#ceph IRC Log


IRC Log for 2012-05-25

Timestamps are in GMT/BST.

[0:10] <elder> Well Dayum.
[0:11] <elder> My newest, best, most questionable changes are fine after all.
[0:11] <elder> I want my afternoon back.
[0:12] * aliguori (~anthony@cpe-70-123-145-39.austin.res.rr.com) Quit (Remote host closed the connection)
[0:21] * joshd (~joshd@p4FECF022.dip.t-dialin.net) Quit (Ping timeout: 480 seconds)
[0:30] * mgalkiewicz (~mgalkiewi@toya.hederanetworks.net) Quit (Ping timeout: 480 seconds)
[0:30] * joshd (~joshd@aon.hq.newdream.net) has joined #ceph
[0:54] * BManojlovic (~steki@ Quit (Quit: Ja odoh a vi sta 'ocete...)
[0:57] * stass (stas@ssh.deglitch.com) has joined #ceph
[1:04] * stass (stas@ssh.deglitch.com) Quit (Quit: leaving)
[1:04] * stass (stas@ssh.deglitch.com) has joined #ceph
[1:06] * joshd (~joshd@aon.hq.newdream.net) Quit (Quit: Leaving.)
[1:11] * adjohn (~adjohn@50-0-164-218.dsl.dynamic.sonic.net) has joined #ceph
[1:24] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[1:36] <jmlowe> this probably isn't worth filing a bug report but the ceph.spec could stand to have libuuid-devel added as a dep
[1:46] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) has joined #ceph
[1:54] * Tv_ (~tv@aon.hq.newdream.net) Quit (Ping timeout: 480 seconds)
[2:03] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) Quit (Remote host closed the connection)
[2:05] <dmick> jmlowe: oddly, that's in 2253c0168dda83c508a68c2fedd97e967bf792a3
[2:05] <dmick> I wonder if it's been since removed....
[2:07] <dmick> ec776f4b859613b3ba785cddf078140f583cc447 I guess
[2:26] <dmick> jmlowe: emailing Holger to ask
[2:26] <dmick> drop me an email at dan.mick@inktank.com and I'll let you know if I hear back
[2:27] * Ryan_Lane (~Adium@ Quit (Quit: Leaving.)
[3:14] * Ryan_Lane (~Adium@c-98-210-205-93.hsd1.ca.comcast.net) has joined #ceph
[3:21] <nhm> gregaf: when you say "non-locally like teuthology suite", do you mean that the suite runs non-locally, or the things that teuthology spawns are non-local? I ask because I wrote a modified version of teuthology-suite that launches things from the command-line.
[3:21] <nhm> gregaf: ie, it's like the way teuthology-suite used to run.
[3:22] <nhm> it just runs from your own machine and dumps the data in an archive directory.
[3:50] * yoshi (~yoshi@p3167-ipngn3601marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[3:55] * joao (~JL@aon.hq.newdream.net) Quit (Remote host closed the connection)
[4:03] <gregaf> nhm: oh, that would be nice to have ?????I was referring to the way it schedules jobs on the teuthology server
[4:28] * chutzpah (~chutz@ Quit (Quit: Leaving)
[4:40] * CristianDM (~CristianD@host217.190-230-240.telecom.net.ar) Quit (Ping timeout: 480 seconds)
[4:41] * CristianDM (~CristianD@201-213-234-191.net.prima.net.ar) has joined #ceph
[4:42] <jmlowe> dmick: pulled down the 0.47.2 tarball in to a minimal centos 6.2 install and tried to build, installed required packages rpmbuild didn't pick up on the missing libuuid but configure did
[4:43] <dmick> certainly at one point we believed it to be a requirement
[4:44] <jmlowe> build fails but I believe I saw somebody working on that in the channel earlier
[4:46] <nhm> g'night all
[5:22] * guido_ (~guido@mx1.hannover.ccc.de) Quit (synthon.oftc.net oxygen.oftc.net)
[5:22] * guido_ (~guido@mx1.hannover.ccc.de) has joined #ceph
[5:24] * yoshi (~yoshi@p3167-ipngn3601marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[5:38] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[5:47] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[5:59] * adjohn (~adjohn@50-0-164-218.dsl.dynamic.sonic.net) Quit (Quit: adjohn)
[7:00] -coulomb.oftc.net- *** Looking up your hostname...
[7:00] -coulomb.oftc.net- *** Checking Ident
[7:00] -coulomb.oftc.net- *** No Ident response
[7:00] -coulomb.oftc.net- *** Found your hostname
[7:00] * CephLogBot (~PircBot@rockbox.widodh.nl) has joined #ceph
[7:00] * dmick (~dmick@aon.hq.newdream.net) Quit (Quit: Leaving.)
[7:32] * adjohn (~adjohn@50-0-164-218.dsl.dynamic.sonic.net) has joined #ceph
[7:32] * adjohn (~adjohn@50-0-164-218.dsl.dynamic.sonic.net) Quit ()
[7:34] * Qu310 (~qgrasso@ppp59-167-157-24.static.internode.on.net) Quit (charon.oftc.net kinetic.oftc.net)
[7:34] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) Quit (charon.oftc.net kinetic.oftc.net)
[7:34] * The_Bishop (~bishop@cable-86-56-102-91.cust.telecolumbus.net) Quit (charon.oftc.net kinetic.oftc.net)
[7:34] * ceph-test (~Runner@mail.lexinter-sa.COM) Quit (charon.oftc.net kinetic.oftc.net)
[7:34] * gregorg (~Greg@ Quit (charon.oftc.net kinetic.oftc.net)
[7:34] * pruby (~tim@leibniz.catalyst.net.nz) Quit (charon.oftc.net kinetic.oftc.net)
[7:34] * NaioN (~stefan@andor.naion.nl) Quit (charon.oftc.net kinetic.oftc.net)
[7:34] * vhasi (vhasi@vha.si) Quit (charon.oftc.net kinetic.oftc.net)
[7:35] * Qu310 (~qgrasso@ppp59-167-157-24.static.internode.on.net) has joined #ceph
[7:38] * cattelan_away is now known as cattelan_away_away
[7:39] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) has joined #ceph
[7:39] * The_Bishop (~bishop@cable-86-56-102-91.cust.telecolumbus.net) has joined #ceph
[7:39] * ceph-test (~Runner@mail.lexinter-sa.COM) has joined #ceph
[7:39] * gregorg (~Greg@ has joined #ceph
[7:39] * NaioN (~stefan@andor.naion.nl) has joined #ceph
[7:39] * vhasi (vhasi@vha.si) has joined #ceph
[7:39] * pruby (~tim@leibniz.catalyst.net.nz) has joined #ceph
[7:41] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) Quit (Ping timeout: 480 seconds)
[7:45] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) has joined #ceph
[7:52] * s[X]_ (~sX]@ppp59-167-154-113.static.internode.on.net) has joined #ceph
[7:55] * tjikkun (~tjikkun@2001:7b8:356:0:225:22ff:fed2:9f1f) Quit (Remote host closed the connection)
[8:17] * andy8788 (837010ac@ircip4.mibbit.com) has joined #ceph
[8:17] <andy8788> hi, anyone is having error 5 when mounting ceph?
[8:18] <andy8788> I'm having this error 5: Input/output error and I haven't been able to figure out why. Any help is appreciated!
[8:44] * verwilst (~verwilst@dD5769628.access.telenet.be) has joined #ceph
[8:50] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[8:57] * BManojlovic (~steki@ has joined #ceph
[8:58] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[9:08] * pmjdebruijn (~pascal@overlord.pcode.nl) Quit (Quit: leaving)
[9:25] * s[X]_ (~sX]@ppp59-167-154-113.static.internode.on.net) Quit (Remote host closed the connection)
[9:29] * steki-BLAH (~steki@bojanka.net) has joined #ceph
[9:34] * BManojlovic (~steki@ Quit (Ping timeout: 480 seconds)
[9:34] * The_Bishop (~bishop@cable-86-56-102-91.cust.telecolumbus.net) Quit (Quit: Wer zum Teufel ist dieser Peer? Wenn ich den erwische dann werde ich ihm mal die Verbindung resetten!)
[9:35] * BManojlovic (~steki@ has joined #ceph
[9:37] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[9:39] * steki-BLAH (~steki@bojanka.net) Quit (Ping timeout: 480 seconds)
[9:41] * CristianDM (~CristianD@201-213-234-191.net.prima.net.ar) Quit ()
[9:56] * Theuni (~Theuni@i59F76847.versanet.de) has joined #ceph
[10:05] * ao (~ao@ has joined #ceph
[10:32] * mgalkiewicz (~mgalkiewi@toya.hederanetworks.net) has joined #ceph
[10:37] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has left #ceph
[10:43] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) Quit (Quit: Leaving)
[10:46] * gohko (~gohko@natter.interq.or.jp) Quit (Read error: Connection reset by peer)
[10:52] * renzhi (~renzhi@raq2064.uk2.net) has joined #ceph
[10:53] * gohko (~gohko@natter.interq.or.jp) has joined #ceph
[10:54] <renzhi> I look up the API function int rados_objects_list_next(rados_list_ctx_t ctx, const char **entry, const char **key), the doc says key is the object locator, what is its purpose?
[10:54] <renzhi> like index key?
[10:55] <renzhi> If it is, which function do I use to set the key?
[11:04] * s[X]_ (~sX]@60-241-151-10.tpgi.com.au) has joined #ceph
[11:19] * s[X]_ (~sX]@60-241-151-10.tpgi.com.au) Quit (Remote host closed the connection)
[11:27] * s[X]_ (~sX]@60-241-151-10.tpgi.com.au) has joined #ceph
[11:29] * s[X]_ (~sX]@60-241-151-10.tpgi.com.au) Quit (Remote host closed the connection)
[11:46] * Theuni1 (~Theuni@i59F76847.versanet.de) has joined #ceph
[11:46] * Theuni (~Theuni@i59F76847.versanet.de) Quit (Read error: Connection reset by peer)
[12:36] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) has joined #ceph
[12:44] * Theuni1 (~Theuni@i59F76847.versanet.de) Quit (Ping timeout: 480 seconds)
[12:45] * nhorman_ (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) has joined #ceph
[12:50] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[13:00] * renzhi (~renzhi@raq2064.uk2.net) Quit (Quit: Leaving)
[13:42] * Theuni (~Theuni@i59F76847.versanet.de) has joined #ceph
[13:49] * Theuni (~Theuni@i59F76847.versanet.de) has left #ceph
[14:18] * s[X]_ (~sX]@60-241-151-10.tpgi.com.au) has joined #ceph
[14:21] * chuanyu (chuanyu@linux4.cs.nctu.edu.tw) Quit (Quit: leaving)
[14:58] * aliguori (~anthony@cpe-70-123-145-39.austin.res.rr.com) has joined #ceph
[15:21] * s[X]_ (~sX]@60-241-151-10.tpgi.com.au) Quit (Remote host closed the connection)
[15:23] * nhorman_ (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) Quit (Remote host closed the connection)
[16:10] * gregorg_taf (~Greg@ has joined #ceph
[16:10] * gregorg (~Greg@ Quit (Read error: Connection reset by peer)
[16:15] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) has joined #ceph
[16:35] * ao (~ao@ Quit (Quit: Leaving)
[16:45] * lofejndif (~lsqavnbok@82VAAD2IT.tor-irc.dnsbl.oftc.net) has joined #ceph
[17:09] * Tv_ (~tv@aon.hq.newdream.net) has joined #ceph
[17:24] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[17:28] * mgalkiewicz (~mgalkiewi@toya.hederanetworks.net) Quit (Quit: Leaving)
[17:28] * bchrisman1 (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[17:51] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[17:55] * Dennis2310 (~dennis@ has left #ceph
[18:08] * BManojlovic (~steki@ Quit (Quit: Ja odoh a vi sta 'ocete...)
[18:16] * Theuni (~Theuni@ has joined #ceph
[18:20] * Theuni (~Theuni@ Quit ()
[18:20] * Theuni (~Theuni@ has joined #ceph
[18:22] * Theuni (~Theuni@ Quit ()
[18:27] * psomas (~psomas@inferno.cc.ece.ntua.gr) Quit (Read error: Connection reset by peer)
[18:29] * bchrisman (~Adium@ has joined #ceph
[18:32] * joao (~JL@aon.hq.newdream.net) has joined #ceph
[18:39] * rturk (~rturk@aon.hq.newdream.net) has joined #ceph
[18:47] * Ryan_Lane (~Adium@c-98-210-205-93.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[18:52] * psomas (~psomas@inferno.cc.ece.ntua.gr) has joined #ceph
[18:57] * joshd (~joshd@aon.hq.newdream.net) has joined #ceph
[19:06] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) has joined #ceph
[19:09] * chutzpah (~chutz@ has joined #ceph
[19:09] * BManojlovic (~steki@ has joined #ceph
[19:18] * Ryan_Lane (~Adium@ has joined #ceph
[19:18] * BManojlovic (~steki@ Quit (Read error: Connection reset by peer)
[19:19] * BManojlovic (~steki@ has joined #ceph
[19:33] * cattelan_away_away is now known as cattelan_away
[19:33] * cattelan_away is now known as cattelan
[19:51] <joao> hey nhm
[19:51] <joao> are you around?
[19:54] * verwilst (~verwilst@dD5769628.access.telenet.be) Quit (Ping timeout: 480 seconds)
[19:54] <nhm> joao: yep, what's up?
[19:55] <joao> when you ran all the btrfs tests
[19:55] <joao> did you specified the nodiratime option?
[19:56] <nhm> joao: I ran with noatime, which I believe implies nodiratime
[19:57] <joao> we should find out if that is true
[19:58] <nhm> joao: Sage and I had a conversation about it a while back
[19:58] <joao> oh
[19:58] <joao> ok
[19:58] <joao> great then :)
[20:01] <nhm> joao: see Andrew Morton's comment in this thread: http://kerneltrap.org/node/14148
[20:02] * rturk (~rturk@aon.hq.newdream.net) has left #ceph
[20:03] <joao> oh
[20:03] <joao> cool
[20:03] <joao> thanks :)
[20:03] <joao> my mind is at ease now
[20:04] <jmlowe> what are the good btrfs options to use, I've just been using noatime?
[20:06] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) Quit (Quit: Konversation terminated!)
[20:06] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) has joined #ceph
[20:19] * MarkDude (~MT@c-71-198-138-155.hsd1.ca.comcast.net) has joined #ceph
[20:23] * mtk (~mtk@ool-44c35967.dyn.optonline.net) Quit (Remote host closed the connection)
[20:25] <gregaf> jmlowe: that's what we use
[20:27] * mtk (~mtk@ool-44c35967.dyn.optonline.net) has joined #ceph
[20:40] <jmlowe> space_cache and autodefrag look like possible good options, I wouldn't want to try it myself without having somebody else test it out thoroughly
[20:49] <nhm> jmlowe: -l 64k -n 64k might be worth using.
[20:51] <nhm> jmlowe: it seems to have pretty dramatically improved (though not eliminated) performance degradation over time on our test systems.
[20:52] <jmlowe> nhm: tradeoffs?
[20:55] <nhm> jmlowe: Good question. I haven't seen any issues caused by using it for Ceph. One of the guys more familiar with the internals of btrfs might be able to answer better.
[20:55] <jmlowe> that does break the pagesize barrier I imagine there are some consequences
[20:55] <guido_> I just set up a new ceph cluster, using cephx auth this time. When trying to write to the mounted cephfs, I get "Permission denied". Does anybody have an idea why?
[20:57] <nhm> jmlowe: Yes, though so far it's only improved my performance tests at multiple request sizes and over several days of aging tests.
[20:58] <jmlowe> nhm: how about autodefrag, looks like a double edged sword?
[20:58] <nhm> jmlowe: Haven't played with it myself yet.
[21:02] <nhm> jmlowe: here's a thread that mentions it: http://comments.gmane.org/gmane.comp.file-systems.ceph.devel/4490
[21:03] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) Quit (Ping timeout: 480 seconds)
[21:06] <jmlowe> hmm, compression
[21:08] <nhm> yeah, I'm just kind of ignoring it for now.
[21:09] <jmlowe> I suppose the question is will I be bottlenecked by io or cpu if I turn it on
[21:10] <jmlowe> and is that different than my current bottlenecks
[21:11] <nhm> what's your setup/performance right now?
[21:12] <jmlowe> I'm only using rbd, no plans to use ceph netowrk fs or radosgw
[21:13] <jmlowe> I figure I get roughly 600MBs
[21:13] <jmlowe> those rbd devices are used for backing kvm vm's
[21:15] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) has joined #ceph
[21:15] <jmlowe> are the rbd objects fully allocated or do they only use as many bytes as are written to them, eg an object that only had 512k written to it would be padded out to 4M or only 512k?
[21:16] <jmlowe> that would make a huge difference
[21:16] <darkfader> normally 512k, i dont know if it's different with rbd (guess no)
[21:17] <jmlowe> especially if people are using the vm's as dev machines and have tons of 1k files
[21:17] <iggy> it's like thin provioning yes
[21:18] <darkfader> that'd just be inside the image you're saving
[21:18] <iggy> which is something you want to watch out/account forin your planning
[21:38] * nhorman (~nhorman@99-127-245-201.lightspeed.rlghnc.sbcglobal.net) Quit (Quit: Leaving)
[21:40] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) Quit (Remote host closed the connection)
[21:41] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) has joined #ceph
[21:43] * andresambrois (~aa@r200-40-114-26.ae-static.anteldata.net.uy) has joined #ceph
[21:43] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) Quit (Read error: Connection reset by peer)
[21:49] * jmlowe (~Adium@c-71-201-31-207.hsd1.in.comcast.net) Quit (Quit: Leaving.)
[21:59] * andresambrois (~aa@r200-40-114-26.ae-static.anteldata.net.uy) Quit (Remote host closed the connection)
[22:01] <gregaf> guido_: so you mount successfully but get permission denied on write?
[22:03] <gregaf> if that's the case it sounds like you have the wrong capabilities on the OSDs
[22:07] * lofejndif (~lsqavnbok@82VAAD2IT.tor-irc.dnsbl.oftc.net) Quit (Ping timeout: 480 seconds)
[22:07] * jmlowe (~Adium@140-182-131-63.dhcp-bl.indiana.edu) has joined #ceph
[22:11] * lofejndif (~lsqavnbok@19NAAI2GV.tor-irc.dnsbl.oftc.net) has joined #ceph
[22:19] * aliguori (~anthony@cpe-70-123-145-39.austin.res.rr.com) Quit (Ping timeout: 480 seconds)
[22:23] <lxo> so what happens if we try to write data in the data pool, but there aren't any PGs in the data pool? apparently, requests just pile up and never get completed, rather than producing errors
[22:24] <lxo> (just found that out by accidentally starting an mds without my local patch to make data pool zero invalid)
[22:25] <gregaf> lxo: yeah???unfortunate, but we couldn't come up with a good fix and eventually we decided it wasn't worth it yet
[22:25] <elder> Anybody know how to give gitbuilder a kick? I pushded a branch about 5 minutes ago and nothing's happening.
[22:25] <elder> http://gitbuilder-kernel-amd64.ceph.newdream.net/
[22:29] <lxo> gregaf, so, would it be any different if I'd actually removed the data pool, rather than creating it empty?
[22:30] <gregaf> umm, yeah, I think it would detect nonexistence of the pool
[22:30] <gregaf> it's just when there are no PGs that it breaks
[22:30] * aliguori (~anthony@cpe-70-123-145-39.austin.res.rr.com) has joined #ceph
[22:30] <lxo> and fail rather than choose some other pool?
[22:30] * aa (~aa@r200-40-114-26.ae-static.anteldata.net.uy) has joined #ceph
[22:31] <lxo> I've been concerned about removing the pool and breaking something really hard ;-)
[22:31] <gregaf> not sure what you mean there
[22:31] <lxo> there are other pools, and it could try to assign the file to them rather than failing
[22:32] <lxo> (I'd prefer that it failed, which is why I left the data pool in place, but empty)
[22:32] <lxo> now, maybe there should be some way to set the default pool, and to set it to -1, so that it'd fail without having to remove the data pool
[22:33] <gregaf> the client doesn't itself choose which pool to put stuff in
[22:33] <gregaf> it gets that information from the MDS, which defaults to "data" for everything
[22:33] <gregaf> you can override it as you please
[22:33] <gregaf> but the client doesn't pick it
[22:33] <gregaf> that's an admin thing
[22:34] <elder> gitbuilder is going, thanks you if you had something to do with it.
[22:37] <lxo> gregaf, but, can the default pool be set to an invalid value such as -1?
[22:37] <lxo> this would enable me to drop this local patch I've been carrying for a long time
[22:38] <gregaf> lxo: hmm, I don't think it can
[22:38] <gregaf> why do you want to do that?
[22:53] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[22:54] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has left #ceph
[22:59] * yehudasa (~yehudasa@aon.hq.newdream.net) Quit (Remote host closed the connection)
[23:01] <lxo> gregaf, because of the bug that drops layout information from directories from time to time, I want to prevent files from being created unless they're created in the right pool
[23:03] * yehudasa (~yehudasa@aon.hq.newdream.net) has joined #ceph
[23:38] * CristianDM (~CristianD@host217.190-230-240.telecom.net.ar) has joined #ceph
[23:42] <gregaf> lxo: ah, I see
[23:42] * dmick (~dmick@aon.hq.newdream.net) has joined #ceph
[23:44] <gregaf> unfortunately the only way we could do that which would be appropriate for upstream would rely on not dropping layout information anyway ;)
[23:55] * aliguori (~anthony@cpe-70-123-145-39.austin.res.rr.com) Quit (Remote host closed the connection)
[23:56] <CristianDM> Know any other system than openstack that use rbd?
[23:57] <gregaf> work's been done on Cloudstack, and I think wido might even have done a third one but I'm not certain

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