| *** vint_bra has joined #openstack-swift | 00:05 | |
| *** itlinux_ has joined #openstack-swift | 00:26 | |
| *** itlinux_ has quit IRC | 00:32 | |
| *** zhurong has joined #openstack-swift | 00:39 | |
| *** vint_bra has quit IRC | 00:54 | |
| *** tovin07_ has joined #openstack-swift | 00:59 | |
| *** klrmn has quit IRC | 01:05 | |
| *** zhengyin has joined #openstack-swift | 01:10 | |
| *** lucasxu has joined #openstack-swift | 01:19 | |
| *** lucasxu has quit IRC | 01:21 | |
| openstackgerrit | Lingxian Kong proposed openstack/swift master: Write-affinity aware object deletion https://review.openstack.org/470158 | 01:29 |
|---|---|---|
| *** bkopilov has quit IRC | 01:32 | |
| *** vint_bra has joined #openstack-swift | 02:13 | |
| tdasilva | mattoliverau, clayg: changing the IP worked for me | 02:15 |
| clayg | 🎉 | 02:16 |
| *** klrmn has joined #openstack-swift | 02:18 | |
| tdasilva | vagrant halt | 02:29 |
| * tdasilva is falling asleep | 02:30 | |
| *** vint_bra has quit IRC | 02:46 | |
| *** bkopilov has joined #openstack-swift | 02:49 | |
| *** vinsh has quit IRC | 03:05 | |
| *** itlinux_ has joined #openstack-swift | 03:10 | |
| *** catintheroof has joined #openstack-swift | 03:15 | |
| mattoliverau | tdasilva: nice, so that's it then :) maybe the makerings should use 127.0.0.* in the SAIO docs then | 03:16 |
| mattoliverau | Or remakerings | 03:16 |
| *** sanchitmalhotra has quit IRC | 03:19 | |
| *** kei_yama has quit IRC | 03:35 | |
| *** catintheroof has quit IRC | 03:47 | |
| *** kei_yama has joined #openstack-swift | 03:48 | |
| mahatic | good morning | 03:53 |
| mahatic | notmyname: thanks for the info on UpgradeImpact flag (yup acoles said the same :) | 03:57 |
| notmyname | mahatic: I think timburke is the one who had the real help ;-) | 03:58 |
| mahatic | notmyname: hehe, sure | 03:58 |
| *** klrmn has quit IRC | 04:03 | |
| *** tonanhngo has joined #openstack-swift | 04:15 | |
| *** gkadam has joined #openstack-swift | 04:21 | |
| *** psachin has joined #openstack-swift | 04:22 | |
| *** zhurong has quit IRC | 04:28 | |
| *** tonanhngo has quit IRC | 04:37 | |
| *** zhurong has joined #openstack-swift | 04:40 | |
| *** tonanhngo has joined #openstack-swift | 04:41 | |
| *** itlinux_ has quit IRC | 04:44 | |
| *** aselius has quit IRC | 04:45 | |
| *** tonanhngo_ has joined #openstack-swift | 04:56 | |
| *** tonanhngo has quit IRC | 04:59 | |
| *** jaosorior has quit IRC | 05:00 | |
| mattoliverau | mahatic: morning | 05:01 |
| mahatic | mattoliverau: o/ | 05:02 |
| *** tonanhngo_ has quit IRC | 05:17 | |
| openstackgerrit | M V P Nitesh proposed openstack/swift master: Fix doc generation for Python3 https://review.openstack.org/457220 | 05:17 |
| *** tonanhngo has joined #openstack-swift | 05:18 | |
| *** tonanhngo has quit IRC | 05:23 | |
| *** skudlik has joined #openstack-swift | 05:29 | |
| *** gyee has quit IRC | 05:39 | |
| *** abhitechie has joined #openstack-swift | 05:42 | |
| *** cshastri has joined #openstack-swift | 05:43 | |
| *** silor has joined #openstack-swift | 06:02 | |
| *** ChubYann has quit IRC | 06:03 | |
| *** rcernin has quit IRC | 06:11 | |
| *** tonanhngo has joined #openstack-swift | 06:28 | |
| *** rcernin has joined #openstack-swift | 06:37 | |
| *** tesseract has joined #openstack-swift | 06:43 | |
| *** silor has quit IRC | 06:44 | |
| *** silor has joined #openstack-swift | 06:45 | |
| *** pcaruana has joined #openstack-swift | 06:58 | |
| *** aselius has joined #openstack-swift | 07:08 | |
| *** jaosorior has joined #openstack-swift | 07:12 | |
| *** tonanhngo has quit IRC | 07:23 | |
| *** abhitechie has quit IRC | 07:33 | |
| *** hoonetorg has joined #openstack-swift | 07:34 | |
| *** abhitechie has joined #openstack-swift | 07:34 | |
| *** geaaru has joined #openstack-swift | 07:41 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Eliminate node_index in Putter https://review.openstack.org/443072 | 07:43 |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Prepare iter_nodes_local_fist able to take parimary nodes in args https://review.openstack.org/474472 | 07:43 |
| *** gkadam is now known as gkadam-brb | 07:48 | |
| openstackgerrit | Kota Tsuyuzaki proposed openstack/swift master: Eliminate node_index in Putter https://review.openstack.org/443072 | 07:58 |
| *** hseipp has joined #openstack-swift | 08:04 | |
| *** dosaboy has quit IRC | 08:11 | |
| *** dosaboy has joined #openstack-swift | 08:19 | |
| *** oshritf_ has joined #openstack-swift | 08:32 | |
| *** psachin has quit IRC | 08:40 | |
| *** psachin has joined #openstack-swift | 08:41 | |
| *** psachin has quit IRC | 08:43 | |
| *** psachin has joined #openstack-swift | 08:43 | |
| *** zhengyin has quit IRC | 08:48 | |
| *** zhengyin has joined #openstack-swift | 08:48 | |
| *** dosaboy has quit IRC | 08:50 | |
| *** abhitechie has quit IRC | 08:51 | |
| *** dosaboy has joined #openstack-swift | 08:51 | |
| *** dosaboy has quit IRC | 08:53 | |
| openstackgerrit | Van Hung Pham proposed openstack/swift master: Add 'rm -f .testrepository/times.dbm' command in testenv https://review.openstack.org/474501 | 09:03 |
| *** dosaboy has joined #openstack-swift | 09:08 | |
| *** silor has quit IRC | 09:16 | |
| *** aselius has quit IRC | 09:18 | |
| *** hseipp has quit IRC | 09:26 | |
| *** psachin has quit IRC | 09:29 | |
| *** ujjain has joined #openstack-swift | 09:30 | |
| *** ujjain has joined #openstack-swift | 09:30 | |
| *** kei_yama has quit IRC | 09:36 | |
| *** abhitechie has joined #openstack-swift | 09:43 | |
| *** abhinavtechie has joined #openstack-swift | 09:45 | |
| *** abhitechie has quit IRC | 09:47 | |
| *** abhitechie has joined #openstack-swift | 09:47 | |
| *** abhinavtechie has quit IRC | 09:47 | |
| *** d0ugal has quit IRC | 09:48 | |
| *** d0ugal has joined #openstack-swift | 09:49 | |
| *** abhitechie has quit IRC | 09:50 | |
| *** abhitechie has joined #openstack-swift | 09:50 | |
| *** mvk has quit IRC | 09:54 | |
| *** tovin07_ has quit IRC | 09:58 | |
| *** abhitechie has quit IRC | 09:59 | |
| *** abhitechie has joined #openstack-swift | 09:59 | |
| *** mvk has joined #openstack-swift | 10:23 | |
| *** abhitechie has quit IRC | 10:34 | |
| *** abhitechie has joined #openstack-swift | 10:45 | |
| *** gkadam-brb is now known as gkadam | 11:00 | |
| *** torgomatic has quit IRC | 11:01 | |
| *** bkopilov has quit IRC | 11:02 | |
| *** chlong has joined #openstack-swift | 11:31 | |
| *** bkopilov has joined #openstack-swift | 12:24 | |
| *** zhurong has quit IRC | 12:26 | |
| *** zhurong has joined #openstack-swift | 12:28 | |
| openstackgerrit | junbo.li proposed openstack/swift master: Using assertIsNone instead of assertEqual(None) https://review.openstack.org/474562 | 12:32 |
| *** vinsh has joined #openstack-swift | 12:32 | |
| *** zhurong has quit IRC | 12:36 | |
| *** zhengyin has quit IRC | 12:41 | |
| *** oshritf_ has quit IRC | 12:52 | |
| *** oshritf_ has joined #openstack-swift | 12:56 | |
| *** catintheroof has joined #openstack-swift | 13:07 | |
| *** catinthe_ has joined #openstack-swift | 13:21 | |
| *** hseipp has joined #openstack-swift | 13:21 | |
| *** catintheroof has quit IRC | 13:24 | |
| *** catintheroof has joined #openstack-swift | 13:25 | |
| *** catinthe_ has quit IRC | 13:25 | |
| *** catinthe_ has joined #openstack-swift | 13:26 | |
| *** catinth__ has joined #openstack-swift | 13:26 | |
| *** catintheroof has quit IRC | 13:26 | |
| *** lucasxu has joined #openstack-swift | 13:26 | |
| *** klamath has joined #openstack-swift | 13:29 | |
| *** klamath has joined #openstack-swift | 13:29 | |
| *** catintheroof has joined #openstack-swift | 13:29 | |
| *** catinthe_ has quit IRC | 13:30 | |
| *** mariusv has quit IRC | 13:30 | |
| *** catinth__ has quit IRC | 13:33 | |
| *** dancn has quit IRC | 13:33 | |
| *** silor has joined #openstack-swift | 13:49 | |
| *** zhurong has joined #openstack-swift | 13:50 | |
| *** vinsh has quit IRC | 13:50 | |
| *** abhitechie has quit IRC | 13:53 | |
| *** jistr is now known as jistr|mtg | 13:59 | |
| *** tonanhngo has joined #openstack-swift | 14:11 | |
| *** tonanhngo has quit IRC | 14:15 | |
| *** zhurong has quit IRC | 14:22 | |
| *** aselius has joined #openstack-swift | 14:29 | |
| *** silor1 has joined #openstack-swift | 14:41 | |
| *** silor has quit IRC | 14:44 | |
| *** silor1 is now known as silor | 14:44 | |
| *** vinsh has joined #openstack-swift | 14:45 | |
| *** silor1 has joined #openstack-swift | 14:49 | |
| *** cshastri has quit IRC | 14:51 | |
| *** silor has quit IRC | 14:52 | |
| *** silor1 is now known as silor | 14:52 | |
| *** jistr|mtg is now known as jistr | 14:59 | |
| *** MVenesio has joined #openstack-swift | 15:01 | |
| *** skudlik has quit IRC | 15:02 | |
| *** gkadam has quit IRC | 15:04 | |
| *** mvk has quit IRC | 15:05 | |
| *** rcernin has quit IRC | 15:09 | |
| *** gyee has joined #openstack-swift | 15:10 | |
| *** oshritf_ has quit IRC | 15:15 | |
| *** silor1 has joined #openstack-swift | 15:18 | |
| *** silor has quit IRC | 15:22 | |
| *** silor1 is now known as silor | 15:22 | |
| *** silor1 has joined #openstack-swift | 15:27 | |
| *** silor has quit IRC | 15:29 | |
| *** silor1 is now known as silor | 15:29 | |
| *** vint_bra has joined #openstack-swift | 15:31 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-swiftclient master: Updated from global requirements https://review.openstack.org/89250 | 15:34 |
| *** silor1 has joined #openstack-swift | 15:34 | |
| *** silor has quit IRC | 15:35 | |
| *** silor1 is now known as silor | 15:35 | |
| *** rcernin has joined #openstack-swift | 15:44 | |
| *** cshastri has joined #openstack-swift | 15:45 | |
| *** rcernin has quit IRC | 16:12 | |
| *** MVenesio has quit IRC | 16:16 | |
| *** itlinux has joined #openstack-swift | 16:26 | |
| *** MVenesio has joined #openstack-swift | 16:31 | |
| *** silor has quit IRC | 16:34 | |
| *** silor has joined #openstack-swift | 16:41 | |
| timur | timburke: I wrote up a patch to do a streaming upload from stdin using SwiftService and then bumped into a comment in service.py on 1903, where the stream uploads are never uploaded as an SLO. I'm thinking of doing this change in two steps: the patch to do stdin single object uploads and then a follow on change to not use SwiftService for the upload and do the SLO part ourselves | 16:44 |
| *** silor has quit IRC | 16:45 | |
| *** tesseract has quit IRC | 16:46 | |
| *** chsc has joined #openstack-swift | 16:52 | |
| *** chsc has joined #openstack-swift | 16:52 | |
| *** MVenesio has quit IRC | 16:56 | |
| *** klrmn has joined #openstack-swift | 17:05 | |
| timburke | timur: " do the SLO part ourselves" -> start using the client.py API from shell.py? hmm... maybe? | 17:09 |
| timburke | maybe we could plumb in some "no, seriously, i know I have a stream but i *want an SLO*" param? idk, to the extent that i've thought much about it, that was a while ago, so you've probably got more of the details loaded into your head than me | 17:09 |
| timur | timburke: right, using the client.py API in shell.py. I can look into the override of the service.py thing, but I think it'd be a little tricky -- I'll try to see how much of a change it would be, although the fact that it's a filesystem path seems to be a baked in assumption passed that check | 17:13 |
| *** pcaruana has quit IRC | 17:16 | |
| timburke | ah...hmm... yeah, might be more trouble than it's worth then. i mean, the big problem (as i see it) is that service really wants to open multiple readers so it can do parallel uploads, which just isn't going to work for stdin. client from shell may well be easier | 17:18 |
| *** hseipp has quit IRC | 17:25 | |
| *** cshastri has quit IRC | 17:35 | |
| *** geaaru has quit IRC | 17:41 | |
| *** ChubYann has joined #openstack-swift | 17:56 | |
| timur | timburke: I also noticed that swiftclient doesn't remove segments on error during an SLO upload. That seems like a desirable behavior? | 17:56 |
| timur | (cleaning them up, that is) | 17:56 |
| timburke | timur: agreed -- see also, bug 1527017 | 17:57 |
| openstack | bug 1527017 in python-swiftclient "Segment upload failures should abort other segment jobs" [Medium,Confirmed] https://launchpad.net/bugs/1527017 - Assigned to Joel Wright (joel-wright-k) | 17:57 |
| timur | oh, good pointer | 17:57 |
| timburke | (slightly different bad behavior, but certainly related) | 17:58 |
| *** skudlik has joined #openstack-swift | 17:58 | |
| timur | right | 17:58 |
| notmyname | good morning | 17:58 |
| *** cshastri has joined #openstack-swift | 18:02 | |
| *** mvk has joined #openstack-swift | 18:03 | |
| *** MVenesio has joined #openstack-swift | 18:17 | |
| *** mwheckmann has joined #openstack-swift | 18:27 | |
| mwheckmann | hello everyone. Simple questions: Are there any precautions or know caveats when using 512e advanced format drives with Swift? | 18:28 |
| mwheckmann | known caveats I meant | 18:28 |
| notmyname | mwheckmann: emulated 512b sector sizes? | 18:29 |
| notmyname | I'm not sure what "advance format" means in this case | 18:29 |
| notmyname | *advanced | 18:29 |
| mwheckmann | yes. exactly. 4kiB native on platter, but 512 logical | 18:29 |
| notmyname | mwheckmann: I've seen issues where a misalignment of physical and logical sector size (not account for in firmware) has caused issues (ie very slow performance) because most reads/writes werent' aligned properly | 18:30 |
| notmyname | mwheckmann: but if the drive firmware is handling that, it shouldn't be a problem | 18:30 |
| notmyname | mwheckmann: or more generally, there's nothing in swift that cares. it's more of a question of the XFS you put on it | 18:31 |
| notmyname | that is, swift isn't doing direct IO or anything like that which would require certain buffer sizes aligned a certain way | 18:31 |
| mwheckmann | notmyname: so if partition alignment is good, which frankly it should be on any modern distro, we're less likely to encounter issues? | 18:31 |
| notmyname | however, I have also seen performance improvements when the buffer sizes line up all the way from the client to the network to the drives | 18:32 |
| mwheckmann | ok. | 18:32 |
| notmyname | mwheckmann: I wouldn't expect you to see any issues when using emulated sector sizes. please do your due diligence of testing first, of course :-) | 18:32 |
| mwheckmann | I'm asking because I noticed that TripleO have validations tests to ensure that you are *not* using 512e due to problems uploading Glance images. There are no other explanations given however. It's seems really odd. | 18:33 |
| notmyname | I think it would depend on how the emulation is working. is the drive managing some table of virtual sub-sectors? how is it handling writes of less than 4k? | 18:34 |
| notmyname | mwheckmann: that does sound really odd | 18:34 |
| mwheckmann | https://github.com/openstack/tripleo-validations/commit/2e1a198c6b5b39cc8dddcc7097696e9e0b22b00e | 18:34 |
| *** cshastri has quit IRC | 18:34 | |
| mwheckmann | I guess I should take it up with them. We don't use TripleO, but it got me scared for Swift. We do have some 512e drives in prod on Swift and I haven't noticed any particular issues. | 18:35 |
| notmyname | cschwede does some tripleO stuff. I wonder if he knows anything about that | 18:36 |
| clarkb | also they may not use swift to back glance | 18:36 |
| clarkb | so could be related to something else entirely? | 18:36 |
| mwheckmann | agreed, but it still got me scared. | 18:37 |
| tdasilva | clarkb: i believe swift the default backend for glance in tripleo, unless they are deploying ceph | 18:38 |
| notmyname | mwheckmann: for a fuller picture, I'd strongly recommend that swift not be deployed on SMR drives (commonly marketed as "archive" drives). not because things "break" so much as everything just gets really really slow | 18:39 |
| notmyname | so while there aren't any restrictions per-se in swift as to what kind of drives you can use, there are certainly ways that the cluster is impacted based on the drives | 18:40 |
| notmyname | but I don't think that 512/4096/512e matters | 18:40 |
| mwheckmann | notmyname: I'll be watchful for SMR drives. Thanks for the heads up and the tips as usual. | 18:52 |
| mwheckmann | If I get more info from the tripleo team, I'll share it here. | 18:52 |
| *** catinthe_ has joined #openstack-swift | 19:07 | |
| *** jaosorior is now known as jaosorior_away | 19:08 | |
| *** Renich has joined #openstack-swift | 19:10 | |
| *** lucasxu has quit IRC | 19:10 | |
| *** catintheroof has quit IRC | 19:11 | |
| *** lucasxu has joined #openstack-swift | 19:11 | |
| *** catintheroof has joined #openstack-swift | 19:22 | |
| *** catinthe_ has quit IRC | 19:25 | |
| *** Sukhdev has joined #openstack-swift | 19:26 | |
| *** vinsh has quit IRC | 19:36 | |
| *** MVenesio has quit IRC | 19:48 | |
| *** itlinux has quit IRC | 19:59 | |
| *** itlinux has joined #openstack-swift | 20:12 | |
| *** joeljwright has joined #openstack-swift | 20:12 | |
| *** ChanServ sets mode: +v joeljwright | 20:12 | |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift master: Fix a socket leak in copy middleware https://review.openstack.org/474767 | 20:14 |
| *** joeljwright has quit IRC | 20:15 | |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift master: Fix a socket leak in copy middleware https://review.openstack.org/474767 | 20:54 |
| *** lucasxu has quit IRC | 20:55 | |
| *** klamath has quit IRC | 21:01 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Close all versioned_writes subrequests' app_iters https://review.openstack.org/307937 | 21:03 |
| timburke | rledisez: thanks for using self.app.unclosed_requests! i was looking for a way to test ^^ -- part of why i never got around to following up on it | 21:04 |
| timburke | TIL | 21:05 |
| *** catinthe_ has joined #openstack-swift | 21:06 | |
| rledisez | timburke: lots of new close() in versioned_writes now :) | 21:06 |
| rledisez | i didn't check who wrote this helper first, but this is brilliant! | 21:06 |
| timburke | so much closing! it'll be great | 21:06 |
| *** skudlik has quit IRC | 21:07 | |
| *** catintheroof has quit IRC | 21:09 | |
| *** catinthe_ has quit IRC | 21:28 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Close all versioned_writes subrequests' app_iters https://review.openstack.org/307937 | 21:34 |
| *** vinsh has joined #openstack-swift | 21:51 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Add support to increase object ring partition power https://review.openstack.org/337297 | 22:11 |
| *** SkyRocknRoll has joined #openstack-swift | 22:51 | |
| *** itlinux has quit IRC | 23:00 | |
| *** itlinux_ has joined #openstack-swift | 23:09 | |
| *** hoonetorg has quit IRC | 23:10 | |
| openstackgerrit | Tim Burke proposed openstack/swift master: Ignore all auditor_status_*.json files in reconstructor https://review.openstack.org/474801 | 23:10 |
| *** itlinux_ has quit IRC | 23:10 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!