his hand will be slapped if he decides to provision SSDs with enough space in the cloud :P | 01:40 | |
| tdasilva | sometimes it is better to ask for forgiveness thank to permission ;) | 01:42 |
| mattoliverau | tdasilva: lol | 01:42 |
| tdasilva | *than | 01:42 |
| clayg | mattoliverau: you can tell 'em pdardeau said it was ok | 01:42 |
| mattoliverau | clayg: ahh, scapegoat... love it | 01:43 |
| *** ouchkernel has quit IRC | 01:43 | |
| *** ouchkernel has joined #openstack-swift | 01:51 | |
| *** CaioBrentano has joined #openstack-swift | 02:10 | |
| *** sams-gleb has joined #openstack-swift | 02:12 | |
| pdardeau | clayg: heh, good luck with that | 02:18 |
| *** sams-gleb has quit IRC | 02:19 | |
| *** SkyRocknRoll has joined #openstack-swift | 02:25 | |
| *** gyee has quit IRC | 02:39 | |
| *** sanchitmalhotra1 has quit IRC | 02:57 | |
| *** sanchitmalhotra has joined #openstack-swift | 02:57 | |
| *** sams-gleb has joined #openstack-swift | 03:16 | |
| *** dmorita has quit IRC | 03:24 | |
| *** sams-gleb has quit IRC | 03:24 | |
| *** dmorita has joined #openstack-swift | 03:26 | |
| *** dmorita has quit IRC | 03:30 | |
| *** sheel has quit IRC | 03:35 | |
| *** sheel has joined #openstack-swift | 03:46 | |
| *** dmorita has joined #openstack-swift | 03:50 | |
| openstackgerrit | Merged openstack/swift: Allow concurrent bulk deletes https://review.openstack.org/311817 | 03:50 |
| *** dmorita has quit IRC | 03:54 | |
| *** links has joined #openstack-swift | 04:01 | |
| clayg | oh snap! | 04:02 |
| *** klrmn has quit IRC | 04:03 | |
| *** ppai has joined #openstack-swift | 04:04 | |
| *** psachin has joined #openstack-swift | 04:07 | |
| *** thumpba has quit IRC | 04:08 | |
| *** sams-gleb has joined #openstack-swift | 04:21 | |
| *** CaioBrentano has quit IRC | 04:24 | |
| openstackgerrit | Merged openstack/swift: Add links to mitaka install guides https://review.openstack.org/320655 | 04:28 |
| clayg | it's a merge party! | 04:28 |
| *** sams-gleb has quit IRC | 04:29 | |
| *** nadeem has joined #openstack-swift | 04:41 | |
| *** dmorita has joined #openstack-swift | 04:41 | |
| openstackgerrit | Matthew Oliver proposed openstack/swift: Add ring-builder dispersion command to admin guide https://review.openstack.org/320775 | 04:44 |
| mattoliverau | clayg: ^^ decided to put it in the admin guide so ops know about it and it's where we give ringbuilder command examples | 04:46 |
| *** dmorita has quit IRC | 04:46 | |
| clayg | mattoliverau: i'll check it out when it builds - i want to see how linkable it is? | 04:48 |
| mattoliverau | cool, if we want it more linkable I guess we just add another sub heading. | 04:48 |
| clayg | mattoliverau: I think i see what you're getting at tho - the rings history doesn't really have any swift-ring-builder commands - and the managing the rings section is probably pretty good reading for new people trying to deploy swift - so there's a good chance folks'll catch that there | 04:50 |
| *** jamielennox is now known as jamielennox|away | 04:50 | |
| mattoliverau | yeah, exactly. overview rings is great as a glossary of terms and history, but if I was an admin I'd look in the admin guide for cool tips, diagnosing dispersion etc. | 04:52 |
| *** jamielennox|away is now known as jamielennox | 05:04 | |
| *** sams-gleb has joined #openstack-swift | 05:26 | |
| *** zaitcev has quit IRC | 05:34 | |
| *** sams-gleb has quit IRC | 05:34 | |
| *** ChubYann has quit IRC | 05:35 | |
| *** chirag has joined #openstack-swift | 05:42 | |
| *** ppai has quit IRC | 05:44 | |
| *** jamielennox is now known as jamielennox|away | 05:54 | |
| *** daemontool has joined #openstack-swift | 05:54 | |
| chirag | In a multinode swift cluster, if a container PUT request is sent to one proxy node and after receiving successful response, with-in few mili seconds a object PUT request is sent via another proxy node. Should it fail with 404 not found error? | 05:55 |
| clayg | i'm trying to imagine a scenario where a container existence check cached a non-existent value and then the PUT missed the cache invalidation because of some inconsistencying the memcache ring | 05:56 |
| clayg | chirag: I would say that it might be possible under failure; but not known to be a common behavior - it shouldn't be stable - if you PUT the container again it's very unlikely you'd be able to continue to observe the failure over the course of a few seconds unless something else is broken in the configuration | 05:59 |
| clayg | like maybe each proxy server only has it's own local memcache in the cache ring? | 06:00 |
| *** ig0r_ has joined #openstack-swift | 06:00 | |
| *** ppai has joined #openstack-swift | 06:02 | |
| chirag | clayg : Yes sending a request again don't fail, but I wonder should why should it fail in my first attempt ? | 06:05 |
| chirag | I too doubt that memcache has not updated globally. | 06:06 |
| clayg | well - how many nodes do you have? are you running seperate memcache pools? | 06:07 |
| chirag | I have 4 a node cluster, no separate memcache pools are running | 06:09 |
| *** jamielennox|away is now known as jamielennox | 06:10 | |
| *** Vinsh has quit IRC | 06:10 | |
| clayg | i wouldn't expect memcache consistency to be that big a problem there really... you sure you have explicitly filled in all four servers in all four proxy-server.conf's on all node? http://docs.openstack.org/developer/swift/deployment_guide.html#memcached-considerations | 06:12 |
| clayg | chirag: in [filter:cache] section you have to set memcache_servers to a comma seperated list of the cluster facing interface for your proxy nodes - and make sure memcache is binding to the cluster facing nic | 06:14 |
| clayg | I think there was a recent fix that allows you to set recheck_[account|container]_existence to 0 at the expense of performance if it's a big issue - you could cirtainly tune it down a fair bit without any big impact | 06:15 |
| clayg | timburke: did the fix with the recheck existence 0 merge? | 06:15 |
| *** ig0r_ has quit IRC | 06:17 | |
| *** jamielennox is now known as jamielennox|away | 06:21 | |
| *** rcernin has joined #openstack-swift | 06:21 | |
| chirag | I am using icehouse release :( will look into it if poosible can change | 06:22 |
| *** jamielennox|away is now known as jamielennox | 06:28 | |
| *** geaaru has joined #openstack-swift | 06:31 | |
| *** nadeem has quit IRC | 06:31 | |
| *** sams-gleb has joined #openstack-swift | 06:31 | |
| *** chirag has quit IRC | 06:36 | |
| *** sams-gleb has quit IRC | 06:40 | |
| *** tesseract has joined #openstack-swift | 06:47 | |
| *** Vinsh has joined #openstack-swift | 06:51 | |
| *** silor has joined #openstack-swift | 06:52 | |
| *** sams-gleb has joined #openstack-swift | 06:58 | |
| *** asettle has joined #openstack-swift | 06:59 | |
| *** SkyRocknRoll has quit IRC | 07:02 | |
| *** hseipp has joined #openstack-swift | 07:23 | |
| *** mmcardle has joined #openstack-swift | 07:35 | |
| *** joeljwright has joined #openstack-swift | 07:41 | |
| *** ChanServ sets mode: +v joeljwright | 07:41 | |
| *** hoonetorg has quit IRC | 07:43 | |
| *** chlong has quit IRC | 07:46 | |
| *** jordanP has joined #openstack-swift | 07:49 | |
| * mattoliverau is calling it a night, night all | 07:51 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Add dispersion --verbose example to admin guide https://review.openstack.org/320818 | 07:53 |
| *** hoonetorg has joined #openstack-swift | 07:56 | |
| *** asettle has quit IRC | 08:01 | |
| *** daemontool_ has joined #openstack-swift | 08:07 | |
| *** daemontool_ has quit IRC | 08:09 | |
| *** daemontool_ has joined #openstack-swift | 08:09 | |
| *** daemontool has quit IRC | 08:10 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/swift: [WIP] Testing latest u-c https://review.openstack.org/318441 | 08:10 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/swift: [WIP] Testing latest u-c https://review.openstack.org/318441 | 08:10 |
| openstackgerrit | Merged openstack/swift: Add ring-builder dispersion command to admin guide https://review.openstack.org/320775 | 08:14 |
| *** ig0r_ has joined #openstack-swift | 08:20 | |
| *** asettle has joined #openstack-swift | 08:22 | |
| *** admin6 has joined #openstack-swift | 08:25 | |
| *** dmk0202 has joined #openstack-swift | 08:28 | |
| *** _fortis has quit IRC | 08:32 | |
| openstackgerrit | Merged openstack/swift: Make SLO manifest copies retain correct content-type https://review.openstack.org/318891 | 08:32 |
| *** acoles_ is now known as acoles | 08:33 | |
| *** Shashikant86 has joined #openstack-swift | 08:41 | |
| *** Shashikant86 has quit IRC | 08:42 | |
| openstackgerrit | Merged openstack/swift: Fix setup of manifest responses in SLO tests https://review.openstack.org/320042 | 08:43 |
| *** Shashikant86 has joined #openstack-swift | 08:43 | |
| *** kei_yama has quit IRC | 08:48 | |
| *** SkyRocknRoll has joined #openstack-swift | 08:51 | |
| *** asettle has quit IRC | 08:58 | |
| *** ouchkernel has quit IRC | 09:02 | |
| *** ouchkernel has joined #openstack-swift | 09:05 | |
| *** daemontool_ has quit IRC | 09:19 | |
| *** daemontool_ has joined #openstack-swift | 09:19 | |
| *** mvk has quit IRC | 09:22 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - test varying proxy pipeline configs https://review.openstack.org/318167 | 09:28 |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - Add "disable_encryption" config option https://review.openstack.org/315588 | 09:29 |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - don't break conditional GETs on unencrypted objects https://review.openstack.org/314740 | 09:32 |
| *** joeljwright1 has joined #openstack-swift | 09:32 | |
| *** joeljwright1 has quit IRC | 09:33 | |
| *** joeljwright has quit IRC | 09:36 | |
| *** ppai has quit IRC | 09:40 | |
| *** Shashikant86 has quit IRC | 09:41 | |
| *** asettle has joined #openstack-swift | 09:42 | |
| *** asettle has quit IRC | 09:42 | |
| *** Shashikant86 has joined #openstack-swift | 09:45 | |
| *** admin6_ has joined #openstack-swift | 09:46 | |
| *** admin6 has quit IRC | 09:48 | |
| *** admin6_ is now known as admin6 | 09:48 | |
| *** Shashikant86 has quit IRC | 09:48 | |
| *** mvk has joined #openstack-swift | 09:55 | |
| *** Shashikant86 has joined #openstack-swift | 09:56 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/swift: [WIP] Testing latest u-c https://review.openstack.org/318441 | 10:10 |
| *** _JZ_ has joined #openstack-swift | 10:41 | |
| *** Shashikant86 has quit IRC | 10:50 | |
| *** sams-gleb has quit IRC | 10:52 | |
| *** sams-gleb has joined #openstack-swift | 10:53 | |
| *** Shashikant86 has joined #openstack-swift | 10:54 | |
| *** sams-gleb has quit IRC | 11:01 | |
| *** sams-gleb has joined #openstack-swift | 11:09 | |
| *** openstackgerrit has quit IRC | 11:18 | |
| *** openstackgerrit has joined #openstack-swift | 11:19 | |
| openstackgerrit | Merged openstack/swift: crypto - test varying proxy pipeline configs https://review.openstack.org/318167 | 11:23 |
| *** Shashikant86 has quit IRC | 11:39 | |
| openstackgerrit | Merged openstack/swift: crypto - Add "disable_encryption" config option https://review.openstack.org/315588 | 11:44 |
| *** jmccarthy has joined #openstack-swift | 11:47 | |
| *** Shashikant86 has joined #openstack-swift | 11:52 | |
| *** Shashikant86 has quit IRC | 11:54 | |
| openstackgerrit | Merged openstack/swift: crypto - don't break conditional GETs on unencrypted objects https://review.openstack.org/314740 | 11:55 |
| *** jamielennox is now known as jamielennox|away | 11:58 | |
| *** hseipp has quit IRC | 12:01 | |
| *** jamielennox|away is now known as jamielennox | 12:07 | |
| *** delatte has quit IRC | 12:12 | |
| *** thumpba has joined #openstack-swift | 12:19 | |
| *** thumpba has quit IRC | 12:19 | |
| *** klamath has quit IRC | 12:24 | |
| *** klamath has joined #openstack-swift | 12:25 | |
| *** jmccarthy has quit IRC | 12:26 | |
| *** hseipp has joined #openstack-swift | 12:26 | |
| *** takashi has joined #openstack-swift | 12:28 | |
| *** jmccarthy has joined #openstack-swift | 12:29 | |
| *** CaioBrentano has joined #openstack-swift | 12:44 | |
| *** SkyRocknRoll has quit IRC | 12:47 | |
| *** asettle has joined #openstack-swift | 12:47 | |
| *** asettle has quit IRC | 12:52 | |
| *** asettle has joined #openstack-swift | 12:54 | |
| *** ppai has joined #openstack-swift | 13:05 | |
| *** ig0r__ has joined #openstack-swift | 13:05 | |
| *** ig0r_ has quit IRC | 13:08 | |
| *** pauloewerton has joined #openstack-swift | 13:12 | |
| *** SkyRocknRoll has joined #openstack-swift | 13:14 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Merge branch 'master' feature/crypto https://review.openstack.org/320432 | 13:24 |
| *** CaioBrentano has quit IRC | 13:24 | |
| *** CaioBrentano has joined #openstack-swift | 13:33 | |
| *** asettle has quit IRC | 13:33 | |
| *** CaioBren_ has joined #openstack-swift | 13:36 | |
| *** CaioBrentano has quit IRC | 13:39 | |
| *** links has quit IRC | 13:48 | |
| *** CaioBrentano has joined #openstack-swift | 13:49 | |
| *** ametts has joined #openstack-swift | 13:51 | |
| *** asettle has joined #openstack-swift | 13:53 | |
| *** CaioBren_ has quit IRC | 13:53 | |
| *** sheel has quit IRC | 14:05 | |
| *** ppai has quit IRC | 14:10 | |
| *** takashi has quit IRC | 14:14 | |
| *** sheel has joined #openstack-swift | 14:16 | |
| *** diogogmt has quit IRC | 14:20 | |
| *** CaioBrentano has quit IRC | 14:24 | |
| *** MikeG451 has joined #openstack-swift | 14:25 | |
| *** CaioBrentano has joined #openstack-swift | 14:25 | |
| *** dmorita has joined #openstack-swift | 14:25 | |
| *** dmorita has quit IRC | 14:30 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Merge branch 'master' feature/crypto https://review.openstack.org/320432 | 14:43 |
| *** takashi has joined #openstack-swift | 14:45 | |
| openstackgerrit | Doron Chen proposed openstack/swift: Added a test for delayed reap. The test schedules a 10-second delayed account reaping. The test checks that no reaping occurs 9 seconds later, and that full reaping occurs 11 seconds later. https://review.openstack.org/321041 | 14:53 |
| *** zaitcev has joined #openstack-swift | 14:54 | |
| *** ChanServ sets mode: +v zaitcev | 14:54 | |
| *** takashi has quit IRC | 14:54 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - purge crypto sysmeta from responses https://review.openstack.org/320579 | 14:55 |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - stop using key id to derive GET path keys https://review.openstack.org/318148 | 14:56 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 15:04 |
| *** arch-nemesis has joined #openstack-swift | 15:05 | |
| *** arch-nemesis has joined #openstack-swift | 15:05 | |
| *** MikeG451 has quit IRC | 15:08 | |
| *** m_kazuhiro has joined #openstack-swift | 15:08 | |
| *** m_kazuhiro has quit IRC | 15:09 | |
| *** m_kazuhiro has joined #openstack-swift | 15:09 | |
| *** m_kazuhiro has quit IRC | 15:10 | |
| CaioBrentano | Is the number of replicas related with the ring "domain"? Replicas in account ring is how many replicas it would be stored of each account db? | 15:15 |
| *** daemontool_ has quit IRC | 15:16 | |
| acoles | CaioBrentano: yes to "Replicas in account ring is how many replicas it would be stored of each account db" | 15:19 |
| *** diogogmt has joined #openstack-swift | 15:19 | |
| CaioBrentano | thanks acoles! is there any downside about having different number of replicas for account, container and object? | 15:22 |
| acoles | CaioBrentano: not that I am aware of. each ring (a, c or o) will have its own level of durability depending on number of replicas chosen. | 15:30 |
| CaioBrentano | acoles nice! thanks! | 15:30 |
| zaitcev | Your durability becomes impossible to calculate because they depend on each other. E.g. accessing object actually checks if account and container exist. | 15:30 |
| zaitcev | Proxy does get_info() on each of them, or at least did last time I checked. | 15:31 |
| *** MikeG451 has joined #openstack-swift | 15:34 | |
| *** klrmn has joined #openstack-swift | 15:37 | |
| *** psachin has quit IRC | 15:47 | |
| *** dmk0202 has quit IRC | 15:49 | |
| *** lakshmiS has joined #openstack-swift | 15:49 | |
| *** esker has joined #openstack-swift | 15:50 | |
| *** nadeem has joined #openstack-swift | 15:54 | |
| *** sams-gleb has quit IRC | 15:56 | |
| *** rcernin has quit IRC | 16:04 | |
| *** links has joined #openstack-swift | 16:10 | |
| *** daemontool has joined #openstack-swift | 16:10 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - remove default root secret (WIP) https://review.openstack.org/318208 | 16:11 |
| *** SkyRocknRoll has quit IRC | 16:13 | |
| *** tesseract has quit IRC | 16:13 | |
| *** klrmn has quit IRC | 16:16 | |
| *** daemontool has quit IRC | 16:20 | |
| *** gyee has joined #openstack-swift | 16:23 | |
| *** esker has quit IRC | 16:23 | |
| *** esker has joined #openstack-swift | 16:24 | |
| *** dmorita has joined #openstack-swift | 16:36 | |
| *** jmccarthy has quit IRC | 16:38 | |
| timburke | clayg: on patch 310216: no, not yet | 16:42 |
| patchbot | timburke: https://review.openstack.org/#/c/310216/ - swift - Make get_info requests useful with recheck_*_exist... | 16:42 |
| *** asettle has quit IRC | 16:46 | |
| notmyname | morning | 16:50 |
| *** Jeffrey4l has quit IRC | 16:51 | |
| *** asettle has joined #openstack-swift | 16:51 | |
| *** lyrrad has joined #openstack-swift | 16:54 | |
| *** sams-gleb has joined #openstack-swift | 16:56 | |
| *** asettle has quit IRC | 17:04 | |
| *** jordanP has quit IRC | 17:08 | |
| *** klrmn has joined #openstack-swift | 17:16 | |
| openstackgerrit | Merged openstack/swift: change etag (currently encypted with object key) to use container key https://review.openstack.org/315372 | 17:17 |
| *** sams-gleb has quit IRC | 17:31 | |
| *** ChubYann has joined #openstack-swift | 17:32 | |
| *** suyash has joined #openstack-swift | 17:41 | |
| *** suyash has quit IRC | 17:42 | |
| *** suyash has joined #openstack-swift | 17:42 | |
| *** vint_bra has joined #openstack-swift | 17:43 | |
| *** vint_bra has left #openstack-swift | 17:47 | |
| *** vint_bra has joined #openstack-swift | 17:47 | |
| *** ig0r__ has quit IRC | 17:51 | |
| *** links has quit IRC | 17:52 | |
| *** asettle has joined #openstack-swift | 18:04 | |
| *** asettle has quit IRC | 18:09 | |
| *** hseipp has quit IRC | 18:11 | |
| openstackgerrit | Merged openstack/swift-specs: Symlink spec updates https://review.openstack.org/242804 | 18:16 |
| *** mvk has quit IRC | 18:20 | |
| *** StraubTW has joined #openstack-swift | 18:23 | |
| *** nadeem has quit IRC | 18:24 | |
| *** nadeem has joined #openstack-swift | 18:26 | |
| *** sams-gleb has joined #openstack-swift | 18:27 | |
| *** mathiasb_ is now known as mathiasb | 18:31 | |
| *** dmorita_ has joined #openstack-swift | 18:53 | |
| *** dmorita has quit IRC | 18:53 | |
| openstackgerrit | Merged openstack/swift: Add dispersion --verbose example to admin guide https://review.openstack.org/320818 | 18:57 |
| clayg | who specs smylink merged? I didn't think that repo worked anymore? | 18:57 |
| clayg | zaitcev: pretty sure read path doesn't bother with container existence check | 18:58 |
| clayg | zaitcev: PUT will of course | 18:58 |
| clayg | zaitcev: GET just serves whatever it finds - doesn't care if the metadata layer thinks it's deleted or w/e | 18:58 |
| zaitcev | clayg: I see... | 18:58 |
| *** sams-gleb has quit IRC | 19:01 | |
| *** mvk has joined #openstack-swift | 19:05 | |
| tdasilva | clayg: my understanding is that we were merging all specs and then 'closing' the repo. | 19:10 |
| clayg | tdasilva: it's all maddness | 19:10 |
| tdasilva | heh | 19:11 |
| tdasilva | let me find notmyname's email again | 19:11 |
| tdasilva | http://lists.openstack.org/pipermail/openstack-dev/2016-May/094026.html | 19:12 |
| timburke | clayg: acls - we need to get_info on the account/container to check that you're even allowed to GET the obj | 19:17 |
| bkeller` | I had thought they were all merged right away but I never actually checked | 19:20 |
| notmyname | clayg: that was the last one that was open. jsut got wedged a while back and didn't get the kick from zuul or something. | 19:26 |
| openstackgerrit | Clay Gerrard proposed openstack/swift: Clarify overload best practices in admin guide https://review.openstack.org/321198 | 19:33 |
| clayg | timburke: well you could try and test it - but I think even if the container_info 404's you can still get the object if you're swift_owner already? | 19:34 |
| tdasilva | my understanding has always been that object read doens't bother with container db. That's how we are able to 'get by' with Swift-on-File, where users might create containers and objects using File access and do a GET on the object. Currently SoF doesn't update ocntainer db if file is written from File access | 19:37 |
| tdasilva | but yeah, acls... | 19:38 |
| *** silor has quit IRC | 19:39 | |
| openstackgerrit | Tim Burke proposed openstack/swift: Make get_info requests useful with recheck_*_existence == 0 https://review.openstack.org/310216 | 19:41 |
| clayg | tdasilva: did you get pyeclib moved over to the openstack namespace? | 19:54 |
| clayg | tdasilva: notmyname: are we in control of relases yet? | 19:54 |
| tdasilva | clayg: yeah | 19:54 |
| notmyname | yeah, I was hoping tdasilva would give us an update at the meeting today | 19:54 |
| clayg | tdasilva: I want to get rid of the stupid log message warnings in my builds - I think timur fixed it upstream but I'd rather build a tagged release than some random sha? | 19:55 |
| clayg | notmyname: FINE | 19:55 |
| notmyname | no, that's just me saying I don't know | 19:55 |
| notmyname | and that's when I was going to ask tdasilva | 19:55 |
| tdasilva | notmyname, clayg: i've had a patch to get the gate builds going for pyeclib but havent had any luck in getting some from infra to look at it: https://review.openstack.org/#/c/317672/ | 19:55 |
| patchbot | tdasilva: patch 317672 - openstack-infra/project-config - add python jobs to pyeclib project | 19:55 |
| notmyname | tdasilva: sounds like the same situation as the rolling upgrade tests patches | 19:56 |
| clayg | why isn't patch 317475 merged!? | 19:56 |
| patchbot | clayg: https://review.openstack.org/#/c/317475/ - swift - Send correct size in POST async update for EC object | 19:56 |
| clayg | tdasilva: so i'm hearing that I'm going to package some random sha because we can't figure out how to cut a release with a merged fix for an annoying bug? | 19:57 |
| tdasilva | clayg: we just don't have automated test. we should be able to send a patch, review and merge it (if we are willing with people saying, it builds on my machine) :P | 19:57 |
| tdasilva | *willing to live* | 19:58 |
| clayg | tdasilva: now we're talking | 19:58 |
| *** sams-gleb has joined #openstack-swift | 19:58 | |
| tdasilva | clayg: https://github.com/openstack/pyeclib | 19:58 |
| clayg | so where are source tarballs published? | 20:00 |
| clayg | i already have a build for 1.2.0 and it doesn't have the fix? | 20:00 |
| clayg | zaitcev: you build packages - do you know how any of this works? | 20:01 |
| zaitcev | clayg: hold on a sec, let me read the scrollback and understand what's up | 20:02 |
| tdasilva | clayg: notmyname is able to cut a release, I imagine once that happens, then they should up in github?? https://github.com/openstack/pyeclib/releases | 20:02 |
| tdasilva | s/should/show | 20:02 |
| notmyname | oh, I can now? you can too, right? | 20:02 |
| clayg | tdasilva: notmyname: YES YES YES! | 20:03 |
| notmyname | is a release simply a tag? | 20:03 |
| notmyname | clayg: you tell me to push a tag somewhere, and I'll push all the tags you want | 20:03 |
| tdasilva | notmyname, honestly I don't know cutting a release works (I imagine it is similar to swift), I just put your name in the release group :P | 20:03 |
| notmyname | ok | 20:03 |
| notmyname | tdasilva: how do you "release" for swiftonfile? | 20:04 |
| tdasilva | I'm sure I could find a doc somewhere | 20:04 |
| openstackgerrit | Tim Burke proposed openstack/swift: Always encrypt etags for container listings https://review.openstack.org/319496 | 20:04 |
| clayg | timur: how you like https://github.com/openstack/pyeclib/commit/855b318042c38a204d3c0ad2d080e62b28cd9f76 for pyeclib 1.2.1 ??? | 20:04 |
| tdasilva | notmyname: ppai does it, he showed me a doc before but I honestly forget, I can look it up again :P | 20:05 |
| notmyname | ok | 20:05 |
| *** sheel has quit IRC | 20:05 | |
| notmyname | so if I'm understanding this right, tdasilva sets up the repos but punts to notmyname and ppai when it comes to releases ;-) | 20:05 |
| acoles | notmyname: I am having trouble with latest merge of master to feature/crypto with the py34 job failing, i think because it fails to install the cryptography lib. can you think of any reason why the recently added other-requirements.txt could cause that? (I removed that file and passed tests) | 20:05 |
| tdasilva | hey hey, I only signed up to get it set up ;) | 20:06 |
| notmyname | acoles: might be some hint in the CI logs. maybe it can't build/install cryptography? do we need to a py3-dev lib or something? | 20:06 |
| acoles | notmyname: i'm clutching at straws, it passed in the check queue once, but has otherwise been failing rechecks. | 20:06 |
| tdasilva | but I'd be glad to come up to speed and figure this out...give me a moment | 20:06 |
| notmyname | oh, that's interesting if it did pass the check queue | 20:06 |
| clayg | notmyname: make timur the boss shot caller | 20:06 |
| acoles | notmyname: we've never failed on feature/crypto, so would the existence of other-requirements.txt perhaps exclude stuff that was installed before but is not now mentioned in other-requirements? | 20:08 |
| notmyname | did the other-requirements.txt just get merged to crypto? | 20:08 |
| tdasilva | love the section title: http://docs.openstack.org/infra/manual/creators.html#make-your-project-useful | 20:08 |
| notmyname | tdasilva: sounds like one of those things that should be an openstack dependency instead of in openstack | 20:09 |
| notmyname | wait, did that sound bitter? | 20:09 |
| tdasilva | hehe | 20:09 |
| *** tqtran has joined #openstack-swift | 20:10 | |
| acoles | notmyname: yes it's included in this latest merge from master | 20:10 |
| notmyname | acoles: ah, ok. so probably we didnt' include some dependency. can you link me some CI logs (or that patch set) that's failing? | 20:10 |
| clayg | tdasilva: it sounds like one of those warnings you ready on a package like "don't try and stick this in your nose" and you're like "wait, wat? did someone try to do that?" | 20:10 |
| tdasilva | and I honestly did ask why we are moving this to openstack infra instead of just moving to github, but I understood that for a lot of folks, they are able to contribute to openstack domain projects but not other projects | 20:10 |
| notmyname | tdasilva: ding ding ding. that's the reason | 20:11 |
| acoles | notmyname: yes. gerrit is being super slow on me right now | 20:11 |
| tdasilva | AFK, be right back | 20:11 |
| clayg | tdasilva: isn't there a law about how every technical decision is at some point beholden to some organizational issue? | 20:12 |
| timburke | acoles: this seems relevant: "openssl/opensslv.h: No such file or directory" - probably need to add libssl-dev? | 20:14 |
| acoles | notmyname: "fatal error: openssl/opensslv.h: No such file or directory" while installing cryptography, http://logs.openstack.org/32/320432/3/check/gate-swift-python34/43d9b84/console.html#_2016-05-25_14_52_47_589 | 20:15 |
| timburke | acoles: yeah, that guy :) | 20:15 |
| acoles | timburke: yeah, that! | 20:15 |
| notmyname | timburke: acoles: so if you propose a patch to crypto and update the other-requirements.txt with libssl-dev (and/or the rhel equiv) then (1) it's self testing in the gate and (2) should get you unblocked for everything else | 20:16 |
| acoles | notmyname: k, other-requirements.txt is not yet on feature/crypto so I'm guessing I need to copy it, add that lib, and patch it onto feature/crypto before I can then merge from master? | 20:19 |
| notmyname | ah, I see. the merge from master isn't working? | 20:19 |
| acoles | yeah :/ | 20:19 |
| acoles | so is other-requirements.txt actually restrictive? (we didn't need to specify libssl-dev anywhere before) | 20:20 |
| notmyname | you could amend the merge commit to include it, I guess. or land it in crypto first | 20:20 |
| acoles | the first is easier :) | 20:20 |
| notmyname | right. other-requirements.txt specifies the packages that will be installed in the base image instead of using the default set of stuff that's from the -infra image | 20:21 |
| notmyname | so it let's us be explicit about our dependencies that aren't installable from pypi (ie the stuff in requirements.txt) | 20:22 |
| *** MikeG451 has quit IRC | 20:25 | |
| acoles | notmyname: got it, thanks | 20:29 |
| *** zul has quit IRC | 20:29 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: Merge branch 'master' feature/crypto https://review.openstack.org/320432 | 20:29 |
| * acoles crosses fingers | 20:30 | |
| *** sams-gleb has quit IRC | 20:30 | |
| *** _JZ_ has quit IRC | 20:42 | |
| *** geaaru has quit IRC | 20:54 | |
| *** jrichli has left #openstack-swift | 20:55 | |
| mattoliverau | Morning | 20:55 |
| *** jrichli has joined #openstack-swift | 20:55 | |
| notmyname | just about meeting time in #openstack-meeting | 20:58 |
| *** asettle has joined #openstack-swift | 20:59 | |
| *** lcurtis has joined #openstack-swift | 21:00 | |
| *** hosanai_ has joined #openstack-swift | 21:00 | |
| *** gyee has quit IRC | 21:07 | |
| *** daemontool has joined #openstack-swift | 21:07 | |
| *** raildo is now known as raildo-afk | 21:09 | |
| *** pauloewerton has quit IRC | 21:17 | |
| *** asettle has quit IRC | 21:23 | |
| *** lakshmiS has quit IRC | 21:23 | |
| *** sams-gleb has joined #openstack-swift | 21:27 | |
| *** ametts has quit IRC | 21:28 | |
| *** daemontool has quit IRC | 21:31 | |
| *** gonzalo2kx has joined #openstack-swift | 21:33 | |
| *** arch-nemesis has quit IRC | 21:37 | |
| *** gonzalo2kx has quit IRC | 21:47 | |
| notmyname | acoles: lurgi? | 21:51 |
| clayg | noic | 21:52 |
| *** arch-nemesis has joined #openstack-swift | 21:52 | |
| *** StraubTW has quit IRC | 21:52 | |
| tdasilva | notmyname: "swift's being used all over the world at large scale in production" but according to EMC it is not ready for enterprise: http://emergingtechblog.emc.com/openstack-swift-object-storage-integration-ecs/ | 21:53 |
| acoles | notmyname: cooties? | 21:53 |
| notmyname | tdasilva: well, that's because they finally solved the "persistent" part of storage ;-) | 21:53 |
| acoles | notmyname: https://en.wikipedia.org/wiki/Cooties | 21:53 |
| notmyname | acoles: https://en.wiktionary.org/wiki/lurgy ? | 21:53 |
| tdasilva | "the platform weaknesses swiftly show up" | 21:53 |
| torgomatic | LOL @ "Further, as you grow across sites you will require ‘strong’ data consistency" | 21:53 |
| tdasilva | who writes this crap | 21:53 |
| torgomatic | they've got one hell of a magic 8-ball over there at EMC | 21:54 |
| clayg | tdasilva: holy shit? | 21:54 |
| notmyname | tdasilva: look at it this way. you've got a major corporation now slinging fud about an open source project. seems like we might be threatening? | 21:54 |
| mattoliverau | acoles: I got lurgi.. must be an English/Australia thing :) | 21:54 |
| clayg | Nisha Narasimhan - i'm looking for you | 21:55 |
| acoles | notmyname: mattoliverau TIL Americans never suffer the dreaded lurgy | 21:55 |
| tdasilva | notmyname, clayg: I need to drive home, can we talk tomorrow about the pyeclib release? | 21:58 |
| acoles | mattoliverau: so do you know the ning nang nong? | 21:59 |
| *** sams-gleb has quit IRC | 22:00 | |
| *** acoles is now known as acoles_ | 22:01 | |
| *** bill_az_ has joined #openstack-swift | 22:02 | |
| mattoliverau | acoles_: where the cows go bong? | 22:02 |
| openstackgerrit | Tim Burke proposed openstack/swift: Always encrypt etags for container listings https://review.openstack.org/319496 | 22:05 |
| openstackgerrit | David Goetz proposed openstack/swift: go: have replicateDevice go routine loop per device https://review.openstack.org/317125 | 22:06 |
| clayg | dfg_: ^ nice! | 22:09 |
| *** hosanai_ has quit IRC | 22:13 | |
| openstackgerrit | David Goetz proposed openstack/swift: go: have replicateDevice go routine loop per device https://review.openstack.org/317125 | 22:14 |
| redbo | dfg_: You fixed the thing I was about to comment on :) | 22:16 |
| dfg_ | was it the commented out line? | 22:16 |
| redbo | Yeah | 22:16 |
| dfg_ | ha :) | 22:16 |
| dfg_ | clayg: ya- thats the biggest change to hbird i've made in a while :) | 22:17 |
| ahale | and you only added 2 new LogError lines ? | 22:21 |
| notmyname | if you don't add LogError lines, there won't be any errors logged to worry about | 22:21 |
| redbo | dfg_: We should build the whole thing with -race and do some benchmarks and make sure it's all happy. | 22:22 |
| *** CaioBrentano has quit IRC | 22:24 | |
| *** diogogmt has quit IRC | 22:27 | |
| *** nadeem has quit IRC | 22:39 | |
| *** nadeem has joined #openstack-swift | 22:41 | |
| *** nadeem has quit IRC | 22:44 | |
| briancline | lol @ the EMC thing | 22:45 |
| briancline | show me one thing that's "enterprise" scale and public cloud friendly that *is* simple to scale to thousands of nodes | 22:46 |
| notmyname | briancline: get out your checkbook and EMC will show you! | 22:46 |
| briancline | lol | 22:47 |
| briancline | and the strong consistency thing... | 22:49 |
| briancline | in the words of peluse, okay, sure | 22:49 |
| briancline | there goes the performance of the entire system | 22:49 |
| *** gyee has joined #openstack-swift | 22:55 | |
| notmyname | http://eavesdrop.openstack.org/meetings/swift/2014/swift.2014-08-06-19.01.log.html | 23:00 |
| notmyname | meeting log from the TC gap analysis of swift (the swift meeting after the TC meeting) | 23:00 |
| *** chlong has joined #openstack-swift | 23:05 | |
| *** lcurtis has quit IRC | 23:14 | |
| notmyname | here's an interesting blog post http://ttx.re/stepping-out-of-the-way.html | 23:15 |
| *** Jeffrey4l has joined #openstack-swift | 23:20 | |
| *** arch-nemesis has quit IRC | 23:22 | |
| briancline | nice post. | 23:33 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!