#ceph IRC Log

Index

IRC Log for 2012-11-29

Timestamps are in GMT/BST.

[0:00] <plut0> how is ceph doing these days?
[0:01] <rweeks> well, next major release is code frozen, should be out soon
[0:01] <plut0> got a changelog yet?
[0:02] <rweeks> that's a developer question… sagewk or elder?
[0:02] <sagewk> git log v0.54..next
[0:02] <joao> lol
[0:06] <plut0> the next release is 0.54?
[0:06] * Leseb (~Leseb@5ED01FAC.cm-7-1a.dynamic.ziggo.nl) Quit (Quit: Leseb)
[0:07] <iggy> 0.54 is already out
[0:07] <plut0> i see that
[0:07] <iggy> 0.55 is supposed to be the next stable iirc
[0:08] <plut0> ok
[0:09] <elder> What sage said means this: "git log" is asking for a log of changes/commits. The "v0.54..next" means between version 0.54 and the current "next" branch, which will end up v0.55.
[0:11] <elder> The result could form the basis for a user-oriented change log, but for now it's something that provides informatino.
[0:11] * gucki (~smuxi@80-218-125-247.dclient.hispeed.ch) Quit (Remote host closed the connection)
[0:12] <elder> sagewk, I'm leaving in about 20 minutes. If you have a few minutes to talk about the "resubmit linger ops" before then let me know.
[0:12] <sagewk> elder: lets do it now
[0:12] <sagewk> hangout?
[0:12] <elder> Sure.
[0:22] * Leseb (~Leseb@5ED01FAC.cm-7-1a.dynamic.ziggo.nl) has joined #ceph
[0:22] * Leseb (~Leseb@5ED01FAC.cm-7-1a.dynamic.ziggo.nl) Quit ()
[0:23] <elder> sagewk, pushed. qa/workunits/rbd/map-unmap.sh
[0:23] <sagewk> elder: sweet
[0:24] * tziOm (~bjornar@ti0099a340-dhcp0628.bb.online.no) Quit (Remote host closed the connection)
[0:25] <sagewk> added to suite
[0:32] <dmick> sweet suite
[0:35] * Steki (~steki@81.18.49.20) Quit (Remote host closed the connection)
[0:35] * Steki (~steki@81.18.49.20) has joined #ceph
[0:39] * Steki (~steki@81.18.49.20) Quit (Read error: Connection reset by peer)
[0:58] * illuminatis (~illuminat@0001adba.user.oftc.net) Quit (Quit: WeeChat 0.3.9.2)
[1:05] * loicd (~loic@magenta.dachary.org) has joined #ceph
[1:15] * ircolle (~ian@c-67-172-132-164.hsd1.co.comcast.net) Quit (Quit: ircolle)
[1:33] * jlogan1 (~Thunderbi@2600:c00:3010:1:742b:7b42:4526:f0f2) Quit (Ping timeout: 480 seconds)
[1:41] * sagelap1 (~sage@169.sub-70-197-130.myvzw.com) has joined #ceph
[1:43] * nwatkins (~Adium@soenat3.cse.ucsc.edu) Quit (Quit: Leaving.)
[1:44] * sagelap (~sage@2607:f298:a:607:f51a:d51:bb88:da91) Quit (Ping timeout: 480 seconds)
[1:46] * maxiz (~pfliu@111.194.202.4) Quit (Quit: Ex-Chat)
[1:49] * sagelap1 (~sage@169.sub-70-197-130.myvzw.com) Quit (Ping timeout: 480 seconds)
[1:56] * rweeks (~rweeks@c-98-234-186-68.hsd1.ca.comcast.net) Quit (Quit: ["Textual IRC Client: www.textualapp.com"])
[1:58] <nhm> wido: I imagine you are sleeping right? ;)
[2:00] * sagelap (~sage@251.sub-70-197-147.myvzw.com) has joined #ceph
[2:01] <joao> nhm, wido got +1 hour on me, and he's by far smarter than I am ;)
[2:02] <nhm> joao: lol
[2:02] <joao> well, looks like I'll have to buy a new fan along with the hdd tomorrow
[2:03] <joao> it just started making that kind of noise that hints something just went wrong
[2:06] <iggy> put it in the freezer
[2:10] * sagelap (~sage@251.sub-70-197-147.myvzw.com) Quit (Quit: Leaving.)
[2:17] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[2:17] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[2:17] <gregaf> sagewk: joao: I just pushed a wip-mon-store-errorcheck branch
[2:18] <joao> looking
[2:18] <gregaf> it passed my primitive "vstart worked, and so did turning it off and back on" test; if one of you can confirm that it looks good (not right away; it's not entirely trivial) then I will figure out how to run a good teuthology monitor test on it tomorrow :)
[2:28] <joao> gregaf, skimmed through the commits and they look fine, but will pay them the proper attention in the morning
[2:30] <joao> got to look better into the ones regarding the put/get functions' return values
[2:32] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[2:32] * loicd (~loic@magenta.dachary.org) has joined #ceph
[3:13] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Ping timeout: 480 seconds)
[3:18] * miroslav (~miroslav@173-228-38-131.dsl.dynamic.sonic.net) Quit (Quit: Leaving.)
[3:29] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[4:26] * deepsa (~deepsa@122.172.220.155) has joined #ceph
[4:40] * plut0 (~cory@pool-96-236-43-69.albyny.fios.verizon.net) has left #ceph
[5:00] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[5:01] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[5:03] * ghbizness (~ghbizness@host-208-68-233-254.biznesshosting.net) Quit (Read error: Connection reset by peer)
[5:03] * ghbizness (~ghbizness@host-208-68-233-254.biznesshosting.net) has joined #ceph
[5:14] <phantomcircuit> i have 4 disks two ssd two conventional
[5:14] <phantomcircuit> i could setup the sdd's through flashcache as writethrough cache
[5:14] <phantomcircuit> but im wondering if that would actually have a significant performance improvement
[6:18] * adjohn (~adjohn@69.170.166.146) Quit (Quit: adjohn)
[6:23] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[6:23] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[6:50] <via> i have an issue with .54 where one of my mds's has crashed and will repeatedly crash upon starting the resolve action: http://pastebin.com/EsVXeWxS
[6:57] <via> aside from the issue, is there a way to have ceph automatically rebalance and use the other mds for ongoing file operation?
[6:59] <via> ...i just set_max_mds down to 1 and it locked up both my client kernel as well as the ceph node carrying the monitor
[7:00] <via> oh, the latter might have been due to it being mounted on that machine incidentally
[7:09] <via> does having multiple mds's mean you can't lose any of them?
[7:17] * chutzpah (~chutz@199.21.234.7) Quit (Quit: Leaving)
[7:22] * dmick (~dmick@2607:f298:a:607:9971:8e07:6a30:2cb8) Quit (Quit: Leaving.)
[7:22] * stass (stas@ssh.deglitch.com) has joined #ceph
[7:22] * ssedov (stas@ssh.deglitch.com) Quit (Read error: Connection reset by peer)
[7:22] <via> i see a code to fix the oops has already been commited, but i still don't know how to even recover from this mds failure
[7:32] <via> all the solutions i can find completely wipe the fs
[7:33] * nwatkins (~Adium@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[7:34] * nwatkins (~Adium@c-50-131-197-174.hsd1.ca.comcast.net) Quit ()
[7:35] * nwatkins (~Adium@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[7:46] * deepsa (~deepsa@122.172.220.155) Quit (Ping timeout: 480 seconds)
[7:46] <iggy> via: the cephfs code itself still isn't production ready, the multi-MDS code even less so I think
[7:46] <iggy> that said, I think it is recoverable, i just don't know the specifics
[7:53] * miroslav (~miroslav@173-228-38-131.dsl.dynamic.sonic.net) has joined #ceph
[7:57] * miroslav (~miroslav@173-228-38-131.dsl.dynamic.sonic.net) Quit ()
[7:59] * nwatkins (~Adium@c-50-131-197-174.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[8:01] * ramsay_za (~ramsay_za@41.223.32.4) has joined #ceph
[8:04] * miroslav (~miroslav@c-98-248-210-170.hsd1.ca.comcast.net) has joined #ceph
[8:05] * Ryan_Lane (~Adium@c-67-160-217-184.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[8:06] * Ryan_Lane (~Adium@c-67-160-217-184.hsd1.ca.comcast.net) has joined #ceph
[8:11] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[8:13] * PerlStalker (~PerlStalk@72.166.192.70) Quit (Ping timeout: 480 seconds)
[8:13] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[8:14] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[8:18] * adjohn (~adjohn@108-225-130-229.lightspeed.sntcca.sbcglobal.net) has joined #ceph
[8:23] * miroslav (~miroslav@c-98-248-210-170.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[8:24] * sjustlaptop (~sam@68-119-138-53.dhcp.ahvl.nc.charter.com) Quit (Ping timeout: 480 seconds)
[8:49] * illuminatis (~illuminat@0001adba.user.oftc.net) has joined #ceph
[8:51] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[8:51] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[8:52] * nosebleedkt (~kostas@kotama.dataways.gr) has joined #ceph
[8:58] * yoshi (~yoshi@p4105-ipngn4301marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[9:03] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Remote host closed the connection)
[9:03] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[9:03] <nosebleedkt> hello everybody
[9:06] * fr0st (~matt@mail.base3.com.au) has joined #ceph
[9:07] <fr0st> Is anyone around that might be able to help with a few performance questions?
[9:19] * verwilst (~verwilst@d5152D6B9.static.telenet.be) has joined #ceph
[9:22] * yoshi (~yoshi@p4105-ipngn4301marunouchi.tokyo.ocn.ne.jp) Quit (Read error: No route to host)
[9:22] * yoshi (~yoshi@p4105-ipngn4301marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[9:23] <todin> morning #ceph
[9:26] <ramsay_za> morning
[9:27] <nosebleedkt> I found a bug. I get a crash while trying to mount ceph from client
[9:27] <nosebleedkt> !
[9:28] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[9:28] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has left #ceph
[9:42] * tnt (~tnt@55.188-67-87.adsl-dyn.isp.belgacom.be) Quit (Ping timeout: 480 seconds)
[9:53] * yoshi (~yoshi@p4105-ipngn4301marunouchi.tokyo.ocn.ne.jp) Quit (Read error: Connection reset by peer)
[9:53] * tnt (~tnt@212-166-48-236.win.be) has joined #ceph
[9:54] * ScOut3R (~ScOut3R@212.96.47.215) has joined #ceph
[9:54] * yoshi (~yoshi@p4105-ipngn4301marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[10:00] * nosebleedkt_ (~kostas@kotama.dataways.gr) has joined #ceph
[10:04] * nosebleedkt (~kostas@kotama.dataways.gr) Quit (Ping timeout: 480 seconds)
[10:07] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) has joined #ceph
[10:13] * jbd_ (~jbd_@34322hpv162162.ikoula.com) has joined #ceph
[10:21] * yoshi (~yoshi@p4105-ipngn4301marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[10:27] <fr0st> Is there a way to disable the journal completely?
[10:27] <fr0st> I think it may help with my SSD performance problems..
[10:32] * adjohn (~adjohn@108-225-130-229.lightspeed.sntcca.sbcglobal.net) Quit (Quit: adjohn)
[10:33] * verwilst (~verwilst@d5152D6B9.static.telenet.be) Quit (Ping timeout: 480 seconds)
[10:35] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[10:42] * Ryan_Lane (~Adium@c-67-160-217-184.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[10:45] * verwilst (~verwilst@d5152D6B9.static.telenet.be) has joined #ceph
[10:49] * loicd (~loic@LAubervilliers-153-52-6-55.w217-128.abo.wanadoo.fr) has joined #ceph
[10:54] * loicd (~loic@LAubervilliers-153-52-6-55.w217-128.abo.wanadoo.fr) Quit ()
[11:11] * kermin (ca037809@ircip3.mibbit.com) has joined #ceph
[11:17] * calebamiles (~caleb@c-24-128-194-192.hsd1.vt.comcast.net) Quit (Quit: Leaving.)
[11:24] * MikeMcClurg1 (~mike@firewall.ctxuk.citrix.com) has joined #ceph
[11:24] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) Quit (Read error: Connection reset by peer)
[11:26] * loicd (~loic@LAubervilliers-153-52-6-55.w217-128.abo.wanadoo.fr) has joined #ceph
[11:30] <kermin> how to use cephfs to mount specific pool to dir
[11:39] <tnt> you can't really do that.
[11:40] <tnt> what you need to do is create a subdir in the cephfs filesystem, then assign that subdir to another pool using cephfs set_layout and then you can mount only that subdir directly.
[11:42] * tziOm (~bjornar@ti0099a340-dhcp0628.bb.online.no) has joined #ceph
[11:46] * loicd (~loic@LAubervilliers-153-52-6-55.w217-128.abo.wanadoo.fr) Quit (Read error: Connection timed out)
[11:51] <kermin> how to do that Sir?
[12:00] * jlogan1 (~Thunderbi@72.5.59.176) has joined #ceph
[12:06] * mengesb (~bmenges@servepath-gw3.servepath.com) has joined #ceph
[12:07] * scalability-junk (~stp@188-193-211-236-dynip.superkabel.de) Quit (Ping timeout: 480 seconds)
[12:08] * jlogan1 (~Thunderbi@72.5.59.176) Quit (Ping timeout: 480 seconds)
[12:09] * andreask1 (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[12:12] <kermin> @tnt : how can I do that ?
[12:12] <cephalobot> kermin: Error: "tnt" is not a valid command.
[12:12] <kermin> tnt : how can I do that ?
[12:14] <tnt> read the doc. I've never done it myself, I just know it's possible.
[12:14] <kermin> actually i talked to sage sir.
[12:14] <kermin> bt I have problem to do it somhow
[12:14] <kermin> :(
[12:27] * nosebleedkt_ (~kostas@kotama.dataways.gr) Quit (Ping timeout: 480 seconds)
[12:28] <kermin> tnt: i am able to ingest data to specific pool
[12:28] <kermin> but problem is I can use only one pool at a time
[12:29] <kermin> I cant use two pools simultaneously
[12:29] * MikeMcClurg1 (~mike@firewall.ctxuk.citrix.com) Quit (Quit: Leaving.)
[12:30] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) has joined #ceph
[12:32] * kermin (ca037809@ircip3.mibbit.com) Quit (Quit: http://www.mibbit.com ajax IRC Client)
[12:33] * nosebleedkt_ (~kostas@kotama.dataways.gr) has joined #ceph
[12:34] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) Quit ()
[12:34] <nosebleedkt_> This means the image is still open or the client using it crashed. Try again after closing/unmapping it or waiting 30s for the crashed client to timeout.
[12:34] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) has joined #ceph
[12:34] <nosebleedkt_> tnt, joao
[12:35] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) Quit ()
[12:36] <nosebleedkt_> root@cephfs:/mnt# rbd rm foo
[12:36] <nosebleedkt_> Removing image: 99% complete...failed.
[12:36] <nosebleedkt_> delete error: image still has watchers
[12:36] <nosebleedkt_> This means the image is still open or the client using it crashed. Try again after closing/unmapping it or waiting 30s for the crashed client to timeout.
[12:36] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) has joined #ceph
[12:36] <nosebleedkt_> what's the problem here ?
[12:38] * andreask1 (~andreas@chello062178013131.5.11.vie.surfer.at) has left #ceph
[12:42] * firaxis (~vortex@unnum-91-196-193-107.domashka.kiev.ua) Quit (Ping timeout: 480 seconds)
[12:45] * firaxis (~vortex@unnum-91-196-193-107.domashka.kiev.ua) has joined #ceph
[12:45] * nosebleedkt__ (~kostas@kotama.dataways.gr) has joined #ceph
[12:47] * mengesb (~bmenges@servepath-gw3.servepath.com) Quit (Quit: Leaving.)
[12:48] * nosebleedkt_ (~kostas@kotama.dataways.gr) Quit (Ping timeout: 480 seconds)
[12:49] * mengesb (~bmenges@servepath-gw3.servepath.com) has joined #ceph
[12:49] * loicd (~loic@LAubervilliers-153-52-6-55.w217-128.abo.wanadoo.fr) has joined #ceph
[12:55] * loicd (~loic@LAubervilliers-153-52-6-55.w217-128.abo.wanadoo.fr) Quit (Quit: Leaving.)
[13:05] * maxiz (~pfliu@222.128.131.156) has joined #ceph
[13:21] <tnt> nosebleedkt__: ... it means "the image is still open or the client using it crashed"
[13:21] <nosebleedkt__> tnt, yes how its open ?
[13:21] <nosebleedkt__> i have umounted it
[13:21] <tnt> you need to unmap it
[13:32] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[13:48] * loicd (~loic@90.84.144.227) has joined #ceph
[13:52] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[13:54] * loicd (~loic@90.84.144.227) Quit (Quit: Leaving.)
[13:56] <nosebleedkt__> tnt, i did it :(
[14:12] * scalability-junk (~stp@188-193-211-236-dynip.superkabel.de) has joined #ceph
[14:30] * iltisanni (d4d3c928@ircip3.mibbit.com) Quit (Quit: http://www.mibbit.com ajax IRC Client)
[14:31] * aliguori (~anthony@cpe-70-123-145-75.austin.res.rr.com) has joined #ceph
[14:43] * silversu_ (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[14:45] * epitron (~epitron@bito.ponzo.net) has joined #ceph
[14:47] * kbad_ (~kbad@malicious.dreamhost.com) has joined #ceph
[14:47] * fc_ (~fc@home.ploup.net) has joined #ceph
[14:48] * __jt___ (~james@jamestaylor.org) has joined #ceph
[14:48] * verwilst (~verwilst@d5152D6B9.static.telenet.be) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * fr0st (~matt@mail.base3.com.au) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * dok (~dok@static-50-53-68-158.bvtn.or.frontiernet.net) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * epitron_ (~epitron@bito.ponzo.net) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * match (~mrichar1@pcw3047.see.ed.ac.uk) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * joshd (~joshd@2607:f298:a:607:221:70ff:fe33:3fe3) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * kbad (~kbad@malicious.dreamhost.com) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * madphoenix (~btimm@128.104.79.82) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * mdxi (~mdxi@74-95-29-182-Atlanta.hfc.comcastbusiness.net) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * fc (~fc@home.ploup.net) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * raso (~raso@deb-multimedia.org) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * jeffhung (~jeffhung@60-250-103-120.HINET-IP.hinet.net) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * exec (~defiler@109.232.144.194) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * __jt__ (~james@jamestaylor.org) Quit (synthon.oftc.net charm.oftc.net)
[14:48] * iggy_ (~iggy@theiggy.com) Quit (Remote host closed the connection)
[14:49] * verwilst (~verwilst@d5152D6B9.static.telenet.be) has joined #ceph
[14:49] * fr0st (~matt@mail.base3.com.au) has joined #ceph
[14:49] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) has joined #ceph
[14:49] * dok (~dok@static-50-53-68-158.bvtn.or.frontiernet.net) has joined #ceph
[14:49] * epitron_ (~epitron@bito.ponzo.net) has joined #ceph
[14:49] * match (~mrichar1@pcw3047.see.ed.ac.uk) has joined #ceph
[14:49] * joshd (~joshd@2607:f298:a:607:221:70ff:fe33:3fe3) has joined #ceph
[14:49] * madphoenix (~btimm@128.104.79.82) has joined #ceph
[14:49] * mdxi (~mdxi@74-95-29-182-Atlanta.hfc.comcastbusiness.net) has joined #ceph
[14:49] * raso (~raso@deb-multimedia.org) has joined #ceph
[14:49] * exec (~defiler@109.232.144.194) has joined #ceph
[14:49] * epitron_ (~epitron@bito.ponzo.net) Quit (Read error: Connection reset by peer)
[14:49] * silversurfer (~silversur@124x35x68x250.ap124.ftth.ucom.ne.jp) Quit (Ping timeout: 480 seconds)
[14:49] * raso (~raso@deb-multimedia.org) Quit (Ping timeout: 480 seconds)
[14:49] * madphoenix (~btimm@128.104.79.82) Quit (Ping timeout: 480 seconds)
[14:50] * fr0st (~matt@mail.base3.com.au) Quit (Ping timeout: 480 seconds)
[14:50] * exec (~defiler@109.232.144.194) Quit (Ping timeout: 480 seconds)
[14:50] * fr0st (~matt@mail.base3.com.au) has joined #ceph
[14:50] * raso (~raso@deb-multimedia.org) has joined #ceph
[14:50] * exec (~defiler@109.232.144.194) has joined #ceph
[14:50] * iggy (~iggy@theiggy.com) Quit (Ping timeout: 483 seconds)
[14:50] * jeffhung (~jeffhung@60-250-103-120.HINET-IP.hinet.net) has joined #ceph
[14:50] * madphoenix (~btimm@128.104.79.82) has joined #ceph
[14:51] <nhm> good morning #ceph
[14:54] * iggy (~iggy@theiggy.com) has joined #ceph
[15:08] <todin> good afternoon nhm
[15:16] * nhorman (~nhorman@nat-pool-rdu.redhat.com) has joined #ceph
[15:26] * The_Bishop_ (~bishop@f052099189.adsl.alicedsl.de) has joined #ceph
[15:27] * mdawson (~chatzilla@host-69-95-126-226.cwon.choiceone.net) has joined #ceph
[15:33] * The_Bishop (~bishop@e179015093.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[15:35] * mdawson (~chatzilla@host-69-95-126-226.cwon.choiceone.net) Quit (Ping timeout: 480 seconds)
[15:38] <jtang> good morning
[15:38] <jtang> its afternoon!
[15:40] <via> iggy: okay, thanks
[15:43] * loicd (~loic@magenta.dachary.org) has joined #ceph
[15:53] * mdawson (~chatzilla@host-69-95-126-226.cwon.choiceone.net) has joined #ceph
[16:04] * calebmiles (~caleb@65-183-137-95-dhcp.burlingtontelecom.net) has joined #ceph
[16:05] * nosebleedkt__ (~kostas@kotama.dataways.gr) Quit (Quit: Leaving)
[16:08] * calebmiles (~caleb@65-183-137-95-dhcp.burlingtontelecom.net) has left #ceph
[16:09] * calebmiles (~caleb@65-183-137-95-dhcp.burlingtontelecom.net) has joined #ceph
[16:31] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[16:31] * loicd (~loic@magenta.dachary.org) has joined #ceph
[16:31] * illuminatis (~illuminat@0001adba.user.oftc.net) Quit (Quit: WeeChat 0.3.9.2)
[16:35] * scuttlemonkey (~scuttlemo@c-69-244-181-5.hsd1.mi.comcast.net) Quit (Ping timeout: 480 seconds)
[16:43] * scuttlemonkey (~scuttlemo@c-69-244-181-5.hsd1.mi.comcast.net) has joined #ceph
[16:43] * ChanServ sets mode +o scuttlemonkey
[17:03] * aliguori (~anthony@cpe-70-123-145-75.austin.res.rr.com) Quit (Remote host closed the connection)
[17:20] * deepsa (~deepsa@101.62.45.61) has joined #ceph
[17:23] * ircolle (~ian@c-67-172-132-164.hsd1.co.comcast.net) has joined #ceph
[17:27] * ScOut3R (~ScOut3R@212.96.47.215) Quit (Remote host closed the connection)
[17:28] * miroslav (~miroslav@173-228-38-131.dsl.dynamic.sonic.net) has joined #ceph
[17:33] * jlogan1 (~Thunderbi@2600:c00:3010:1:e12a:776f:2a6d:8a8) has joined #ceph
[17:34] * verwilst (~verwilst@d5152D6B9.static.telenet.be) Quit (Quit: Ex-Chat)
[17:36] * sagewk (~sage@2607:f298:a:607:e116:e786:b94f:5586) Quit (Ping timeout: 480 seconds)
[17:36] * aliguori (~anthony@32.97.110.59) has joined #ceph
[17:37] * deepsa (~deepsa@101.62.45.61) Quit (Ping timeout: 480 seconds)
[17:38] * mdawson (~chatzilla@host-69-95-126-226.cwon.choiceone.net) Quit (Quit: ChatZilla 0.9.89 [Firefox 17.0/20121119183901])
[17:47] * sagewk (~sage@2607:f298:a:607:611c:760a:a2d9:a4b1) has joined #ceph
[17:54] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[17:54] * loicd (~loic@magenta.dachary.org) has joined #ceph
[17:55] * scuttlemonkey (~scuttlemo@c-69-244-181-5.hsd1.mi.comcast.net) Quit (Quit: Leaving)
[17:55] * scuttlemonkey (~scuttlemo@c-69-244-181-5.hsd1.mi.comcast.net) has joined #ceph
[17:55] * ChanServ sets mode +o scuttlemonkey
[18:08] * gucki (~smuxi@80-218-125-247.dclient.hispeed.ch) has joined #ceph
[18:14] * calebmiles (~caleb@65-183-137-95-dhcp.burlingtontelecom.net) Quit (Ping timeout: 480 seconds)
[18:25] * tnt (~tnt@212-166-48-236.win.be) Quit (Ping timeout: 480 seconds)
[18:25] <elder> sagewk, question about lists used for osd requests, whenever you're around.
[18:33] <sagewk> elder: yeah
[18:33] <sagewk> hangout?
[18:33] <elder> Sure
[18:34] * calebmiles (~caleb@65-183-137-95-dhcp.burlingtontelecom.net) has joined #ceph
[18:44] * tnt (~tnt@83.164-67-87.adsl-dyn.isp.belgacom.be) has joined #ceph
[18:47] * nwatkins (~Adium@soenat3.cse.ucsc.edu) has joined #ceph
[18:56] * jbd_ (~jbd_@34322hpv162162.ikoula.com) has left #ceph
[19:03] * nwatkins (~Adium@soenat3.cse.ucsc.edu) has left #ceph
[19:03] * fc_ (~fc@home.ploup.net) Quit (Remote host closed the connection)
[19:03] * fc (~fc@home.ploup.net) has joined #ceph
[19:06] * chutzpah (~chutz@199.21.234.7) has joined #ceph
[19:14] * fc (~fc@home.ploup.net) Quit (Quit: leaving)
[19:15] * brambles (lechuck@s0.barwen.ch) has joined #ceph
[19:17] * sjustlaptop (~sam@68-119-138-53.dhcp.ahvl.nc.charter.com) has joined #ceph
[19:20] * rweeks (~rweeks@c-24-4-66-108.hsd1.ca.comcast.net) has joined #ceph
[19:26] * BManojlovic (~steki@81.18.49.20) has joined #ceph
[19:32] * jjgalvez (~jjgalvez@cpe-76-175-17-226.socal.res.rr.com) has joined #ceph
[19:32] * Ryan_Lane (~Adium@216.38.130.164) has joined #ceph
[19:49] * brambles (lechuck@s0.barwen.ch) Quit (Quit: leaving)
[19:50] * BManojlovic (~steki@81.18.49.20) Quit (Ping timeout: 480 seconds)
[19:52] * brambles (lechuck@s0.barwen.ch) has joined #ceph
[20:01] * adjohn (~adjohn@69.170.166.146) has joined #ceph
[20:02] * The_Bishop_ (~bishop@f052099189.adsl.alicedsl.de) Quit (Ping timeout: 480 seconds)
[20:03] * The_Bishop_ (~bishop@f052099189.adsl.alicedsl.de) has joined #ceph
[20:03] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) Quit (Quit: Leaving.)
[20:13] * ssedov (stas@ssh.deglitch.com) has joined #ceph
[20:15] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) has joined #ceph
[20:15] * stass (stas@ssh.deglitch.com) Quit (Ping timeout: 480 seconds)
[20:17] * jjgalvez (~jjgalvez@cpe-76-175-17-226.socal.res.rr.com) Quit (Quit: Leaving.)
[20:19] * MikeMcClurg (~mike@firewall.ctxuk.citrix.com) Quit ()
[20:25] * jpieper (~josh@209-6-86-62.c3-0.smr-ubr2.sbo-smr.ma.cable.rcn.com) has joined #ceph
[20:29] <jefferai> I'm wondering -- if I have rbd-backed kvm virtual machines, will I see performance increases by doing things like raid-0 the virtual disks inside the VM?
[20:30] <jefferai> I could see the answer going either way
[20:30] <jefferai> either, no, because it will use as much bandwidth as possible already
[20:30] <jefferai> or, yes, because it will be able to multiplex requests to the RBD devices (and thus the underlying rados pool)
[20:31] <joshd> it can potentially help, since each new device instantiates independent communications with the cluster
[20:32] <joshd> it depends where the bottleneck is, of course. if you're client-nic limited, it won't matter
[20:32] <jefferai> sure
[20:32] <jefferai> I have active-active 10gbe, so I'm guessing not that :-)
[20:34] * jjgalvez (~jjgalvez@12.248.40.138) has joined #ceph
[20:35] * BManojlovic (~steki@81.18.49.20) has joined #ceph
[20:38] <sjustlaptop> jefferai: we have seen cases where two clients on the same machines can outperform a single client due to inefficiencies in the messenger layer, it might be worth trying a raid0
[20:40] * jjgalvez (~jjgalvez@12.248.40.138) Quit (Read error: Connection reset by peer)
[20:41] * BManojlovic (~steki@81.18.49.20) Quit (Read error: Connection reset by peer)
[20:42] * BManojlovic (~steki@81.18.49.20) has joined #ceph
[20:51] * Steki (~steki@gprswap.mts.telekom.rs) has joined #ceph
[20:54] * BManojlovic (~steki@81.18.49.20) Quit (Ping timeout: 480 seconds)
[20:59] <jefferai> sjustlaptop: cool, thanks for the info
[20:59] * nhorman (~nhorman@nat-pool-rdu.redhat.com) Quit (Quit: Leaving)
[21:09] <elder> sagewk, that kernel+patches from Nate is really old stuff. Each time I try to add something to get the latest two patches to build, something new is missing.
[21:11] * MikeMcClurg (~mike@cpc10-cmbg15-2-0-cust205.5-4.cable.virginmedia.com) has joined #ceph
[21:14] * aliguori (~anthony@32.97.110.59) Quit (Ping timeout: 480 seconds)
[21:15] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[21:15] * loicd (~loic@magenta.dachary.org) has joined #ceph
[21:21] * calebmiles (~caleb@65-183-137-95-dhcp.burlingtontelecom.net) Quit (Quit: Leaving.)
[21:25] <sagewk> elder: nate?
[21:26] <jefferai> what is the "sector size" of an RBD disk? Would it be the same 4MB as the chunk size for the underlying rados pool?
[21:26] <elder> I mean Nick.
[21:26] <jefferai> I figure it would be good to align operating system filesystem operations to that sector size
[21:26] <elder> Sector size of kernel rbd is 512 bytes.
[21:26] <elder> It reports back that 4 MB is the optimal alignment though.
[21:26] <elder> (Or whatever your object order is)
[21:27] <jefferai> ah, ok
[21:27] <jefferai> so setting 4MB as my filesystem pool sector size wouoldn't be a bad idea
[21:27] <jefferai> it can always break that into 512 byte chunks
[21:27] <jefferai> unless that would cause a lot more traffic than necessary for many operations
[21:29] <jefferai> (right? :-) )
[21:29] <elder> I'm sorry I don't know the answer to that. (Is this for btrfs or something?)
[21:32] <sagewk> elder: ah. we can just provide the 2 patches for him to add to his tree..
[21:33] * cephalobot (~ceph@ds2390.dreamservers.com) Quit (Remote host closed the connection)
[21:33] * cephalobot (~ceph@ds2390.dreamservers.com) has joined #ceph
[21:34] * nhorman (~nhorman@hmsreliant.think-freely.org) has joined #ceph
[21:34] <elder> No the problem is they don't apply cleanly. But in any case I have it now.
[21:38] * morse (~morse@supercomputing.univpm.it) Quit (Ping timeout: 480 seconds)
[21:39] * glowell (~glowell@c-98-210-224-250.hsd1.ca.comcast.net) Quit (Ping timeout: 480 seconds)
[21:40] <jefferai> elder: it's zfs, but it's not the only thing that lets you pick sizes
[21:40] <jefferai> for instance, straight up LVM lets you choose how big to make its extents
[21:40] <sagewk> elder: oh i see
[21:41] <elder> Well jefferai I have no experience with ZFS so I don't know how to answer your question.
[21:41] <jefferai> it wasn't meant as a ZFS specific question, which is why I didn't mention ZFS
[21:41] <jefferai> since it applies to e.g. mdraid and lvm
[21:41] <jefferai> and various other filesystems
[21:41] <jefferai> all of those let you specify chunk/extent sizes
[21:42] <elder> I think that 4MB is a rather large chunk.
[21:42] <elder> But if that won't cause you problems, such as internal fragmentation, then maybe that's the right thing to use.
[21:42] <joshd> jefferai: it depends on whether that chunk size is treated as the minimum atomic size, or just a guideline for the largest size to try to combine requests to
[21:43] <jefferai> that's a really good point
[21:43] <elder> In the end, 512 bytes is the smallest I/O size supported by the Linux block I/O system. Any other block size should be a power of two multiple of that.
[21:43] <joshd> the latter is what you'd want to set based on the object size
[21:43] <jefferai> right
[21:43] <gregaf> sagewk: comment on wip-mds-ls; unfortunately the request_kill() change is going to need to be more complicated
[21:43] <gregaf> all the cleanups look fine, though
[21:44] <sagewk> k
[21:44] <elder> So yes, what Josh said. More than 4MB (or your rbd image object order) probably isn't beneficial, so that can be your maximum (and alignment) size.
[21:45] <elder> In between those two limits I'm not sure how much difference it's really going to make for rbd.
[21:45] <gregaf> sagewk: might not need to be super-complicated, but the request_kill caller is iterating over the session requests so refusing to pop them off that list will make it infinite loop from the dispatch thread
[21:45] <gregaf> I don't remember the structure lifetimes well enough to know if we can just pop them off the list without doing the rest of the request cleanup, or if the loop should check and see if they're killed, or what
[21:45] <jefferai> elder: joshd: I'll do some quick benchmarks of each
[21:45] <sagewk> whats strange is that i hammered on this way longer than i usually did and nothing got hung.
[21:46] <jefferai> each being 512bytes and 4MB
[21:46] <sagewk> i'll look
[21:46] <gregaf> well, maybe I'm missing something, but I wouldn't expect killing to actually happen very frequently — how are you inducing it?
[21:47] <sagewk> kernel client closes session without waiting for mds ops to commit. ffsb is the workload i've been using.
[21:48] * glowell (~glowell@c-98-210-224-250.hsd1.ca.comcast.net) has joined #ceph
[21:48] <sagewk> oh i bet the cleanups masked it. still using unallocated memory, but the killed flag is there instead of looking STL structures
[21:49] <gregaf> I seem to remember that we've had issues where the clients weren't actually dispatching the CEPH_SESSION_REQUEST_CLOSE message, maybe it ran into that?
[21:50] <gregaf> I don't think the cleanups could have masked it — I'm just looking at request_kill() and the journal_close_session() loop that calls it
[21:50] <gregaf> if we get into that loop with committing requests, are we not stuck?
[21:51] <joao> sagewk, gregaf, whenever you have the time, would be nice to have an okay on wip-mon-osd-create-fix's two top commits
[21:51] <joao> fairly simple ones
[21:51] <joao> just want to make sure it makes sense to someone other than me
[21:52] <sagewk> gregaf: hrm yeah.
[21:52] * aliguori (~anthony@32.97.110.59) has joined #ceph
[21:54] <gregaf> joao: not sure we want to prevent "01" from parsing as 1, I suspect some people with large clusters will find that useful so they can conflate the name and ID while keeping everything three digits
[21:55] <joao> gregaf, yeah, makes sense
[21:56] <gregaf> oh, and the first commit isn't EINVAL on incorrect UUID, it's EINVAL on not-a-uuid
[21:56] <gregaf> okay
[21:56] <gregaf> should fix that commit message ;)
[21:56] <joao> right
[21:57] <joao> preventing people from using +001 or makes sense though, no? :p
[21:57] <gregaf> yeah, works for me!
[21:57] <joao> although I'm yet to understand why would one even try that
[21:58] <gregaf> heh
[21:58] <joao> I'm just glad that strtol doesn't parse 10e10
[21:58] <gregaf> it did seem a bit extreme to me, but hey, it doesn't hurt us at all
[22:05] <joao> cool, all taken care of :)
[22:05] <joao> thanks
[22:07] * nhorman (~nhorman@hmsreliant.think-freely.org) Quit (Quit: Leaving)
[22:25] * madphoenix (~btimm@128.104.79.82) Quit (Quit: Leaving)
[22:26] * ron-slc (~Ron@173-165-129-125-utah.hfc.comcastbusiness.net) Quit (Quit: Leaving)
[22:32] * maxiz_ (~pfliu@114.245.250.70) has joined #ceph
[22:33] * loicd (~loic@magenta.dachary.org) Quit (Quit: Leaving.)
[22:33] * loicd (~loic@magenta.dachary.org) has joined #ceph
[22:34] * tziOm (~bjornar@ti0099a340-dhcp0628.bb.online.no) Quit (Remote host closed the connection)
[22:34] <phantomcircuit> i want to change the journal device for an osd
[22:35] <phantomcircuit> can i simply stop the osd
[22:35] <phantomcircuit> set the osd journal value
[22:35] <phantomcircuit> and restart it?
[22:36] <sagewk> stop, ceph-osd --flush-journal, change config, ceph-osd --mkjournal, start
[22:38] * maxiz (~pfliu@222.128.131.156) Quit (Ping timeout: 480 seconds)
[22:39] <phantomcircuit> sagewk, thanks
[22:39] * maxiz_ (~pfliu@114.245.250.70) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * glowell (~glowell@c-98-210-224-250.hsd1.ca.comcast.net) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * cephalobot (~ceph@ds2390.dreamservers.com) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * jpieper (~josh@209-6-86-62.c3-0.smr-ubr2.sbo-smr.ma.cable.rcn.com) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * The_Bishop_ (~bishop@f052099189.adsl.alicedsl.de) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * rweeks (~rweeks@c-24-4-66-108.hsd1.ca.comcast.net) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * sjustlaptop (~sam@68-119-138-53.dhcp.ahvl.nc.charter.com) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * chutzpah (~chutz@199.21.234.7) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * jlogan1 (~Thunderbi@2600:c00:3010:1:e12a:776f:2a6d:8a8) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * mdxi (~mdxi@74-95-29-182-Atlanta.hfc.comcastbusiness.net) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * joshd (~joshd@2607:f298:a:607:221:70ff:fe33:3fe3) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * match (~mrichar1@pcw3047.see.ed.ac.uk) Quit (resistance.oftc.net synthon.oftc.net)
[22:39] * dok (~dok@static-50-53-68-158.bvtn.or.frontiernet.net) Quit (resistance.oftc.net synthon.oftc.net)
[22:40] <phantomcircuit> journal on ssd is a massive improvement heh
[22:40] * maxiz_ (~pfliu@114.245.250.70) has joined #ceph
[22:40] * glowell (~glowell@c-98-210-224-250.hsd1.ca.comcast.net) has joined #ceph
[22:40] * cephalobot (~ceph@ds2390.dreamservers.com) has joined #ceph
[22:40] * jpieper (~josh@209-6-86-62.c3-0.smr-ubr2.sbo-smr.ma.cable.rcn.com) has joined #ceph
[22:40] * The_Bishop_ (~bishop@f052099189.adsl.alicedsl.de) has joined #ceph
[22:40] * rweeks (~rweeks@c-24-4-66-108.hsd1.ca.comcast.net) has joined #ceph
[22:40] * sjustlaptop (~sam@68-119-138-53.dhcp.ahvl.nc.charter.com) has joined #ceph
[22:40] * chutzpah (~chutz@199.21.234.7) has joined #ceph
[22:40] * jlogan1 (~Thunderbi@2600:c00:3010:1:e12a:776f:2a6d:8a8) has joined #ceph
[22:40] * dok (~dok@static-50-53-68-158.bvtn.or.frontiernet.net) has joined #ceph
[22:40] * match (~mrichar1@pcw3047.see.ed.ac.uk) has joined #ceph
[22:40] * joshd (~joshd@2607:f298:a:607:221:70ff:fe33:3fe3) has joined #ceph
[22:40] * mdxi (~mdxi@74-95-29-182-Atlanta.hfc.comcastbusiness.net) has joined #ceph
[22:43] <elder> phantomcircuit, /dev/null is an even faster journal.
[22:43] <elder> :)
[22:45] <rweeks> as opposed to tmpfs?
[22:45] * lurbs heard that /dev/null was also webscale.
[22:45] * rweeks goes off and develops an exascale /dev/null
[22:46] <phantomcircuit> ahah
[22:46] <phantomcircuit> elder, im sure i could convince a lot of "developers" that
[22:46] <phantomcircuit> s/that/of that/
[22:47] * gucki (~smuxi@80-218-125-247.dclient.hispeed.ch) Quit (Remote host closed the connection)
[22:48] <lurbs> rweeks: http://www.youtube.com/watch?v=b2F-DItXtZs
[22:49] <rweeks> wow that is awful voice synthesis
[22:49] <lurbs> Yep.
[22:50] <gregaf> you're showing off that mongodb mockery, aren't you
[22:50] <lurbs> Maybe a little.
[22:50] <ircolle> Why you gotta hate on xtranormal? ;-)
[22:52] <ircolle> Best. Line. Ever. "Does /dev/null support sharding?"
[22:54] <gregaf> /dev/null supports arbitrary sharding at arbitrary times! even better, there is no synchronization overhead no matter how many shards you create or how many shards any operation involves!
[22:57] <elder> Shards are the secret ingredient in the web scale sauce. They just work.
[22:57] <Robe> :D
[22:57] <Robe> When I hear sharding I always have to think of theo schlossnagles wise words
[22:58] <Robe> http://www.slideshare.net/postwait/velocity-2010-scalable-internet-architectures/30
[23:08] <nhm> elder: we should make a videogame that involves finding lost shards and plugging them int othe right slights to keep the filesystem going fast.
[23:08] <nhm> slots rather
[23:08] <nhm> ugh
[23:09] <elder> ShardRunner
[23:09] <rweeks> Angry Shards!
[23:09] <lurbs> Or a driving game, Race Condition.
[23:09] <elder> Call of Duty, Shard Ops
[23:09] <nhm> lurbs: that's awful and glorious.
[23:10] <Robe> nhm: I think that's already implemented
[23:10] <Robe> nhm: get a SSD, pull out the NAND-chips and zap the translation-table
[23:11] <rweeks> Shard Fortress 2
[23:11] <Robe> this way you get a funny and entertaining puzzle with ssdsize * underprovisioningfactor / eraseblocksize pieces!
[23:11] <nhm> lol
[23:12] <Robe> bonus points if it's a sandforce-based chip
[23:12] <Robe> then you can also deal with compression and AES ;)
[23:15] <nhm> Robe: I ended up modifying the IOR benchmark so that it would really quickly generate pseudo-random data just to get around sandforce compression. :)
[23:28] * dmick (~dmick@2607:f298:a:607:7596:c46:d062:a53d) has joined #ceph
[23:37] * aliguori (~anthony@32.97.110.59) Quit (Remote host closed the connection)
[23:47] * nwatkins (~nwatkins@c-50-131-197-174.hsd1.ca.comcast.net) has joined #ceph
[23:48] <nwatkins> How do I access performance counters tracked in client context (e.g. in the objector) without a admin socket?

These logs were automatically created by CephLogBot on irc.oftc.net using the Java IRC LogBot.