| clayg | timur: it's an interesting idea IMHO, i left some feedback - it's obvious it was just a POC - but it might be easier to reason about if you go deeper describing the use-case (commit message or review comments)? | 00:00 |
|---|---|---|
| *** mwheckmann has quit IRC | 00:14 | |
| timur | clayg: thanks for glancing at it. About gating, I wasn't sure how to tell the difference between a request coming from a proxy node vs a request coming in internally. Agreed that exposing this outward is probably not a good idea. I read the suggestion of X-Backend- is that the preferred way to do such things? | 00:16 |
| timur | I'll keep expanding the change, but wanted to validate that this is roughly a reasonable direction | 00:16 |
| *** gyee has quit IRC | 00:20 | |
| *** itlinux has quit IRC | 00:29 | |
| *** lyrrad has quit IRC | 00:34 | |
| kota_ | good morning | 00:46 |
| kota_ | acoles: thanks for adding my review into summit discussion^^ | 00:47 |
| *** baojg has joined #openstack-swift | 00:54 | |
| *** itlinux has joined #openstack-swift | 00:58 | |
| kota_ | hew! looking at the back log of my irc bouncer, great information to download specific version of gerrit review (that's in between acoles and notmyname) | 01:03 |
| *** trifon has quit IRC | 01:06 | |
| *** klrmn has quit IRC | 01:12 | |
| *** takashi has joined #openstack-swift | 01:16 | |
| takashi | good morning | 01:17 |
| *** Jeffrey4l__ has joined #openstack-swift | 01:24 | |
| *** dmorita has quit IRC | 01:38 | |
| *** dmorita has joined #openstack-swift | 01:40 | |
| *** forcebanana has quit IRC | 01:57 | |
| *** asettle has quit IRC | 01:57 | |
| *** dmorita has quit IRC | 01:58 | |
| *** dmorita has joined #openstack-swift | 02:00 | |
| *** dmorita has quit IRC | 02:00 | |
| *** dmorita has joined #openstack-swift | 02:00 | |
| *** vinsh_ has joined #openstack-swift | 02:05 | |
| *** vinsh has quit IRC | 02:05 | |
| *** asettle has joined #openstack-swift | 02:05 | |
| *** baojg has quit IRC | 02:06 | |
| *** itlinux has quit IRC | 02:06 | |
| *** bapalm has quit IRC | 02:06 | |
| *** baojg has joined #openstack-swift | 02:09 | |
| *** klrmn has joined #openstack-swift | 02:11 | |
| *** M4urice has quit IRC | 02:12 | |
| *** bapalm has joined #openstack-swift | 02:13 | |
| *** hurricanerix has quit IRC | 02:13 | |
| *** tristanC has quit IRC | 02:13 | |
| *** wanghua has joined #openstack-swift | 02:13 | |
| *** Guest88568 has quit IRC | 02:13 | |
| *** tristanC has joined #openstack-swift | 02:14 | |
| *** Jeffrey4l__ has quit IRC | 02:20 | |
| *** scott has joined #openstack-swift | 02:27 | |
| *** hurricanerix has joined #openstack-swift | 02:27 | |
| *** scott is now known as Guest42678 | 02:28 | |
| *** trifon has joined #openstack-swift | 02:28 | |
| *** trifon has quit IRC | 02:33 | |
| *** Jeffrey4l__ has joined #openstack-swift | 02:33 | |
| *** trifon has joined #openstack-swift | 02:35 | |
| *** citytiming has joined #openstack-swift | 02:37 | |
| *** trifon has quit IRC | 02:41 | |
| *** Jeffrey4l__ has quit IRC | 02:45 | |
| *** citytiming has quit IRC | 02:56 | |
| *** asettle has quit IRC | 03:01 | |
| *** dmorita has quit IRC | 03:03 | |
| *** furlongm has quit IRC | 03:07 | |
| *** links has joined #openstack-swift | 03:20 | |
| *** asettle has joined #openstack-swift | 03:33 | |
| *** asettle has quit IRC | 03:38 | |
| *** furlongm has joined #openstack-swift | 03:45 | |
| *** treaki__ has joined #openstack-swift | 03:57 | |
| *** asettle has joined #openstack-swift | 03:58 | |
| *** treaki_ has quit IRC | 04:01 | |
| *** sheel has joined #openstack-swift | 04:12 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Add Pros/Cons table for global clusetr consieration https://review.openstack.org/304382 | 04:12 |
| kota_ | takashi: hello | 04:12 |
| kota_ | ugh, typo in the patch title :/ | 04:13 |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift: Add Pros/Cons table for global clusetr consideration https://review.openstack.org/304382 | 04:13 |
| takashi | kota_: o/ | 04:31 |
| *** nadeem has joined #openstack-swift | 04:49 | |
| *** rcernin has joined #openstack-swift | 04:52 | |
| *** dmorita has joined #openstack-swift | 04:56 | |
| *** dmorita has quit IRC | 05:01 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:06 | |
| *** ppai has joined #openstack-swift | 05:13 | |
| *** silor has joined #openstack-swift | 05:18 | |
| *** zaitcev has quit IRC | 05:19 | |
| *** silor1 has joined #openstack-swift | 05:23 | |
| *** silor has quit IRC | 05:27 | |
| *** silor1 is now known as silor | 05:27 | |
| *** asettle has quit IRC | 05:27 | |
| *** asettle has joined #openstack-swift | 05:30 | |
| *** tamizh_geek has quit IRC | 05:33 | |
| mattoliverau | pdardeau: my parents are over (from interstate) visiting their new granddaughter, so am currently in a holiday house, away from the world. Having the week "off" work, but will be online from time to time to start preparing and doing community work in the hopes to be useful when in Austin ;) | 05:35 |
| mattoliverau | internet here is mostly non-existant.. so will do what I can ;) | 05:36 |
| *** tamizh_geek has joined #openstack-swift | 05:39 | |
| kota_ | mattoliverau: enjoy your time! | 05:45 |
| *** baojg has quit IRC | 05:45 | |
| *** baojg has joined #openstack-swift | 05:46 | |
| *** jethrosun has quit IRC | 05:51 | |
| *** jethrosun has joined #openstack-swift | 05:51 | |
| *** jethrosun has quit IRC | 05:52 | |
| *** jethrosun has joined #openstack-swift | 05:52 | |
| *** natarej_ has joined #openstack-swift | 05:54 | |
| *** natarej has quit IRC | 05:57 | |
| *** haomaiwang has joined #openstack-swift | 05:59 | |
| *** haomaiwang has quit IRC | 05:59 | |
| *** tesseract has joined #openstack-swift | 06:24 | |
| *** tesseract is now known as Guest15115 | 06:24 | |
| *** SkyRocknRoll has quit IRC | 06:25 | |
| *** asettle has quit IRC | 07:00 | |
| mattoliverau | kota_: thanks man | 07:03 |
| *** mingdang1 has quit IRC | 07:05 | |
| *** jaakkos has quit IRC | 07:07 | |
| *** jaakkos has joined #openstack-swift | 07:08 | |
| *** mmcardle has joined #openstack-swift | 07:10 | |
| *** klrmn has quit IRC | 07:14 | |
| *** SkyRocknRoll has joined #openstack-swift | 07:16 | |
| *** rledisez has joined #openstack-swift | 07:19 | |
| *** nadeem has quit IRC | 07:21 | |
| *** nadeem has joined #openstack-swift | 07:28 | |
| *** asettle has joined #openstack-swift | 07:30 | |
| ntt | Hi, is there any script for automatic distribution of rings across all nodes in a swift cluster? | 07:33 |
| *** asettle has quit IRC | 07:34 | |
| *** jmccarthy has quit IRC | 07:36 | |
| *** jmccarthy has joined #openstack-swift | 07:37 | |
| *** geaaru has joined #openstack-swift | 07:41 | |
| *** jamielennox is now known as jamielennox|away | 07:42 | |
| *** trifon has joined #openstack-swift | 07:46 | |
| *** daemontool_ has joined #openstack-swift | 07:50 | |
| *** siva_krishnan_ has joined #openstack-swift | 07:50 | |
| *** daemontool has quit IRC | 07:53 | |
| *** siva_krishnan_ has quit IRC | 08:00 | |
| *** siva_krishnan_ has joined #openstack-swift | 08:01 | |
| *** siva_krishnan_ has quit IRC | 08:03 | |
| *** siva_krishnan_ has joined #openstack-swift | 08:04 | |
| *** oshritf has joined #openstack-swift | 08:07 | |
| *** siva_krishnan_ has quit IRC | 08:11 | |
| *** siva_krishnan_ has joined #openstack-swift | 08:13 | |
| *** oshritf has quit IRC | 08:13 | |
| *** oshritf has joined #openstack-swift | 08:14 | |
| *** siva_kri- has joined #openstack-swift | 08:15 | |
| *** siva_krishnan_ has quit IRC | 08:17 | |
| *** baojg has quit IRC | 08:21 | |
| *** baojg has joined #openstack-swift | 08:22 | |
| ahale | ntt: we use https://github.com/pandemicsyn/swift-ring-master , or sometimes just scp | 08:22 |
| ntt | thank you ahale. can I ask you some questions related to replicas? | 08:24 |
| ahale | sure | 08:24 |
| *** jistr has joined #openstack-swift | 08:25 | |
| ntt | actually I have a cluster with only 1replica, 1 node, 1 zone and 4 devices but I'd like to add more replicas and zones | 08:25 |
| ntt | on the storage node I have 4 devices with a weight of 5000 for each one | 08:25 |
| ahale | (just stepping away from keyboard for a sec) | 08:26 |
| *** jordanP has joined #openstack-swift | 08:26 | |
| ntt | my new node has 4 disks with a size of 8TB (the current node has 4 disks with a size of 5TB) | 08:26 |
| ntt | my idea is to change the weight of the 4 disks from 5000 to 8000 (even tough they are of 5 TB). At that point, using swift-ring-builder, I want add r1z2 4 times for adding new zone | 08:27 |
| ntt | and before rebalance I want to set_replicas 2 | 08:28 |
| ntt | then I distribute rings across the cluster | 08:28 |
| ntt | is this correct? | 08:28 |
| *** acoles_ is now known as acoles | 08:29 | |
| ahale | yeah i think so | 08:30 |
| ntt | my idea is that I need to change the weight on the first node because the number of partitions are not equal on each replica | 08:31 |
| ahale | yeah, im figuring you want to end up with the two boxes having pretty much the same data | 08:32 |
| ntt | What happens in the cluster when I change the weight of all devices simultaneously? I think no data will be moved | 08:32 |
| ntt | yes | 08:32 |
| ahale | im not sure , that would be the logical thing to happen | 08:33 |
| ahale | you could also just add the 8T disks with weight 5000 | 08:33 |
| *** siva_kri- has left #openstack-swift | 08:33 | |
| ntt | yes, you're right.... but my idea is that I want to remove the node where I have disks with the weight of 5000 | 08:34 |
| ntt | so... if I set the weight of disks of 8TB to 5000 in the future I have to adjust this value (I think) | 08:34 |
| acoles | tdasilva: jrichli i can do any time Weds except 1600-1730 UTC. Would prefer to be before that time but could do later e.g prior to weekly meeting, so 2000UTC. thanks! | 08:36 |
| *** avgoor has joined #openstack-swift | 08:36 | |
| *** siva_kri- has joined #openstack-swift | 08:37 | |
| *** siva_kri- has left #openstack-swift | 08:38 | |
| ntt | ahale: so my question is: what happens in a cluster with 1 region, 1zone, 1replica, 4 devices when I change the weight of all devices from 5000 to 8000? | 08:38 |
| *** siva_kri- has joined #openstack-swift | 08:39 | |
| *** oshritf has quit IRC | 08:39 | |
| *** siva_kri- has left #openstack-swift | 08:40 | |
| acoles | tdasilva: thanks for +A on the merge from master | 08:40 |
| *** siva_kri- has joined #openstack-swift | 08:41 | |
| ahale | yeah just tested that ntt , looks like nothing happens with an equal weight change on a 1 replica 1 box test ring | 08:41 |
| *** siva_kri- has left #openstack-swift | 08:41 | |
| ahale | like https://gist.github.com/racker-andrew-hale/e20ae4a204b8cfde43847a034f927101 | 08:42 |
| ntt | ok ahale.... also I have just tested this morning but when I do the second rebalance I use --force and I rewrite .ring.gz files | 08:43 |
| ntt | is this correct? | 08:43 |
| *** siva_kri- has joined #openstack-swift | 08:43 | |
| *** siva_kri- has left #openstack-swift | 08:43 | |
| ahale | uh maybe.. :) | 08:43 |
| ntt | ok.... I think I will move to production soon... | 08:44 |
| ntt | however... swift is a really flexible software! | 08:44 |
| *** siva_kri- has joined #openstack-swift | 08:44 | |
| *** siva_kri- has left #openstack-swift | 08:45 | |
| ahale | it is- its pretty awesome | 08:45 |
| *** nadeem has quit IRC | 08:49 | |
| acoles | looks like jobs in gate were lost when gerrit went offline yesterday, I resubmitted those I could see pending but you may need to recheck anything you were expecting to merge. | 09:04 |
| *** stantonnet has quit IRC | 09:04 | |
| *** stantonnet has joined #openstack-swift | 09:07 | |
| openstackgerrit | Alistair Coles proposed openstack/swift: POST update to round-trip unit tests https://review.openstack.org/274530 | 09:07 |
| openstackgerrit | Denis V. Meltsaykin proposed openstack/swift: Change pids' type to set https://review.openstack.org/303056 | 09:08 |
| *** SkyRocknRoll has quit IRC | 09:14 | |
| *** openstackgerrit has quit IRC | 09:17 | |
| *** openstackgerrit has joined #openstack-swift | 09:18 | |
| *** asettle has joined #openstack-swift | 09:18 | |
| cschwede | acoles: notmyname: timburke: i updated https://etherpad.openstack.org/p/swift-rolling-upgrade-multinode-testing, linked two patches to devstack and devstack-gate that make it possible to build a consistent ring on all (sub-)nodes | 09:20 |
| cschwede | next patch to add is the one enabling multinode tests with swift, but i want to get some feedback on the latest two patches first | 09:21 |
| *** asettle has quit IRC | 09:23 | |
| *** yatin has joined #openstack-swift | 09:30 | |
| mahatic | tdasilva: acoles the call is about todo's on trello? /me is still catching up since coming back from vacation | 09:37 |
| acoles | cschwede: ack. thanks for all your work on that! | 09:37 |
| acoles | mahatic: welcome back! yes call to discuss status on trello. I guess earlier UTC times are better for you? | 09:38 |
| mahatic | great to see lots of progress on encryption! | 09:38 |
| mahatic | acoles: thanks. yeah, I can stretch until 17:00 UTC, which wil lbe my 10:30 pm | 09:40 |
| *** haomaiwang has joined #openstack-swift | 09:42 | |
| *** haomaiwang has quit IRC | 09:42 | |
| acoles | mahatic: ok we'll keep you posted | 09:43 |
| mahatic | acoles: great thanks! | 09:43 |
| acoles | mahatic: i see you have a talk at summit - that's good! | 09:43 |
| mahatic | acoles: yeah, that's a first for me. and not one but two! both on thursday. Another one is on Outreachy (an internship program that i was a prot of) | 09:44 |
| mahatic | acoles: looking forward to yours and jrichli's crypto talk! | 09:44 |
| mahatic | part* | 09:44 |
| *** yatin has quit IRC | 09:44 | |
| acoles | mahatic: it's going to be all jrichli ! i'm just going to advance the powerpoint. | 09:48 |
| acoles | JK | 09:48 |
| mahatic | lol. good luck with that prep then ;) | 09:49 |
| *** kei_yama has quit IRC | 09:55 | |
| *** openstackstatus has quit IRC | 09:57 | |
| *** openstack has joined #openstack-swift | 10:00 | |
| *** hosanai has quit IRC | 10:14 | |
| *** baojg has quit IRC | 10:16 | |
| *** _JZ_ has joined #openstack-swift | 10:18 | |
| acoles | jrichli: mahatic tdasilva I just managed to repeat jrichli's "all func tests passing with crypto" with EC and replica policies - fairly easy cherry-pick of the copy middleware patches onto patch 257603 :) | 10:18 |
| patchbot | acoles: https://review.openstack.org/#/c/257603/ - swift (feature/crypto) - Conditional GETs Fix | 10:18 |
| mahatic | acoles: that's awesome :) | 10:19 |
| openstackgerrit | Alistair Coles proposed openstack/swift: Conditional GETs Fix https://review.openstack.org/257603 | 10:22 |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto: use cipher required iv length for random iv https://review.openstack.org/304066 | 10:22 |
| *** zul has quit IRC | 10:23 | |
| mahatic | acoles: 156923 and 257603 are pending with copy that need to be cherry-picked right? | 10:23 |
| acoles | mahatic: yes, 257603 is the one significant patch we need to merge on feature/crypto (so if you can review that would be great), then we need copy patches to merge on master - i did the cherry-pick to prove to myself it all worked. | 10:24 |
| mahatic | acoles: sure, will do. Is 257603 ready to be merged? Because I will start to look at it today but don't think will be able to finish it today itself | 10:28 |
| acoles | mahatic: imho it is ready to merge but I made some changes so would value other reviews | 10:29 |
| openstackgerrit | Merged openstack/swift: Modified REPLICATE request to use replication_ip https://review.openstack.org/301963 | 10:29 |
| acoles | mahatic: we could aim to merge it tomorrow | 10:29 |
| mahatic | acoles: sounds good | 10:29 |
| acoles | thanks | 10:30 |
| *** mingdang1 has joined #openstack-swift | 10:36 | |
| openstackgerrit | Merged openstack/swift: Make DirectClientException report correct ip and port https://review.openstack.org/302171 | 10:37 |
| *** zul has joined #openstack-swift | 10:39 | |
| *** daemontool_ has quit IRC | 10:54 | |
| *** daemontool has joined #openstack-swift | 10:54 | |
| *** asettle has joined #openstack-swift | 10:59 | |
| *** asettle has quit IRC | 10:59 | |
| *** takashi has quit IRC | 11:01 | |
| openstackgerrit | Merged openstack/swift: Add unit tests for swift.account.reaper https://review.openstack.org/239105 | 11:04 |
| *** asettle has joined #openstack-swift | 11:53 | |
| *** asettle has quit IRC | 12:04 | |
| *** openstack has quit IRC | 12:04 | |
| *** openstack has joined #openstack-swift | 12:06 | |
| *** dmorita has quit IRC | 12:09 | |
| *** km has quit IRC | 12:13 | |
| notmyname | good morning | 12:16 |
| notmyname | welcome back mahatic! | 12:16 |
| notmyname | cschwede: thanks for the update on rolling upgrades | 12:16 |
| mahatic | notmyname: thanks. good to hear about san antonio visit | 12:18 |
| notmyname | yeah, I'm about to have to get ready and get over there | 12:19 |
| *** klamath has joined #openstack-swift | 12:19 | |
| *** klamath has joined #openstack-swift | 12:20 | |
| tdasilva | good morning | 12:21 |
| pdardeau | good morning | 12:22 |
| notmyname | clayg: FYI I started on some notes on the review guidelines doc yesterday on the plane. my goal is to work on those when I find time this week and I hope we can land something this week | 12:22 |
| notmyname | time to go find some breakfast | 12:23 |
| notmyname | pdardeau: I'll be at the castle at 9ish, assuming traffic isn't too bad | 12:23 |
| pdardeau | notmyname: sounds great | 12:24 |
| *** jamielennox|away is now known as jamielennox | 12:26 | |
| *** jaakkos has quit IRC | 12:26 | |
| *** jaakkos has joined #openstack-swift | 12:28 | |
| *** links has quit IRC | 12:52 | |
| *** mwheckmann has joined #openstack-swift | 12:54 | |
| *** geaaru has quit IRC | 12:58 | |
| *** cdelatte has quit IRC | 13:00 | |
| *** pauloewerton has joined #openstack-swift | 13:05 | |
| tdasilva | acoles, mahatic, jrichli: I'm thinking either 1400 or 1500 UTC time. How does that sound? | 13:10 |
| *** geaaru has joined #openstack-swift | 13:11 | |
| mahatic | tdasilva: sounds good to me | 13:14 |
| *** itlinux has joined #openstack-swift | 13:19 | |
| *** cdelatte has joined #openstack-swift | 13:22 | |
| *** mingdang1 has quit IRC | 13:27 | |
| *** SkyRocknRoll has joined #openstack-swift | 13:28 | |
| *** links has joined #openstack-swift | 13:36 | |
| *** asettle has joined #openstack-swift | 13:42 | |
| *** ametts has joined #openstack-swift | 13:43 | |
| *** asettle has quit IRC | 13:46 | |
| *** jhesketh has left #openstack-swift | 13:56 | |
| *** joeljwright has joined #openstack-swift | 14:00 | |
| *** ChanServ sets mode: +v joeljwright | 14:00 | |
| *** joeljwright has quit IRC | 14:00 | |
| openstackgerrit | Denis V. Meltsaykin proposed openstack/swift: Change pids' type to set https://review.openstack.org/303056 | 14:07 |
| *** links has quit IRC | 14:08 | |
| *** bkeller` has joined #openstack-swift | 14:09 | |
| *** bkeller` has quit IRC | 14:14 | |
| openstackgerrit | Merged openstack/swift: Convert CONTRIBUTING.md to CONTRIBUTING.rst https://review.openstack.org/303660 | 14:15 |
| *** bkeller` has joined #openstack-swift | 14:15 | |
| *** ppai has quit IRC | 14:17 | |
| *** NM has quit IRC | 14:20 | |
| jrichli | tdasilva: 1400 or 1500 UTC tomorrow? That is fine with me as well. | 14:23 |
| jrichli | mahatic: good to have you back! | 14:23 |
| acoles | tdasilva: eithet time works, thanks | 14:24 |
| *** ppai has joined #openstack-swift | 14:28 | |
| tdasilva | acoles, jrichli, mahatic: cool, let's settle for 1400 then. Here's a link: https://bluejeans.com/966565529 | 14:29 |
| *** ppai has quit IRC | 14:30 | |
| mahatic | jrichli: hello, thanks! tdasilva: we don't need to sign up on blue jeans right? | 14:30 |
| *** ppai has joined #openstack-swift | 14:30 | |
| *** avgoor has quit IRC | 14:30 | |
| *** cbartz has joined #openstack-swift | 14:31 | |
| tdasilva | mahatic: nope | 14:31 |
| mahatic | tdasilva: nice, thanks | 14:31 |
| tdasilva | mahatic: but you might need a plugin | 14:32 |
| tdasilva | ;) | 14:32 |
| mahatic | tdasilva: :) shouldn't be a problem | 14:32 |
| *** SkyRocknRoll has quit IRC | 14:32 | |
| *** tongli has joined #openstack-swift | 14:40 | |
| acoles | tdasilva: thanks | 14:40 |
| *** vinsh_ is now known as Vinsh | 14:40 | |
| acoles | tdasilva: does that link have a max capacity or can others join? | 14:41 |
| jrichli | tdasilva: thanks for setting that up | 14:42 |
| *** mvk has quit IRC | 14:43 | |
| tdasilva | acoles: others can join | 14:46 |
| tdasilva | acoles: i think max capacity is pretty high | 14:46 |
| *** cbartz has left #openstack-swift | 14:49 | |
| *** NM has joined #openstack-swift | 14:51 | |
| *** ametts has quit IRC | 14:59 | |
| *** bkeller` has quit IRC | 14:59 | |
| *** mvk has joined #openstack-swift | 15:00 | |
| *** jethrosun has quit IRC | 15:02 | |
| *** bkeller` has joined #openstack-swift | 15:02 | |
| *** geaaru has quit IRC | 15:04 | |
| *** openstackgerrit has quit IRC | 15:18 | |
| *** joeljwright has joined #openstack-swift | 15:18 | |
| *** openstackgerrit has joined #openstack-swift | 15:18 | |
| *** klrmn has joined #openstack-swift | 15:18 | |
| *** ChanServ sets mode: +v joeljwright | 15:18 | |
| *** asettle has joined #openstack-swift | 15:30 | |
| *** SkyRocknRoll has joined #openstack-swift | 15:32 | |
| *** logan- has quit IRC | 15:34 | |
| *** asettle has quit IRC | 15:35 | |
| *** jethrosun has joined #openstack-swift | 15:35 | |
| *** yarkot_ has joined #openstack-swift | 15:38 | |
| *** logan- has joined #openstack-swift | 15:40 | |
| *** zaitcev has joined #openstack-swift | 15:42 | |
| *** ChanServ sets mode: +v zaitcev | 15:42 | |
| *** fesp has joined #openstack-swift | 15:44 | |
| Vinsh | Hi @torgomatic Do you know if "splice" is supported in swift when using EC? I'm getting some logging showing such things as "zero_copy_setnd() blew up... Required key not available" This is on a 4.2 kernel.. so should support splice. | 15:44 |
| Vinsh | I'm wondering if that code gets confused sending ec fragments instead of complete objects. | 15:45 |
| *** jethrosun has quit IRC | 15:45 | |
| *** fesp has quit IRC | 15:45 | |
| Vinsh | A get can fail with "md5sum != etag" in this state | 15:46 |
| *** jethrosun has joined #openstack-swift | 15:50 | |
| openstackgerrit | Gleb Samsonov proposed openstack/swift: go: send to statsd more metrics (timings,pendings etc) https://review.openstack.org/280695 | 15:56 |
| *** gyee has joined #openstack-swift | 16:00 | |
| *** jethrosun has quit IRC | 16:00 | |
| *** esker has joined #openstack-swift | 16:01 | |
| *** jmccarthy1 has joined #openstack-swift | 16:04 | |
| *** bkeller^ has joined #openstack-swift | 16:11 | |
| *** ppai has quit IRC | 16:12 | |
| *** jethrosun has joined #openstack-swift | 16:12 | |
| *** bkeller` has left #openstack-swift | 16:12 | |
| *** bkeller` has quit IRC | 16:12 | |
| *** bkeller^ is now known as bkeller` | 16:12 | |
| *** esker has quit IRC | 16:12 | |
| *** bkeller^ has joined #openstack-swift | 16:13 | |
| *** bkeller^ has quit IRC | 16:13 | |
| *** lyrrad has joined #openstack-swift | 16:19 | |
| *** nadeem has joined #openstack-swift | 16:21 | |
| *** BigWillie has joined #openstack-swift | 16:25 | |
| Vinsh | @torgomatic more data: Apr 12 16:25:20 localhost swift-object-server: zero_copy_send() blew up: #012Traceback (most recent call last):#012 File "/usr/lib/python2.7/dist-packages/swift/obj/server.py", line 1018, in zero_copy_iter#012 app_iter.zero_copy_send(wsockfd)#012 File "/usr/lib/python2.7/dist-packages/swift/obj/diskfile.py", line 1349, in zero_copy_send#012 bin_checksum = os.read(md5_sockfd, 16)#012OSError: [ | 16:26 |
| Vinsh | Errno 126] Required key not available (txn: txe36d6146fcde4e6095934-00570d2170) | 16:26 |
| *** joeljwright has quit IRC | 16:26 | |
| *** mingdang1 has joined #openstack-swift | 16:28 | |
| *** joeljwright has joined #openstack-swift | 16:28 | |
| *** ChanServ sets mode: +v joeljwright | 16:28 | |
| *** joeljwright has quit IRC | 16:31 | |
| *** mingdang1 has quit IRC | 16:32 | |
| *** nadeem has quit IRC | 16:35 | |
| *** nadeem has joined #openstack-swift | 16:36 | |
| *** dmorita has joined #openstack-swift | 16:36 | |
| Vinsh | Change log reads "(AF_ALG sockets must be supported)" Not sure how to even determine that. Would expect the 4.2 HWE Ubuntu kernel to have most useful things. | 16:36 |
| jethrosun | Hello, my patch(change#300710) to fix bug #1558754 is at the status of "Needs workflow", could someone take a look and also give a review? Thx | 16:42 |
| openstack | bug 1558754 in OpenStack Object Storage (swift) "remove devices before min-part-hours requires --force" [Undecided,Fix released] https://launchpad.net/bugs/1558754 - Assigned to Jethro Sun (shwsun) | 16:42 |
| jethrosun | https://review.openstack.org/#/c/300710/ | 16:42 |
| patchbot | jethrosun: patch 300710 - swift - Fix the errors in swift/common/ring/builder.py and... | 16:42 |
| *** jmccarthy1 has quit IRC | 16:51 | |
| *** chsc has joined #openstack-swift | 16:53 | |
| *** rledisez has quit IRC | 16:54 | |
| *** Guest15115 has quit IRC | 16:55 | |
| *** mmcardle has quit IRC | 16:59 | |
| Vinsh | Yes. I have the module "af_alg".. this might be just that swift + ec + splice is not supported. | 17:09 |
| *** SkyRocknRoll has quit IRC | 17:14 | |
| *** jistr has quit IRC | 17:15 | |
| *** jordanP has quit IRC | 17:17 | |
| Vinsh | I find what appears to be support.. maybe not tested though. http://docs.openstack.org/developer/swift/object.html#swift.obj.diskfile.ECDiskFileReader | 17:20 |
| *** jethrosun has quit IRC | 17:22 | |
| *** dencaval has joined #openstack-swift | 17:23 | |
| *** jethrosun has joined #openstack-swift | 17:24 | |
| *** jethrosun has quit IRC | 17:32 | |
| *** jethrosun has joined #openstack-swift | 17:32 | |
| *** jethrosun has quit IRC | 17:37 | |
| *** ametts has joined #openstack-swift | 17:45 | |
| *** jethrosun has joined #openstack-swift | 17:50 | |
| *** rcernin has quit IRC | 17:53 | |
| *** BigWillie has quit IRC | 17:56 | |
| *** BigWillie has joined #openstack-swift | 17:57 | |
| *** yarkot_ has quit IRC | 17:57 | |
| *** dmorita has quit IRC | 17:59 | |
| *** dmorita has joined #openstack-swift | 17:59 | |
| *** ChubYann has joined #openstack-swift | 18:03 | |
| *** bkeller` has quit IRC | 18:05 | |
| *** bkeller` has joined #openstack-swift | 18:08 | |
| Vinsh | https://bugs.launchpad.net/swift/+bug/1569495 | 18:09 |
| openstack | Launchpad bug 1569495 in OpenStack Object Storage (swift) "zero_copy_send() 'blew up' on EC cluster" [Undecided,New] | 18:09 |
| NM | Hello! Does anyone here have any thoughts about multi-region and memcache? I'm thinking about performance if a node consults the memcache from another region. Do you think that is relevant? | 18:10 |
| openstackgerrit | Alistair Coles proposed openstack/swift: crypto - cleanup decrypter exception handling https://review.openstack.org/304806 | 18:11 |
| *** acoles is now known as acoles_ | 18:11 | |
| clayg | ohhh spice | 18:18 |
| clayg | splice | 18:18 |
| clayg | NM: yeah it's very wasted overhead - best to have each region's proxies have a different list of memcache servers | 18:19 |
| Vinsh | spicey splice | 18:19 |
| clayg | NM: this requires using some sort of authoratative auth server (e.g. keystone, or something like LDAP-integration) OR you have to be careful that you use the token on the region that handed it to you. | 18:20 |
| Vinsh | @clayg really? Won't that cause more call's back to keystone or whatever? | 18:20 |
| clayg | Vinsh: i mean each region would have to ask the auth system to validate the token once then it will cache the response in that regions pool | 18:20 |
| Vinsh | I run both regions sharing the same memcahe list, the RTT between them isn't soo bad though. | 18:20 |
| clayg | Vinsh: IME clients tend to stick to a single region - jumping back and forth rapidly isn't as common | 18:21 |
| Vinsh | Interesting thought to explore... splitting that. Thanks for sharing. | 18:21 |
| Vinsh | I agree | 18:21 |
| clayg | Vinsh: we should write it down somewhere | 18:21 |
| Vinsh | Maybe a section in the new-ish "runbook" ? | 18:21 |
| clayg | Vinsh: maybe! I was thinking admin guide for geo-clusters -> https://bugs.launchpad.net/swift/+bug/1569512 | 18:25 |
| openstack | Launchpad bug 1569512 in OpenStack Object Storage (swift) "Geo Cluster docs should discuss memcache" [Wishlist,New] | 18:25 |
| Vinsh | +1 | 18:25 |
| clayg | Vinsh: could you add a comment if you think runbook might also be a good place? | 18:25 |
| NM | clayg: Thanks for the answer. Running keystone here. I was wondering if different pools of memcache (one for each reagion) would be a problem. | 18:26 |
| Vinsh | geo clusters doc is a better fit | 18:26 |
| clayg | words are harder than code right? So lp bug #1569512 is wishlist but not low-hanging-fruit? | 18:26 |
| openstack | Launchpad bug 1569512 in OpenStack Object Storage (swift) "Geo Cluster docs should discuss memcache" [Wishlist,New] https://launchpad.net/bugs/1569512 | 18:26 |
| clayg | NM: no you're good with the keystones | 18:27 |
| clayg | doo eet! | 18:27 |
| NM | clayg: Great! I'll work on that. Tks. | 18:34 |
| *** mvk has quit IRC | 18:37 | |
| *** silor has quit IRC | 18:42 | |
| *** dmorita has quit IRC | 18:43 | |
| *** daemontool has quit IRC | 18:45 | |
| *** dmorita has joined #openstack-swift | 18:45 | |
| *** dmorita has quit IRC | 18:45 | |
| *** yarkot_ has joined #openstack-swift | 18:46 | |
| clayg | so lp bug #1568650 is reproducible for me - but I'm not really digging into it at the moment - i'm sure it's highish priority - but probably not critical since it requires at least a slow client + a getting killed backend storage node. | 18:54 |
| openstack | Launchpad bug 1568650 in OpenStack Object Storage (swift) "Connection between client and proxy service does not closes" [High,Confirmed] https://launchpad.net/bugs/1568650 | 18:54 |
| clayg | but idk, that's just a judgement call - it's confirmed anyway | 18:54 |
| *** ametts has quit IRC | 18:58 | |
| *** dmorita has joined #openstack-swift | 19:03 | |
| *** mmcardle has joined #openstack-swift | 19:08 | |
| openstackgerrit | Michael Barton proposed openstack/swift: go: end-user client https://review.openstack.org/304823 | 19:09 |
| openstackgerrit | David Goetz proposed openstack/swift: go: hummingbird rescueparts https://review.openstack.org/303095 | 19:11 |
| *** gyee has quit IRC | 19:24 | |
| *** gyee has joined #openstack-swift | 19:26 | |
| *** gyee has quit IRC | 19:31 | |
| *** mvk has joined #openstack-swift | 19:36 | |
| *** dmorita has quit IRC | 19:36 | |
| *** dmorita has joined #openstack-swift | 19:37 | |
| *** esker has joined #openstack-swift | 19:38 | |
| *** dmorita has quit IRC | 19:41 | |
| *** sc68cal has joined #openstack-swift | 19:44 | |
| *** sc68cal is now known as Guest48873 | 19:44 | |
| *** daemontool has joined #openstack-swift | 19:50 | |
| *** Guest48873 has quit IRC | 19:54 | |
| *** esker has quit IRC | 19:55 | |
| *** yarkot_ has quit IRC | 19:55 | |
| *** esker has joined #openstack-swift | 19:56 | |
| *** sc68cal_ has joined #openstack-swift | 19:57 | |
| *** cdelatte has quit IRC | 20:00 | |
| *** asettle has joined #openstack-swift | 20:00 | |
| *** tqtran has joined #openstack-swift | 20:00 | |
| *** tongli has quit IRC | 20:01 | |
| *** sc68cal_ has left #openstack-swift | 20:02 | |
| tqtran | hello, quick question. if i wanted to contribute tempest tests to swift, how would i go about doing it? | 20:03 |
| *** asettle has quit IRC | 20:05 | |
| openstackgerrit | Merged openstack/swift: POST update to round-trip unit tests https://review.openstack.org/274530 | 20:08 |
| *** rickyrem has joined #openstack-swift | 20:09 | |
| *** dmorita has joined #openstack-swift | 20:10 | |
| *** esker has quit IRC | 20:33 | |
| *** esker has joined #openstack-swift | 20:34 | |
| *** yarkot_ has joined #openstack-swift | 20:35 | |
| *** mmcardle has quit IRC | 20:38 | |
| openstackgerrit | Merged openstack/swift: go: hummingbird rescueparts https://review.openstack.org/303095 | 20:40 |
| glange | you want to add swift tests to tempest? | 20:40 |
| glange | you want to take existing tempest tests for swift to the swift code base? | 20:41 |
| *** esker has quit IRC | 20:44 | |
| tqtran | glange: i think i found what i was looking for. adding swift tests to tempest was what i wanted. looks like its in the tempest, not swift | 20:45 |
| glange | ok, yeah | 20:45 |
| glange | cool | 20:45 |
| tqtran | https://github.com/openstack/tempest/tree/master/tempest/api/object_storage is the right place yes? | 20:45 |
| glange | I don't know much about tempest, but looks like it | 20:45 |
| glange | maybe there is a tempest irc channel | 20:46 |
| tqtran | :) ok | 20:46 |
| *** jethrosun has quit IRC | 20:46 | |
| *** gyee has joined #openstack-swift | 20:47 | |
| *** BigWillie has quit IRC | 20:48 | |
| *** cdelatte has joined #openstack-swift | 20:48 | |
| *** mvk_ has joined #openstack-swift | 20:49 | |
| *** mvk has quit IRC | 20:53 | |
| *** asettle has joined #openstack-swift | 21:00 | |
| bkeller` | i think #openstack-infra might be the closest | 21:01 |
| *** pauloewerton has quit IRC | 21:01 | |
| *** ametts has joined #openstack-swift | 21:06 | |
| clarkb | #openstack-qa | 21:06 |
| *** trifon has quit IRC | 21:12 | |
| tdasilva | jrichli: nice picture on the User Survey Report | 21:19 |
| *** esker has joined #openstack-swift | 21:20 | |
| *** esker has quit IRC | 21:20 | |
| *** asettle has quit IRC | 21:22 | |
| *** esker has joined #openstack-swift | 21:24 | |
| *** NM has quit IRC | 21:36 | |
| *** sheel has quit IRC | 21:37 | |
| *** asettle has joined #openstack-swift | 21:51 | |
| *** asettle has quit IRC | 21:55 | |
| *** asettle has joined #openstack-swift | 21:55 | |
| *** itlinux has quit IRC | 21:58 | |
| *** trifon has joined #openstack-swift | 22:00 | |
| *** trifon has quit IRC | 22:05 | |
| *** dmorita has quit IRC | 22:20 | |
| *** mingdang1 has joined #openstack-swift | 22:28 | |
| *** jethrosun has joined #openstack-swift | 22:33 | |
| *** ametts has quit IRC | 22:37 | |
| *** lakshmiS_ has joined #openstack-swift | 22:41 | |
| *** itlinux has joined #openstack-swift | 22:43 | |
| *** nadeem has quit IRC | 22:44 | |
| openstackgerrit | Tin Lam proposed openstack/swift: Convert README.md to README.rst https://review.openstack.org/303657 | 22:46 |
| torgomatic | Vinsh: I posted a test script on that bug you filed about zero-copy stuff. | 22:50 |
| *** mingdang1 has quit IRC | 22:50 | |
| *** mwheckmann has quit IRC | 22:52 | |
| *** dmorita has joined #openstack-swift | 22:55 | |
| *** lcurtis has quit IRC | 22:57 | |
| *** km has joined #openstack-swift | 23:00 | |
| *** treaki__ has quit IRC | 23:03 | |
| *** dmorita has quit IRC | 23:05 | |
| *** jamielennox is now known as jamielennox|away | 23:09 | |
| *** jamielennox|away is now known as jamielennox | 23:13 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 23:14 |
| lakshmiS_ | we are trying to integrate swift with searchlight for sending info on account/container/object to be indexed for searching. what's the best place in swift for the searchlight client to hook into? | 23:18 |
| *** klamath has quit IRC | 23:19 | |
| *** itlinux has quit IRC | 23:20 | |
| *** trifon has joined #openstack-swift | 23:22 | |
| *** chsc has quit IRC | 23:25 | |
| *** itlinux has joined #openstack-swift | 23:28 | |
| *** kei_yama has joined #openstack-swift | 23:29 | |
| *** trifon has quit IRC | 23:29 | |
| Vinsh | torgomatic: OK. I will give that a test first thing tomorrow. Thank you for looking! | 23:31 |
| torgomatic | @Vinsh thanks for debugging with me :) | 23:32 |
| *** _JZ_ has quit IRC | 23:36 | |
| *** dmorita has joined #openstack-swift | 23:36 | |
| *** dmorita has quit IRC | 23:37 | |
| *** dmorita has joined #openstack-swift | 23:37 | |
| *** lakshmiS_ has quit IRC | 23:40 | |
| *** lakshmiS has joined #openstack-swift | 23:40 | |
| *** asettle has quit IRC | 23:47 | |
| *** hosanai has joined #openstack-swift | 23:49 | |
| *** ChanServ sets mode: +v hosanai | 23:49 | |
| *** kei_yama has quit IRC | 23:50 | |
| *** kei_yama has joined #openstack-swift | 23:50 | |
| *** tqtran has quit IRC | 23:53 | |
| *** yarkot_ has quit IRC | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!