| *** asettle has joined #openstack-swift | 00:22 | |
| *** asettle has quit IRC | 00:27 | |
| *** _JZ_ has quit IRC | 00:46 | |
| *** klrmn has quit IRC | 01:07 | |
| *** chromatin has joined #openstack-swift | 01:19 | |
| chromatin | Has anyone deployed Seagate Kinetic drives in production with Swift ? | 01:22 |
|---|---|---|
| *** asettle has joined #openstack-swift | 01:32 | |
| *** asettle has quit IRC | 01:37 | |
| *** klrmn has joined #openstack-swift | 02:21 | |
| *** asettle has joined #openstack-swift | 03:12 | |
| *** asettle has quit IRC | 03:16 | |
| *** adu has joined #openstack-swift | 03:24 | |
| *** acoles_ has quit IRC | 03:49 | |
| *** asettle has joined #openstack-swift | 04:02 | |
| *** asettle has quit IRC | 04:07 | |
| *** adu has quit IRC | 04:27 | |
| *** vills has joined #openstack-swift | 04:34 | |
| vills | Hi guys! Maybe someone know existed solution or have some thoughts of implementing so. What i want is to move old objects (like older than 7 days) to another container. And i want it work on servers side, not on client. Any ideas? | 04:36 |
| *** Dragon535 has joined #openstack-swift | 04:39 | |
| *** acoles has joined #openstack-swift | 04:49 | |
| *** ChanServ sets mode: +v acoles | 04:49 | |
| *** nadeem has joined #openstack-swift | 04:49 | |
| *** vills has quit IRC | 04:55 | |
| *** nadeem has quit IRC | 05:01 | |
| *** rickflare2 has joined #openstack-swift | 05:02 | |
| *** rickflare has quit IRC | 05:04 | |
| *** rickflare2 is now known as rickflare | 05:04 | |
| *** chromatin has quit IRC | 05:16 | |
| *** Dragon535 has left #openstack-swift | 05:24 | |
| *** nadeem has joined #openstack-swift | 05:25 | |
| *** rcernin has joined #openstack-swift | 05:32 | |
| *** psachin has joined #openstack-swift | 05:33 | |
| *** d0ugal has joined #openstack-swift | 05:50 | |
| *** psachin has quit IRC | 05:55 | |
| *** itlinux has quit IRC | 06:01 | |
| *** garthb has joined #openstack-swift | 06:04 | |
| *** chsc has joined #openstack-swift | 06:08 | |
| *** chsc has joined #openstack-swift | 06:08 | |
| *** zaitcev has quit IRC | 06:09 | |
| *** klrmn has quit IRC | 06:09 | |
| *** d0ugal has quit IRC | 06:09 | |
| *** chsc has quit IRC | 06:14 | |
| *** nadeem has quit IRC | 06:17 | |
| *** psachin has joined #openstack-swift | 06:25 | |
| *** SkyRocknRoll has joined #openstack-swift | 06:32 | |
| *** rcernin has quit IRC | 06:42 | |
| *** rcernin has joined #openstack-swift | 06:42 | |
| *** natarej_ has joined #openstack-swift | 06:51 | |
| *** silor has joined #openstack-swift | 06:51 | |
| *** SkyRocknRoll has quit IRC | 06:54 | |
| *** natarej has quit IRC | 06:54 | |
| *** silor1 has joined #openstack-swift | 06:57 | |
| *** silor has quit IRC | 06:59 | |
| *** silor1 is now known as silor | 06:59 | |
| *** asettle has joined #openstack-swift | 07:35 | |
| *** asettle has quit IRC | 07:40 | |
| *** mmcardle has joined #openstack-swift | 08:00 | |
| *** mmcardle has quit IRC | 08:13 | |
| *** Jeffrey4l has joined #openstack-swift | 08:25 | |
| *** garthb has quit IRC | 08:26 | |
| *** d0ugal has joined #openstack-swift | 08:43 | |
| *** siva_krish has joined #openstack-swift | 08:48 | |
| *** d0ugal has quit IRC | 08:53 | |
| *** ouchkernel has quit IRC | 09:04 | |
| *** ouchkernel has joined #openstack-swift | 09:09 | |
| *** Vr__ has joined #openstack-swift | 09:19 | |
| *** Vr__ has quit IRC | 09:24 | |
| *** siva_krish has quit IRC | 09:34 | |
| *** pcaruana has joined #openstack-swift | 09:43 | |
| *** abhitechie has quit IRC | 10:29 | |
| *** joeljwright has joined #openstack-swift | 10:30 | |
| *** ChanServ sets mode: +v joeljwright | 10:30 | |
| acoles | tdasilva: jrichli timburke I added my notes from symlinks discussion here https://etherpad.openstack.org/p/swift_symlinks | 11:10 |
| *** oshritf has joined #openstack-swift | 11:17 | |
| acoles | timburke: I didn't take any notes about object versioning w/ symlinks so that section is blank :) all I remembered was that symlinks had to be left of encryption :) | 11:17 |
| * acoles afk til weds | 11:18 | |
| *** acoles is now known as acoles_ | 11:19 | |
| *** oshritf has quit IRC | 11:32 | |
| *** asettle has joined #openstack-swift | 11:41 | |
| *** _JZ_ has joined #openstack-swift | 11:45 | |
| *** asettle has quit IRC | 11:45 | |
| kota_ | acoles: nice! thanks! | 12:04 |
| *** joeljwright1 has joined #openstack-swift | 12:15 | |
| *** joeljwright has quit IRC | 12:19 | |
| *** Jeffrey4l has quit IRC | 12:33 | |
| *** Jeffrey4l has joined #openstack-swift | 12:33 | |
| *** mmcardle has joined #openstack-swift | 12:34 | |
| *** vint_bra has joined #openstack-swift | 12:47 | |
| *** vint_bra has quit IRC | 12:47 | |
| *** chromatin has joined #openstack-swift | 13:11 | |
| *** SkyRocknRoll has joined #openstack-swift | 13:30 | |
| *** SkyRocknRoll has quit IRC | 14:11 | |
| *** mmcardle has quit IRC | 14:28 | |
| *** mmcardle has joined #openstack-swift | 14:31 | |
| *** nottrobin has quit IRC | 14:38 | |
| *** nottrobin has joined #openstack-swift | 14:40 | |
| *** adu has joined #openstack-swift | 15:04 | |
| *** adu has quit IRC | 15:28 | |
| *** ozialien10 has quit IRC | 15:53 | |
| *** ozialien10 has joined #openstack-swift | 15:53 | |
| *** chromatin has quit IRC | 15:53 | |
| *** SkyRocknRoll has joined #openstack-swift | 16:07 | |
| *** psachin has quit IRC | 16:38 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 16:53 |
| *** chromatin has joined #openstack-swift | 17:22 | |
| openstackgerrit | Nandini Tata proposed openstack/swift: swift-ring-builder output corrected for ipv6 https://review.openstack.org/319387 | 17:25 |
| *** adu has joined #openstack-swift | 17:28 | |
| *** manous has joined #openstack-swift | 17:30 | |
| *** adu has quit IRC | 17:33 | |
| *** jmccarthy has quit IRC | 17:37 | |
| *** jmccarthy has joined #openstack-swift | 17:38 | |
| chromatin | Does the use of erasure coding obviate the need to split large files , if splitting is done for parallel read reasons? | 17:49 |
| *** klrmn has joined #openstack-swift | 17:52 | |
| *** psachin has joined #openstack-swift | 18:04 | |
| timburke | clayg: so you might be right that a py3 migration might just take a weekend (or week) and a bottle of whiskey. waiting for my plane, i at least got the proxy-server to start and respond to /info -- it wasn't as bad i i feared https://github.com/openstack/swift/compare/master...tipabu:py3-proxy | 18:24 |
| timburke | still can't get it to talk to the backend servers, and at least one header was coming out with b'...' around it, but it's a start | 18:26 |
| notmyname | chromatin: no. erasure coding (EC) is related to how durability is done. large files (ie SLO and DLO) are from the end-user perspective. they're different things | 18:28 |
| notmyname | chromatin: were you asking about using swift for genomics the other day? | 18:28 |
| chromatin | notmyname: Yes, that was me, and I am not sure I agree with your first answer. THat is to say, I understand EC is for durability, and I understand multipart uploads are performed on client end. I am asking about the functional consequences of each achieving a common goal | 18:29 |
| notmyname | right | 18:29 |
| notmyname | chromatin: oh, if you're interested in genomics stuff with swift, you might be interested in https://www.swiftstack.com/resources | 18:30 |
| chromatin | Thanks for the link! | 18:30 |
| notmyname | chromatin: so if an object is stored with EC, that doesn't affect the client-perspective size of the object | 18:30 |
| notmyname | chromatin: if you store a 1GB BAM file, the client will see 1GB, regardless of how is stored--EC or replicas | 18:31 |
| notmyname | just internally, the data will be split up differently with replication and EC | 18:31 |
| notmyname | s/split up/persisted/ | 18:32 |
| chromatin | notmyname: Agree, and perhaps this is covered in the book, but genomics use case is less focused on many-clients parallel reading smaller objects, but few users reading large files. Even 200 MB/sec (single disk throughput) is unacceptable. I see both EC as well as splitting large files on the client end as both achieving the goal of increasing read throughput to a single client for a large object | 18:32 |
| chromatin | (but that is my naive, uninformed view, so welcome criticism) | 18:35 |
| notmyname | yeah, that makes sense. I'm thinking about it | 18:36 |
| notmyname | I think my answer is "maybe" ;-) | 18:37 |
| chromatin | =) I will do some reading on my end | 18:37 |
| notmyname | but you can tell from that answer that the overall throughput for a single stream isn't the primary design goal of EC | 18:37 |
| notmyname | what's your use case? how are you looking to use swift (or using it today)? | 18:39 |
| chromatin | Am not using yet (currently using a couple of NAS without a distributed file system). Recognized that most of our data is rarely accessed / archive and I think object storage makes sense as our growth is still high (maybe not exponential). Right now very small, ~50 TB, but expect to increase by at least that much yearly | 18:40 |
| notmyname | chromatin: very cool :-) | 18:41 |
| notmyname | chromatin: definitely check out that genomics book from the link. it covers a lot of workflows and how swift can fit in | 18:42 |
| chromatin | Will definitely do, thanks again | 18:42 |
| *** hogepodge has quit IRC | 19:04 | |
| *** hogepodge has joined #openstack-swift | 19:05 | |
| *** klrmn1 has joined #openstack-swift | 19:17 | |
| *** klrmn has quit IRC | 19:19 | |
| *** Jeffrey4l_ has joined #openstack-swift | 19:32 | |
| *** Jeffrey4l has quit IRC | 19:34 | |
| jrichli | acoles: excellent symlink write-up. thank you! | 19:38 |
| *** chromatin has quit IRC | 19:42 | |
| *** _JZ_ has quit IRC | 19:56 | |
| *** manous has quit IRC | 20:04 | |
| *** garthb has joined #openstack-swift | 20:07 | |
| openstackgerrit | Christian Schwede proposed openstack/swift: Add support to increase object ring partition power https://review.openstack.org/337297 | 20:12 |
| *** joeljwright1 has quit IRC | 20:19 | |
| *** SkyRocknRoll has quit IRC | 20:20 | |
| *** chromatin has joined #openstack-swift | 20:21 | |
| *** manous has joined #openstack-swift | 20:29 | |
| *** manous has quit IRC | 20:36 | |
| *** manous has joined #openstack-swift | 20:49 | |
| *** garthb has quit IRC | 20:51 | |
| *** silor has quit IRC | 21:06 | |
| manous | hi | 21:42 |
| *** manous has quit IRC | 22:02 | |
| *** _JZ_ has joined #openstack-swift | 22:27 | |
| *** manous has joined #openstack-swift | 22:32 | |
| *** Raymii has quit IRC | 23:01 | |
| *** Raymii has joined #openstack-swift | 23:05 | |
| *** vinsh has quit IRC | 23:06 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!