| *** itlinux has joined #openstack-swift | 00:01 | |
| clayg | I can't rebase on patch 484375 - it's hella old!? | 00:13 |
|---|---|---|
| patchbot | https://review.openstack.org/#/c/484375/ - swift - Make dict deletion idempotent in dump_recon_cache | 00:13 |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Make dict deletion idempotent in dump_recon_cache https://review.openstack.org/484375 | 00:15 |
| *** klamath has quit IRC | 00:19 | |
| *** frank_young has joined #openstack-swift | 00:22 | |
| *** frank_young has quit IRC | 00:26 | |
| openstackgerrit | Clay Gerrard proposed openstack/swift master: Add multiple worker processes strategy to reconstructor https://review.openstack.org/478416 | 00:35 |
| *** lucasxu has joined #openstack-swift | 00:45 | |
| *** lucasxu has quit IRC | 00:45 | |
| *** tovin07_ has joined #openstack-swift | 00:49 | |
| *** klrmn has quit IRC | 01:13 | |
| *** kiennt_ has joined #openstack-swift | 01:19 | |
| *** frank_young has joined #openstack-swift | 01:21 | |
| *** RayLei has joined #openstack-swift | 01:35 | |
| *** vint_bra has joined #openstack-swift | 01:39 | |
| *** bkopilov has quit IRC | 01:52 | |
| *** JimCheung has quit IRC | 01:52 | |
| kota_ | good morning | 01:54 |
| kota_ | acoles: ack, will look at patch 477000 | 01:57 |
| patchbot | https://review.openstack.org/#/c/477000/ - swift - Ring rebalance respects co-builders' last_part_moves | 01:57 |
| kota_ | timburke: it's a perspective that all of ec logic into c (i.e. liberasurecode actually) could make us easy to make a new golang bindings | 01:58 |
| kota_ | on the other hand, in my understanding, the main purpose for StripingDriver is to reduce the python -> c translation overhead as possible. | 01:59 |
| kota_ | hmmm | 02:01 |
| *** clayg_ has joined #openstack-swift | 02:06 | |
| *** ChanServ sets mode: +v clayg_ | 02:06 | |
| *** samueldmq_ has joined #openstack-swift | 02:06 | |
| *** kiennt_ has quit IRC | 02:10 | |
| *** kiennt_ has joined #openstack-swift | 02:10 | |
| *** _d34dh0r53_ has joined #openstack-swift | 02:12 | |
| *** tdasilva- has joined #openstack-swift | 02:13 | |
| *** dhellmann has quit IRC | 02:13 | |
| *** d34dh0r53 has quit IRC | 02:13 | |
| *** tdasilva has quit IRC | 02:13 | |
| *** clayg has quit IRC | 02:13 | |
| *** samueldmq has quit IRC | 02:13 | |
| *** glb has quit IRC | 02:13 | |
| *** clayg_ is now known as clayg | 02:14 | |
| *** samueldmq_ is now known as samueldmq | 02:14 | |
| *** klrmn has joined #openstack-swift | 02:14 | |
| *** dhellmann has joined #openstack-swift | 02:19 | |
| *** vint_bra has quit IRC | 02:26 | |
| openstackgerrit | Tuan Luong-Anh proposed openstack/swift master: Update URL home-page in documents according to document migration https://review.openstack.org/483755 | 02:29 |
| *** winggundamth has joined #openstack-swift | 02:35 | |
| *** winggundamth has quit IRC | 02:37 | |
| *** raginbajin has quit IRC | 02:41 | |
| *** winggundamth has joined #openstack-swift | 02:46 | |
| *** raginbajin has joined #openstack-swift | 03:00 | |
| timburke | kota_: i could see that -- just have one copy from python string to python string instead of three (python string to c string, c string to c string, c string to python string) but something tells me we'd rarely if ever find ourselves bound by the in-memory copy speed | 03:03 |
| *** RayLei has quit IRC | 03:04 | |
| timburke | doesn't really matter much to me, though -- i find the "k of k+m fragments" abstraction much more interesting | 03:04 |
| timburke | (which actually makes the flat_xor_hd backend a little annoying -- you need k+m-2 or k+m-3 [depending on hd] fragments to reconstruct) | 03:06 |
| *** chlong has quit IRC | 03:11 | |
| *** bkopilov has joined #openstack-swift | 03:20 | |
| openstackgerrit | Merged openstack/swift master: Add Timestamp.now() helper https://review.openstack.org/460754 | 03:27 |
| openstackgerrit | Merged openstack/swift master: Add tests for Recon's object replication_time time unit https://review.openstack.org/286032 | 03:28 |
| openstackgerrit | Merged openstack/swift master: Fix StopIteration noise in obj/test_replicator.py https://review.openstack.org/482488 | 03:28 |
| openstackgerrit | Merged openstack/swift master: Make dict deletion idempotent in dump_recon_cache https://review.openstack.org/484375 | 03:28 |
| notmyname | yay merges | 03:29 |
| mattoliverau | Afternoon | 03:30 |
| notmyname | hello mattoliverau | 03:30 |
| mattoliverau | Sorry I'm a bit quiet, I sent the wife away on a break, so am full time Dading with the toddler. She's finally taking a nap :) | 03:31 |
| notmyname | :-) | 03:31 |
| mattoliverau | *this week | 03:31 |
| *** RayLei has joined #openstack-swift | 03:42 | |
| *** OrionX has joined #openstack-swift | 03:51 | |
| *** OrionX_ has joined #openstack-swift | 03:52 | |
| *** links has joined #openstack-swift | 03:55 | |
| *** OrionX has quit IRC | 03:56 | |
| *** frank_young has quit IRC | 04:11 | |
| *** klrmn has quit IRC | 04:13 | |
| *** gkadam has joined #openstack-swift | 04:18 | |
| *** frank_young has joined #openstack-swift | 04:22 | |
| *** RayLei has quit IRC | 04:24 | |
| *** frank_young has quit IRC | 04:26 | |
| *** kiennt_ is now known as kiennt | 04:38 | |
| *** gyee has quit IRC | 04:44 | |
| *** cschwede_ has joined #openstack-swift | 04:44 | |
| *** psachin has joined #openstack-swift | 05:03 | |
| *** frank_young has joined #openstack-swift | 05:06 | |
| *** frank_young has quit IRC | 05:11 | |
| kota_ | hmm... exhausive tests for isa_l_cauchy k+m >= 25 seems too much expensive to run all cases. | 05:23 |
| kota_ | i've been running the tests by k + m < 32 but around 24~25 = k + m patterns looks to be running for a week with aws ec2 m3.medium instance | 05:24 |
| kota_ | so to run fast, i have to run for the bigger parameters in parallel but it seems expensive... :/ | 05:25 |
| *** frank_young has joined #openstack-swift | 05:27 | |
| *** rcernin has joined #openstack-swift | 05:29 | |
| *** cshastri has joined #openstack-swift | 05:40 | |
| *** kiennt has quit IRC | 05:43 | |
| *** kiennt has joined #openstack-swift | 06:00 | |
| *** OrionX_ has quit IRC | 06:15 | |
| *** OrionX has joined #openstack-swift | 06:15 | |
| *** OrionX has quit IRC | 06:20 | |
| *** SkyRocknRoll has joined #openstack-swift | 06:21 | |
| *** xrb has joined #openstack-swift | 06:26 | |
| *** ChubYann has quit IRC | 06:36 | |
| *** itlinux has quit IRC | 06:57 | |
| *** frank_young has quit IRC | 07:00 | |
| *** frank_young has joined #openstack-swift | 07:00 | |
| openstackgerrit | M V P Nitesh proposed openstack/swift master: Fix doc generation for Python3 https://review.openstack.org/457220 | 07:04 |
| *** nathaniel has joined #openstack-swift | 07:15 | |
| *** nathaniel is now known as Guest19161 | 07:16 | |
| *** tesseract has joined #openstack-swift | 07:18 | |
| Guest19161 | Does the object node run a caching process that uses main memory instead of operating system caching? | 07:27 |
| kota_ | Guest19161: AFAIK, Swift object node doesn't cache the object in the main memory, basically. Especially, swift works to run avoid os caching as possible. | 07:44 |
| openstackgerrit | junbo.li proposed openstack/swift master: [Api-ref] fix response status https://review.openstack.org/484254 | 07:44 |
| *** oshritf has joined #openstack-swift | 07:51 | |
| openstackgerrit | Gábor Antal proposed openstack/swift master: Use more specific asserts in test/unit/account tests https://review.openstack.org/342766 | 07:56 |
| *** geaaru has joined #openstack-swift | 08:26 | |
| *** mvk has quit IRC | 08:27 | |
| * kota_ is on patch 477000 | 08:35 | |
| patchbot | https://review.openstack.org/#/c/477000/ - swift - Ring rebalance respects co-builders' last_part_moves | 08:35 |
| *** mvk has joined #openstack-swift | 08:55 | |
| *** cbartz has joined #openstack-swift | 09:00 | |
| *** oshritf has quit IRC | 09:08 | |
| *** oshritf has joined #openstack-swift | 09:09 | |
| *** oshritf has quit IRC | 09:09 | |
| *** ma9_1 has joined #openstack-swift | 09:12 | |
| *** frank_young has quit IRC | 09:16 | |
| *** frank_young has joined #openstack-swift | 09:20 | |
| *** oshritf has joined #openstack-swift | 09:26 | |
| xrb | hi all, | 09:28 |
| xrb | trying to restrict Swift's container access to other users of the same tenant.. | 09:28 |
| xrb | Not sure if it is possible (e.g. using ACLs, I thought ACLs are rather a mean to extend access).. | 09:29 |
| *** mvk has quit IRC | 09:33 | |
| kota_ | xrb: I'm not an expert on ACL but, AFAIK, operators can manage the access via Swift ACL in the tenant | 09:40 |
| kota_ | xrb: the expected access control works as tested in https://github.com/openstack/swift/blob/master/test/functional/test_access_control.py | 09:41 |
| kota_ | so that, you could see the condition of the account management and the expected ACL result there | 09:41 |
| kota_ | the condition for the test is at https://github.com/openstack/swift/blob/master/test/functional/test_access_control.py#L75-L89 which includes "a same project but different role'ed users" | 09:43 |
| *** mvk has joined #openstack-swift | 09:47 | |
| *** Dinesh_Bhor has quit IRC | 09:50 | |
| xrb | kota_: thx! Looking at the link. There even seem to be a role (ResellerAdmin) who gets cross-account access... | 09:53 |
| xrb | kota_: does this mean cross tenant access? | 09:53 |
| kota_ | xrb: true, AFAIK. | 09:56 |
| kota_ | xrb: ResellerAdmin can access whole account/container/objects in the Swift system like as super user. | 09:56 |
| *** tovin07_ has quit IRC | 09:57 | |
| *** kei_yama has quit IRC | 10:16 | |
| *** Dinesh_Bhor has joined #openstack-swift | 10:21 | |
| *** kei_yama has joined #openstack-swift | 10:22 | |
| *** kei_yama has quit IRC | 10:24 | |
| *** frank_young has quit IRC | 10:38 | |
| *** bkopilov has quit IRC | 10:46 | |
| *** frank_young has joined #openstack-swift | 10:48 | |
| *** winggundamth has quit IRC | 10:52 | |
| *** frank_young has quit IRC | 10:52 | |
| *** tonanhngo has joined #openstack-swift | 10:55 | |
| *** tonanhngo has quit IRC | 10:56 | |
| *** tonanhngo has joined #openstack-swift | 10:57 | |
| *** frank_young has joined #openstack-swift | 11:02 | |
| *** jistr is now known as jistr|tpb | 11:05 | |
| *** OrionX has joined #openstack-swift | 11:16 | |
| *** OrionX has quit IRC | 11:20 | |
| *** jistr|tpb is now known as jistr | 11:28 | |
| kota_ | acoles: I added my thought on patch 477000, please see and comments your idea on that. Thanks for working on that. | 11:34 |
| patchbot | https://review.openstack.org/#/c/477000/ - swift - Ring rebalance respects co-builders' last_part_moves | 11:34 |
| *** saint_ has joined #openstack-swift | 11:39 | |
| *** RayLei has joined #openstack-swift | 11:58 | |
| RayLei | If there are two data centers, how is swift accurate and timely aware that one of the data centers has been down and the data is temporarily written to the hand-off node? When the swap out of the data center recovery, swift how accurate and timely know that the data center has been restored? | 11:59 |
| *** chlong has joined #openstack-swift | 11:59 | |
| *** kiennt has quit IRC | 12:14 | |
| *** JimCheung has joined #openstack-swift | 12:20 | |
| *** JimCheung has quit IRC | 12:24 | |
| *** RayLei has quit IRC | 12:24 | |
| *** MVenesio has joined #openstack-swift | 12:29 | |
| *** frank_young has quit IRC | 12:29 | |
| *** caiobrentano has joined #openstack-swift | 12:31 | |
| *** bkopilov has joined #openstack-swift | 12:46 | |
| *** gkadam has quit IRC | 12:47 | |
| *** catintheroof has joined #openstack-swift | 12:51 | |
| *** catintheroof has quit IRC | 12:58 | |
| *** catintheroof has joined #openstack-swift | 13:06 | |
| *** catintheroof has quit IRC | 13:11 | |
| *** frank_young has joined #openstack-swift | 13:13 | |
| *** frank_young has quit IRC | 13:17 | |
| *** frank_young has joined #openstack-swift | 13:19 | |
| *** lucasxu has joined #openstack-swift | 13:31 | |
| *** JimCheung has joined #openstack-swift | 13:32 | |
| *** ma9_ has joined #openstack-swift | 13:35 | |
| *** JimCheung has quit IRC | 13:36 | |
| *** links has quit IRC | 13:38 | |
| *** ma9_1 has quit IRC | 13:38 | |
| *** ma9_ has quit IRC | 13:40 | |
| *** ma9_ has joined #openstack-swift | 13:41 | |
| *** ma9_ has quit IRC | 13:44 | |
| *** ntt has joined #openstack-swift | 13:45 | |
| ntt | Hi, I have a swift cluster with 3 zones and 3 different servers (with replica=3). One of my servers goes down and now I have to restart after 10 days of inactivity. Should I do some particular operation or can I simply restart services on the node and resync starts automatically? | 13:47 |
| *** ma9_ has joined #openstack-swift | 13:48 | |
| *** ma9_1 has joined #openstack-swift | 13:51 | |
| *** catintheroof has joined #openstack-swift | 13:51 | |
| *** ma9_ has quit IRC | 13:55 | |
| openstackgerrit | junbo.li proposed openstack/swift master: [Api-ref] fix response status https://review.openstack.org/484254 | 13:59 |
| *** klamath has joined #openstack-swift | 13:59 | |
| *** klamath has quit IRC | 14:00 | |
| *** klamath has joined #openstack-swift | 14:00 | |
| *** ma9_1 has quit IRC | 14:00 | |
| *** Guest19161 has quit IRC | 14:05 | |
| *** vint_bra has joined #openstack-swift | 14:17 | |
| *** vint_bra has quit IRC | 14:24 | |
| *** vint_bra has joined #openstack-swift | 14:27 | |
| caiobrentano | ntt You can simply restart services. Swift will do all the necessary sync | 14:36 |
| ntt | caiobrentano: thank you | 14:37 |
| openstackgerrit | junbo.li proposed openstack/swift master: [Api-ref] fix response status https://review.openstack.org/484254 | 14:39 |
| *** frank_young has quit IRC | 14:44 | |
| *** frank_young has joined #openstack-swift | 14:54 | |
| xrb | Re cross-tenant access with reseller-admin role: it works (using Swift API). I needed to add a proper user-role 'reseller-admin' specifying the target tenant (not the original tenant), and now I can access the bucket data.. | 14:57 |
| xrb | Only caveat: is works only with the Swift API and not using S3 API... | 14:58 |
| *** frank_young has quit IRC | 14:59 | |
| *** cshastri has quit IRC | 15:01 | |
| *** gkadam has joined #openstack-swift | 15:03 | |
| *** ukaynar has joined #openstack-swift | 15:03 | |
| *** tovin07 has joined #openstack-swift | 15:04 | |
| notmyname | good morning | 15:08 |
| notmyname | kota_: thanks for the cauchy info. let me know if you need anything (another server?) | 15:10 |
| *** vinsh has joined #openstack-swift | 15:11 | |
| xrb | Ok, found a way: working also with S3. I just need to generate a new set of ec2-credentials for the other tenant. | 15:11 |
| xrb | The only drawback: using the reseller-admin role with S3: I cannot access at once cross tenant containers. I need another ec2-credential for each one (hence separate access).. | 15:13 |
| notmyname | ntt: it depends on your replication cycle time and your reclaim age setting. if the server has been off for 10 days, but you're replicating quickly and have a reclaim age of 1 week, then you shouldn't plug in the old data drives. if you do, you have a chance of having previously-deleted data be resurrected (without being referenced in any container listing). | 15:13 |
| notmyname | ntt: instead of plugging in the old data drives, it's better to plug in empty drives and let the replicator make sure the data is durable | 15:14 |
| notmyname | ntt: of course, my first comment depends a little on your use case (how many DELETEs you have, what the settings in the cluster are, etc) | 15:14 |
| *** OrionX has joined #openstack-swift | 15:30 | |
| *** rcernin has quit IRC | 15:32 | |
| *** logan- has joined #openstack-swift | 15:32 | |
| *** mvk has quit IRC | 15:34 | |
| tovin07 | notmyname: hello | 15:35 |
| tovin07 | recently, cschwede_ said that he will review this patch https://review.openstack.org/#/c/468316/ | 15:36 |
| patchbot | patch 468316 - swift - OSprofiler in OpenStack Swift | 15:36 |
| *** OrionX has quit IRC | 15:36 | |
| tovin07 | sadly, there was only one comment from timburke for a long time ago | 15:36 |
| *** itlinux has joined #openstack-swift | 15:36 | |
| notmyname | tovin07: have you talked to cschwede_ in here about it? | 15:38 |
| tovin07 | yes | 15:40 |
| tovin07 | and in recent 0700 meeting | 15:40 |
| cschwede_ | tovin07, i'm still working on it | 15:42 |
| tovin07 | cschwede_: whao, thanks :D | 15:42 |
| *** frank_young has joined #openstack-swift | 15:43 | |
| *** logan- has left #openstack-swift | 15:44 | |
| *** frank_young has quit IRC | 15:48 | |
| notmyname | this is interesting http://www.anandtech.com/show/11639/viking-ships-uhcsilo-ssds-50-tb-capacity-emlc-sas | 15:50 |
| *** frank_young has joined #openstack-swift | 15:52 | |
| *** cbartz has quit IRC | 16:01 | |
| *** rcernin has joined #openstack-swift | 16:02 | |
| timburke | good morning | 16:09 |
| *** tovin07 has left #openstack-swift | 16:20 | |
| *** JimCheung has joined #openstack-swift | 16:22 | |
| *** JimCheung has quit IRC | 16:22 | |
| *** JimCheung has joined #openstack-swift | 16:23 | |
| *** oshritf has quit IRC | 16:42 | |
| *** frank_young has quit IRC | 16:59 | |
| *** psachin has quit IRC | 16:59 | |
| *** jistr is now known as jistr|off | 17:06 | |
| *** ukaynar has quit IRC | 17:10 | |
| *** frank_young has joined #openstack-swift | 17:14 | |
| *** frank_young has quit IRC | 17:18 | |
| *** ChubYann has joined #openstack-swift | 17:20 | |
| *** klrmn has joined #openstack-swift | 17:20 | |
| *** cschwede_ has quit IRC | 17:26 | |
| *** gkadam has quit IRC | 17:42 | |
| *** ukaynar has joined #openstack-swift | 17:52 | |
| *** klamath has quit IRC | 17:55 | |
| *** klamath has joined #openstack-swift | 17:56 | |
| notmyname | I'm working on some mostly internal stuff today, but I'm planning on starting on release stuff for swift and swift client tomorrow. | 17:56 |
| notmyname | please continue to look at the patches on https://wiki.openstack.org/wiki/Swift/PriorityReviews | 17:56 |
| *** klamath has quit IRC | 17:57 | |
| *** tesseract has quit IRC | 17:58 | |
| *** klamath has joined #openstack-swift | 17:58 | |
| *** frank_young has joined #openstack-swift | 18:00 | |
| *** frank_young has quit IRC | 18:04 | |
| *** deep-book-gk_ has joined #openstack-swift | 18:11 | |
| *** deep-book-gk_ has left #openstack-swift | 18:12 | |
| *** klamath has quit IRC | 18:27 | |
| *** geaaru has quit IRC | 18:28 | |
| *** ukaynar has quit IRC | 18:37 | |
| *** frank_young has joined #openstack-swift | 18:40 | |
| *** frank_young has quit IRC | 18:44 | |
| *** ukaynar has joined #openstack-swift | 18:46 | |
| *** chlong_ has joined #openstack-swift | 19:03 | |
| *** chlong has quit IRC | 19:05 | |
| *** Renich has joined #openstack-swift | 19:15 | |
| *** gyee has joined #openstack-swift | 19:17 | |
| *** SkyRocknRoll has quit IRC | 19:25 | |
| *** oshritf has joined #openstack-swift | 19:44 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift master: Updated from global requirements https://review.openstack.org/88736 | 19:46 |
| *** remix_tj has quit IRC | 19:49 | |
| *** remix_tj has joined #openstack-swift | 19:51 | |
| *** oshritf has quit IRC | 19:56 | |
| *** tdasilva- is now known as tdasilva | 20:09 | |
| *** ChanServ sets mode: +v tdasilva | 20:09 | |
| *** Dinesh_Bhor has quit IRC | 20:17 | |
| *** frank_young has joined #openstack-swift | 20:49 | |
| *** frank_young has quit IRC | 20:53 | |
| *** rcernin has quit IRC | 20:57 | |
| *** lucasxu has quit IRC | 21:02 | |
| *** lucasxu has joined #openstack-swift | 21:06 | |
| *** MVenesio has quit IRC | 21:09 | |
| *** lucasxu has quit IRC | 21:17 | |
| *** Renich has quit IRC | 21:18 | |
| *** caiobrentano has quit IRC | 21:30 | |
| *** ChubYann has quit IRC | 21:34 | |
| *** MVenesio has joined #openstack-swift | 21:50 | |
| *** MVenesio has quit IRC | 21:50 | |
| *** frank_young has joined #openstack-swift | 21:58 | |
| *** frank_young has quit IRC | 22:02 | |
| timburke | kota_: notmyname: fwiw, i've checked `k+m <= 28` and `k <= 12, k+m <= 32` for cauchy using something like https://gist.github.com/tipabu/7d7c50dd02f557e30e38c8ebc46bcfca | 22:17 |
| timburke | it's fun pinning 5 or 6 cores for a while! | 22:18 |
| *** gyee has quit IRC | 22:20 | |
| *** vint_bra has quit IRC | 22:26 | |
| *** itlinux has quit IRC | 22:37 | |
| *** gyee has joined #openstack-swift | 22:38 | |
| *** ukaynar_ has joined #openstack-swift | 22:49 | |
| *** ukaynar has quit IRC | 22:51 | |
| *** JimCheung has quit IRC | 23:01 | |
| *** JimCheung has joined #openstack-swift | 23:01 | |
| *** catintheroof has quit IRC | 23:25 | |
| openstackgerrit | Akihiro Motoki proposed openstack/swift master: Turn on warning-is-error in doc build https://review.openstack.org/484981 | 23:33 |
| *** itlinux has joined #openstack-swift | 23:54 | |
| *** lucasxu has joined #openstack-swift | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!