#ceph IRC Log

Index

IRC Log for 2012-08-22

Timestamps are in GMT/BST.

[0:02] * EmilienM (~EmilienM@vau75-1-81-57-77-50.fbx.proxad.net) Quit (Remote host closed the connection)
[0:03] * Leseb (~Leseb@5ED01FAC.cm-7-1a.dynamic.ziggo.nl) Quit (Quit: Leseb)
[0:07] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) Quit (Quit: Leaving)
[1:08] * pentabular (~sean@adsl-71-141-232-146.dsl.snfc21.pacbell.net) has left #ceph
[1:10] * BManojlovic (~steki@212.200.243.134) Quit (Quit: Ja odoh a vi sta 'ocete...)
[1:31] * bchrisman (~Adium@108.60.121.114) Quit (Quit: Leaving.)
[1:34] * Merdin (18416e82@ircip3.mibbit.com) has joined #ceph
[1:34] <Merdin> hi
[1:34] <nhm> good evening
[1:34] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:34] <Merdin> WHAAHAHAHAHAAAA
[1:35] <nhm> awesome
[1:35] <Merdin> when
[1:35] <Merdin> will
[1:35] <Merdin> it
[1:35] <Merdin> come
[1:35] <Merdin> out
[1:35] <Merdin> please
[1:35] <Merdin> tell
[1:35] <Merdin> me
[1:35] <Merdin> pleaseeee
[1:35] <Merdin> i cant stand it
[1:35] <elder> Merdin, they canceled it. It will not be coming out.
[1:35] <Merdin> FUUU
[1:35] <nhm> and christmas too.
[1:35] <Merdin> FUUU
[1:35] <Merdin> FUUU
[1:35] <Merdin> FUU
[1:35] <Merdin> FUU
[1:35] <Merdin> FUU
[1:35] <Merdin> FUUU
[1:35] <Merdin> FUUU
[1:35] <Merdin> no ur lying
[1:35] <Merdin> XD
[1:36] <Merdin> nice try
[1:36] <Merdin> but i know it WILL come out
[1:36] <Merdin> u.u
[1:36] <Merdin> cant trick me
[1:36] <Merdin> =T
[1:36] <Merdin> :/
[1:37] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:37] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:37] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:37] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:37] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:37] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:37] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:37] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:37] <Merdin> i need a medic to tell me WHEN WILL MARI0 SE COME OUTTTTT
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:37] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> PLEEEEASSE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> tell ME WHEN Mari0 SE WILL COME OUT PLEASE
[1:38] <Merdin> PLEEEEASSE
[1:38] <Merdin> PLEEEEASSE
[1:38] <Merdin> PLEEEEASSE
[1:38] <Merdin> PLEEEEASSE
[1:38] <Merdin> PLEEEEASSE
[1:38] <Merdin> PLEEEEASSE
[1:38] <Merdin> PLEEEEASSE
[1:38] <Merdin> te-e-e-ell me
[1:38] <Merdin> if u dont u wont be here
[1:38] <Merdin> ./ ban
[1:38] <Merdin> ./ban elder
[1:38] <Merdin> damnit
[1:38] <ajm> wtf
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> damnit
[1:38] <Merdin> damnit
[1:38] <Merdin> damnit
[1:38] <Merdin> damnit
[1:38] <Merdin> damnit
[1:38] <Merdin> damnit
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:38] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> damnit
[1:39] <Merdin> damnit
[1:39] <Merdin> damnit
[1:39] <Merdin> damnit
[1:39] <Merdin> damnit
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:39] <Merdin> I WANT MARI0 SE
[1:40] <Merdin> I WANT MARI0 SE
[1:40] <Merdin> I WANT MARI0 SE
[1:40] <Merdin> I WANT MARI0 SE
[1:40] <Merdin> I WANT MARI0 SE
[1:40] <Merdin> I WANT MARI0 SE
[1:40] <Merdin> I WANT MARI0 SE
[1:40] <Merdin> I WANT MARI0 SE
[1:40] <Merdin> I WANT MARI0 SE PLEASE
[1:40] <Merdin> I WANT MARI0 SE PLEASE
[1:40] <Merdin> I WANT MARI0 SE PLEASE
[1:40] <Merdin> I WANT MARI0 SE PLEASE
[1:40] <Merdin> I WANT MARI0 SE PLEASE
[1:40] <Merdin> YOU guys suck dick now
[1:40] <Merdin> mine to be exact
[1:40] * Merdin (18416e82@ircip3.mibbit.com) Quit (Quit: http://www.mibbit.com ajax IRC Client)
[1:40] * Tv_ (~tv@2607:f298:a:607:38b3:897f:20fd:72b9) Quit (Read error: Operation timed out)
[1:40] <ajm> well then.
[1:40] <Cube> okay...
[1:41] <ajm> good to know where we stand I guess?
[1:41] <elder> MARI0 SE would be cool to have though.
[1:42] <nhm> damn, and I was finally bothering to go find a network op.
[1:42] <elder> I was looking for a medic.
[1:42] <ajm> i was just googling it, it actually looks kinda fun
[1:42] <ajm> but not quite THAT fun
[1:42] <elder> Evidently a medic would have been able to tell us all when it comes out.
[1:43] <nhm> speaking of fun, it's time to go mow the lawn.
[1:43] <ajm> allergies make mowing lawns distinctly !fun
[1:43] <nhm> or rather weeds. mow the weeds.
[1:43] <Cube> nhm: Before you go did you see the latest updates on the the spreadsheet?
[1:44] <nhm> Cube: nope, been trying to get the new test nodes on the network using a wireless bridge... What's the latest?
[1:45] <Cube> Went to 6 OSD's then started adding VM's Looks like the performance scales pretty linearly
[1:46] <Cube> Hitting network limitation, might need to switch it over to the 10gb interface :)
[1:50] * yoshi (~yoshi@p22043-ipngn1701marunouchi.tokyo.ocn.ne.jp) has joined #ceph
[1:51] <maelfius> Cube: are you using RBD, CephFS, or ?? for the VMs
[1:54] <Cube> RBD
[1:59] <maelfius> cube: cool.
[2:01] * loicd1 (~loic@brln-4d0ce674.pool.mediaWays.net) Quit (Quit: Leaving.)
[2:03] <andreask> Cube: qemu-rbd?
[2:03] <Cube> yes
[2:04] <andreask> and osds are also virtualization hosts?
[2:05] <Cube> No
[2:07] <andreask> and your disk setup on osds?
[2:09] * didders (~didders@142.196.239.240) has joined #ceph
[2:09] <Cube> yupp
[2:14] <andreask> but using osds as virtualization hosts is "save" when using qemu-rbd?
[2:15] * John (~john@2607:f298:a:697:bc77:5d6a:fa95:960) Quit (Ping timeout: 480 seconds)
[2:19] <nhm> cube: yeah, definitely time to switch over to the 10G. :)
[2:20] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) Quit (Quit: Leaving.)
[2:22] * Cube (~Adium@12.248.40.138) Quit (Ping timeout: 480 seconds)
[2:29] * senner (~Wildcard@68-113-228-89.dhcp.stpt.wi.charter.com) Quit (Quit: Leaving.)
[2:44] <maelfius> if there anything special I need to do to get a layout (with cephfs) assigned to a directory? It seems to always return invalid argument when run against a directory, whereas it works fine on a file. If I perform a show_layout on any directory that isn't the root I get "layout not specified".
[2:45] <maelfius> sorry, not assigned using 'set_layout'
[3:01] <maelfius> ugh. wow, i am having a hard time being clear. using 'cephfs set_layout' (there clearer)
[3:21] * dmick (~dmick@38.122.20.226) Quit (Quit: Leaving.)
[3:36] * Ryan_Lane (~Adium@216.38.130.164) Quit (Quit: Leaving.)
[3:40] * maelfius (~mdrnstm@66.209.104.107) Quit (Quit: Leaving.)
[3:44] <joao> In case someone missed it, Disney's take on OSS: http://www.youtube.com/watch?v=AiVnMazRIII
[3:48] * glowell1 (~Adium@c-98-210-226-131.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[3:50] * senner (~Wildcard@24-196-37-56.dhcp.stpt.wi.charter.com) has joined #ceph
[3:50] * senner (~Wildcard@24-196-37-56.dhcp.stpt.wi.charter.com) Quit ()
[4:12] * chutzpah (~chutz@100.42.98.5) Quit (Quit: Leaving)
[4:22] * The_Bishop (~bishop@2a01:198:2ee:0:c57c:a03b:8941:5585) Quit (Quit: Wer zum Teufel ist dieser Peer? Wenn ich den erwische dann werde ich ihm mal die Verbindung resetten!)
[4:44] * exec (~defiler@109.232.144.194) Quit (Ping timeout: 480 seconds)
[5:05] * deepsa (~deepsa@117.203.5.239) has joined #ceph
[5:13] * Ryan_Lane (~Adium@c-67-160-217-184.hsd1.ca.comcast.net) has joined #ceph
[5:19] <Tobarja1> if my app supports both s3 and swift, is there a ceph-ish reason to choose one over the other?
[5:49] * deepsa_ (~deepsa@115.241.228.210) has joined #ceph
[5:50] * deepsa (~deepsa@117.203.5.239) Quit (Ping timeout: 480 seconds)
[5:51] * deepsa_ is now known as deepsa
[6:03] * Cube (~Adium@184.253.165.242) has joined #ceph
[6:11] <didders> does rbd multipath automatically?
[6:13] * Cube (~Adium@184.253.165.242) Quit (Quit: Leaving.)
[6:25] <didders> If I want to use rbd across multiple hypervisors for migration support, ocfs2 is the thing to use? Overall this is assuming I am using file based VMs? Is there another scenario? I read there is snapshot support in future and kvm could directly use the block device via virtio which is faster iirc; how would one fence this off? pacemaker-foo?
[6:26] <didders> looking at HA is difficult with libvirt imo because you have to diy, rather than something like ganeti which has it built in over multicast
[6:27] <didders> but of course, most of the API's and industry tools have been rally around libvirt so its hard to change horses now
[6:28] <womble> I think it's rather unfair to compare libvirt to ganeti -- they're different tools solving different problems.
[6:28] <womble> We just use Pacemaker to manage all our HA stuff, and it works well enough.
[6:28] <didders> probably
[6:28] <didders> ok so pacemaker is the ha tool of choice with libvirt
[6:29] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[6:29] <womble> I don't know about "in general", but it is for us.
[6:29] <womble> Not because it's particularly awesome, but just because everything else is so much worse, and we Know Pacemaker.
[6:30] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[6:42] * didders (~didders@142.196.239.240) Quit (Ping timeout: 480 seconds)
[6:48] * lxo (~aoliva@lxo.user.oftc.net) Quit (Ping timeout: 480 seconds)
[6:48] * lxo (~aoliva@lxo.user.oftc.net) has joined #ceph
[7:05] * exec (~defiler@109.232.144.194) has joined #ceph
[8:04] * bchrisman (~Adium@c-76-103-130-94.hsd1.ca.comcast.net) has joined #ceph
[8:20] * grk (~grk@staticline56607.toya.net.pl) has joined #ceph
[8:25] * grk (~grk@staticline56607.toya.net.pl) Quit ()
[9:05] * BManojlovic (~steki@91.195.39.5) has joined #ceph
[9:08] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[9:13] * tnt (~tnt@89.40-67-87.adsl-dyn.isp.belgacom.be) Quit (Read error: Operation timed out)
[9:29] * tnt (~tnt@212-166-48-236.win.be) has joined #ceph
[9:44] * loicd (~loic@p5B2C523C.dip.t-dialin.net) has joined #ceph
[9:45] * verwilst (~verwilst@d5152FEFB.static.telenet.be) has joined #ceph
[9:48] * Leseb (~Leseb@193.172.124.196) has joined #ceph
[9:53] * renzhi (~renzhi@180.169.73.90) Quit (Ping timeout: 480 seconds)
[10:01] * renzhi (~renzhi@66.197.225.53) has joined #ceph
[10:08] * jbd_ (~jbd_@34322hpv162162.ikoula.com) has joined #ceph
[10:30] * yoshi (~yoshi@p22043-ipngn1701marunouchi.tokyo.ocn.ne.jp) Quit (Remote host closed the connection)
[10:32] * Cube (~Adium@cpe-76-95-223-199.socal.res.rr.com) has joined #ceph
[11:14] * EmilienM (~EmilienM@vau75-1-81-57-77-50.fbx.proxad.net) has joined #ceph
[11:27] * renzhi (~renzhi@66.197.225.53) Quit (Ping timeout: 480 seconds)
[11:31] * grk (~grk@staticline56607.toya.net.pl) has joined #ceph
[11:40] * renzhi (~renzhi@180.169.73.90) has joined #ceph
[12:16] * grk (~grk@staticline56607.toya.net.pl) Quit (Quit: grk)
[12:16] * grk (~grk@staticline56607.toya.net.pl) has joined #ceph
[12:42] * nhorman (~nhorman@hmsreliant.think-freely.org) has joined #ceph
[12:53] * renzhi (~renzhi@180.169.73.90) Quit (Quit: Leaving)
[13:03] * morse (~morse@supercomputing.univpm.it) Quit (Remote host closed the connection)
[13:07] * morse (~morse@supercomputing.univpm.it) has joined #ceph
[13:33] * ao (~ao@85.183.4.97) has joined #ceph
[13:50] * deepsa (~deepsa@115.241.228.210) Quit (Ping timeout: 480 seconds)
[13:51] * deepsa (~deepsa@115.240.121.199) has joined #ceph
[13:59] * loicd (~loic@p5B2C523C.dip.t-dialin.net) Quit (Quit: Leaving.)
[14:04] * kblin (~kai@kblin.org) has joined #ceph
[14:04] <kblin> hi folks
[14:08] <kblin> I'm trying to play with ceph a bit on a debian squeeze machine, but as my root filesystem isn't btrfs (or ext4 for that matter), the quickstart guide fails
[14:08] <kblin> now, I've got a btrfs mounted on /vol, and I'd like to share stuff below /vol/share
[14:10] <kblin> I've set "osd data = /vol/share" in ceph.conf, but I'm getting "filestore(/vol/share) mount initial op seq is 0; something is wrong" from mkcephfs
[14:10] <kblin> what am I doing wrong?
[14:30] * Cube (~Adium@cpe-76-95-223-199.socal.res.rr.com) Quit (Quit: Leaving.)
[14:36] <NaioN> kblin: you're trying mkcephfs?
[14:37] <kblin> yeah
[14:37] <NaioN> ok
[14:37] <kblin> I can pastebin the full error if needed
[14:37] <NaioN> do you want mkcephfs to also format/mount the btrfs volumes or not?
[14:38] <NaioN> with mkcephfs --mkbtrfs (or something)
[14:38] <kblin> no, it's formatted and mounted already
[14:38] <NaioN> ok
[14:38] <kblin> I'd like to keep the ceph stuff in a subdirectory
[14:38] <NaioN> yeah that's a good idea
[14:38] <NaioN> we use /data/osd.X dirs
[14:39] <NaioN> and mount the volumes under the right /data/osd.X directory
[14:39] <NaioN> ok so you have a btrfs volume mounted under /vol/share?
[14:40] <NaioN> could you pastebin the ceph.conf and the error from mkcephfs?
[14:52] * loicd (~loic@brln-4d0ce674.pool.mediaWays.net) has joined #ceph
[14:57] <kblin> NaioN: sure, just a sec
[14:57] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) Quit (Ping timeout: 480 seconds)
[14:57] <kblin> http://cpaste.org/1203/ is the error output
[14:58] * prometheanfire (~promethea@rrcs-24-173-105-83.sw.biz.rr.com) has left #ceph
[14:58] <kblin> http://cpaste.org/1204/ is the ceph.conf
[14:58] <kblin> and /vol is the btrfs share that is mounted
[14:59] <NaioN> oh not /vol/share?
[15:00] <NaioN> I think it expects the mount under /vol/share and not under /vol
[15:02] <kblin> so I need a dedicated partition for ceph?
[15:02] <kblin> that's probably doable, but I wasn't aware of that requirement
[15:04] <NaioN> well if i'm correct it isn't a requirement
[15:04] <NaioN> but i think mkcephfs can't deal with it
[15:05] <NaioN> but besides that's the recommended setup
[15:05] <NaioN> an osd daemon per volume
[15:06] <NaioN> we use osd nodes with 24 disks per node and we mount the disks with xfs filesystem (no partitions) under /data/osd.X and run an osd daemon per disk
[15:07] <kblin> I see
[15:08] <kblin> well, at this stage I'm more looking at experimenting with ceph
[15:08] <NaioN> besides I wouldn't recommend the use of btrfs in your setup, you use a fairly old kernel :)
[15:08] <NaioN> in terms of btrfs...
[15:09] <kblin> eventually I'd like to use it as the shared data store for a set of number crunching machines
[15:09] <kblin> but as all the number crunching will take care in ramdisk, speed isn't much of an issue
[15:09] <NaioN> it sounds as an ideal case for ceph
[15:10] <kblin> the current setup is on nfs3, but I'd like to make this a bit more resilient
[15:10] <NaioN> hehe :)
[15:11] <kblin> I can do xfs, too, I just thought while I'm experimenting, I'd experiment with btrfs, too
[15:11] <NaioN> well in that case I would recommend using Ubuntu 12.04 LTS with Ceph 0.48.1 (stable release)
[15:11] <NaioN> I've done a lot of expirementing with btrfs but last time I checked it wasn't stable enough
[15:12] <NaioN> and later on the setup didn't crash anymore but the performance collapsed due to btrfs fragmantation
[15:12] <kblin> hm, this is a debian system, I've had the heck of a time trying to keep ubuntu 10.04 running with the way ubuntu kept breaking kernel features
[15:13] <NaioN> But I follow the list a bit and they have made progress on that point, but I'll wait a bit longer before trying aagin with btrfs
[15:13] <NaioN> hmmm never had any trouble with it
[15:13] <kblin> they've broken linux container support three times in the last two years
[15:13] <NaioN> we used debian a lot but Ubuntu with the LTS releases is really nice...
[15:14] <kblin> ABI changes all the time, on LTS releases
[15:14] <kblin> not quite what you expect to bite you on a "stable" system
[15:17] <kblin> hm, I've just realized that one of the other testing machines runs on 2.6.39, if that's any better in terms of kernel support
[15:17] <NaioN> a bit :)
[15:19] <NaioN> as you see ceph wants to use some nice features of btrfs, but some of them aren't yet (fully) available in your kernel, hence the warning
[15:22] <kblin> yeah, fair enough.
[15:22] <kblin> 2.6.32 is a bit on the old side
[15:22] <kblin> heck, it's a 2.6
[15:28] <kblin> I wonder how far off the next debian release is :)
[15:30] <kblin> anyway, thanks for the hints, I'll reformat /vol to xfs and give it a spin with osd data = /vol in the config
[15:31] <kblin> that should work for testing even if I've got two osds set up, right?
[15:51] <pmjdebruijn> Debian Wheezy will have 3.2.x like ubuntu 12.04 LTS
[15:51] <pmjdebruijn> at IIRC
[15:51] <pmjdebruijn> kblin: I'm wasn't aware LXC was usable yet?
[15:52] <pmjdebruijn> btw, how exactly did they break the ABI?
[15:54] <kblin> pmjdebruijn: I'd have to dig up the bug report. basically they broke cgroups to make some other program work that was buggy
[15:56] <kblin> I don't remember the details
[15:56] <NaioN> kblin: yeah that's should work
[15:57] <kblin> pmjdebruijn: but basically the way the bug reports were handled didn't improve my faith in the stability of LTS
[15:59] <pmjdebruijn> ok
[16:00] <pmjdebruijn> kblin: well, with every patch there is risk of regression :)
[16:01] <pmjdebruijn> we've had many times with debian, that issues didn't get fixed because they didn't want to risk regression
[16:01] <pmjdebruijn> we ended up locally patches up _many_ packages
[16:01] <pmjdebruijn> so I guess there are two sides to that coin
[16:02] <pmjdebruijn> kblin: the problem with cgroups were brand spanking new in 10.04, and little used at that time
[16:04] <kblin> well, the thing is that they were working, than they broke it, then they ignored bug reports about the breakage for a year
[16:05] <kblin> then they told everybody to use the backport kernels for this, then they broke those as well
[16:05] <pmjdebruijn> in the same way?
[16:05] <kblin> then finally some other user sat down and figured out what was wrong and provided a patch
[16:05] <darkfader> maybe it was decided that "originally broken feature works" is a regression? :)
[16:06] <kblin> well, apparently the breakage was required to make vsftp or proftp or somesuch work
[16:06] <pmjdebruijn> well vsftpd is rather major
[16:06] <pmjdebruijn> so I could see how they would prioritize that
[16:08] <kblin> sure, all fair, but looking at how often LTS updates the kernel, running a box where I depend on things in the kernel that apparently aren't tested by the ubuntu folks doesn't sound too smart
[16:11] <kblin> I don't run debian testing on production machines either ;)
[16:11] <pmjdebruijn> with something as large as the linux kernel, it's silly to think that 100% coverage test would be remotely possible
[16:11] <pmjdebruijn> but then again, that doesn't mean some problems could have been avoided
[16:12] <kblin> well, "network works in virtual machine" sounds testable
[16:13] <kblin> and it's not the breakage itself that left me with a bad feeling, but how things were handled afterwards
[16:13] <pmjdebruijn> I could see that
[16:13] <kblin> programs break, that's just life
[16:14] <kblin> but ignoring multiple bug reports about a kernel regression for a year, and then closing all of them telling people to use a backport kernel, that's just not the level of professionalism I expect from a distro I want to run on my servers
[16:14] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[16:14] <pmjdebruijn> kblin: Canonical's compeditors fix this by requiring you to have a couple thousand of dollars per server every year :)
[16:14] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) Quit ()
[16:15] <kblin> pmjdebruijn: I'm comparing to debian, mostly
[16:15] <pmjdebruijn> erhm... "pay" should have gone into that somewhere
[16:15] * dspano (~dspano@rrcs-24-103-221-202.nys.biz.rr.com) has joined #ceph
[16:15] <pmjdebruijn> kblin: yeah like I said, debian is extremely overconservative
[16:15] <pmjdebruijn> which has it's uses I guess
[16:15] <kblin> sure
[16:16] <pmjdebruijn> but their "stability" comes at a great cost
[16:16] <pmjdebruijn> for example fibrechannel multipathing was pretty much broken in lenny
[16:16] <pmjdebruijn> and e2fsprogs/blkid has some nasty devicemapper related bugs too
[16:16] <pmjdebruijn> all fixed upstream
[16:17] <kblin> sure
[16:17] <kblin> I'm running a backport kernel on one of the machines because that un-breaks memory limits for cgroups
[16:18] <pmjdebruijn> heh
[16:18] <pmjdebruijn> kblin: lenny's kernel was completely useless to us to begin with :)
[16:19] <kblin> anyhow, I'm completely aware that as a user, I want the latest features without any of my stuff breaking, and then I'd also like a different color on my free pony
[16:20] <pmjdebruijn> indeed
[16:20] <pmjdebruijn> red stripes for speed!
[16:28] <kblin> anyway, diverting back to ceph for a moment...
[16:28] <kblin> osd data probably should be a per-osd setting, right:
[16:28] <kblin> ?
[16:29] <NaioN> well doesn't have to
[16:29] <NaioN> you can use something like osd data = /data/osd.$i
[16:29] <NaioN> hmmm $id i mean
[16:29] <kblin> ah
[16:30] * rlr219 (43c87e04@ircip2.mibbit.com) Quit (Quit: http://www.mibbit.com ajax IRC Client)
[16:31] <kblin> thanks :)
[16:40] <kblin> ok, apparently I don't have an rbd module... :)
[16:40] <kblin> perhaps I should play with a more recent distro after all
[16:42] <NaioN> hehe :)
[16:42] <NaioN> from 2.6.38 upwardes if i'm correct
[16:43] <kblin> yeah, the machine with the 2.6.39 kernel has it
[16:43] <nhm> ugh, the interference from the fans in this server keep drowning out my wifi.
[16:50] * EmilienM (~EmilienM@vau75-1-81-57-77-50.fbx.proxad.net) Quit (Remote host closed the connection)
[16:55] * ao (~ao@85.183.4.97) Quit (Quit: Leaving)
[16:57] * elder (~elder@c-71-195-31-37.hsd1.mn.comcast.net) Quit (Ping timeout: 480 seconds)
[17:02] * senner (~Wildcard@68-113-228-89.dhcp.stpt.wi.charter.com) has joined #ceph
[17:06] * BManojlovic (~steki@91.195.39.5) Quit (Quit: Ja odoh a vi sta 'ocete...)
[17:15] * verwilst (~verwilst@d5152FEFB.static.telenet.be) Quit (Quit: Ex-Chat)
[17:16] <kblin> ok, mount.ceph seems to fail still, but I can create and map block devices, that works for me
[17:26] * loicd1 (~loic@brln-d9bad56f.pool.mediaWays.net) has joined #ceph
[17:30] * loicd (~loic@brln-4d0ce674.pool.mediaWays.net) Quit (Ping timeout: 480 seconds)
[17:30] * BManojlovic (~steki@212.200.243.134) has joined #ceph
[17:33] * didders (~didders@rrcs-71-43-128-65.se.biz.rr.com) has joined #ceph
[17:41] * elder (~elder@c-71-195-31-37.hsd1.mn.comcast.net) has joined #ceph
[17:41] * grk (~grk@staticline56607.toya.net.pl) Quit (Quit: grk)
[17:43] * tnt (~tnt@212-166-48-236.win.be) Quit (Ping timeout: 480 seconds)
[17:50] <elder> sagewk, you updated the testing branch, right?
[17:50] <elder> ceph-client/testing
[17:51] <elder> Looks like maybe a rebase of some kind?
[17:52] * Tv_ (~tv@2607:f298:a:607:38b3:897f:20fd:72b9) has joined #ceph
[17:54] * tnt (~tnt@89.40-67-87.adsl-dyn.isp.belgacom.be) has joined #ceph
[18:08] * senner (~Wildcard@68-113-228-89.dhcp.stpt.wi.charter.com) Quit (Ping timeout: 480 seconds)
[18:14] * gregaf (~Adium@38.122.20.226) Quit (Quit: Leaving.)
[18:15] * senner (~Wildcard@68-113-228-89.dhcp.stpt.wi.charter.com) has joined #ceph
[18:35] * deepsa_ (~deepsa@115.242.27.38) has joined #ceph
[18:36] * deepsa (~deepsa@115.240.121.199) Quit (Ping timeout: 480 seconds)
[18:36] * deepsa_ is now known as deepsa
[18:38] * senner (~Wildcard@68-113-228-89.dhcp.stpt.wi.charter.com) Quit (Ping timeout: 480 seconds)
[18:39] * senner (~Wildcard@68-113-228-89.dhcp.stpt.wi.charter.com) has joined #ceph
[18:40] * Cube (~Adium@12.248.40.138) has joined #ceph
[18:43] * gregaf (~Adium@2607:f298:a:607:6198:95e2:4274:ad1d) has joined #ceph
[18:45] <sagewk> elder: yep. just rebased again on top of the patches i sent to linus.
[18:45] * senner1 (~Wildcard@68-113-228-89.dhcp.stpt.wi.charter.com) has joined #ceph
[18:46] <elder> OK.
[18:46] <elder> I figured it out--all the additions were there, but in a different order.
[18:46] <elder> Another question sagewk.
[18:46] <elder> I am trying to create a snapshot using format 2.
[18:46] * maelfius (~mdrnstm@pool-71-160-33-115.lsanca.fios.verizon.net) has joined #ceph
[18:46] <elder> It's giving me an error from librbd.
[18:47] <elder> Is that due to using the wrong version of the rbd user space command, perhaps?
[18:47] <sagewk> the rbd command line tool is?
[18:47] <elder> Yes.
[18:47] * senner (~Wildcard@68-113-228-89.dhcp.stpt.wi.charter.com) Quit (Ping timeout: 480 seconds)
[18:47] <elder> librbd: Error reading mutable metadata: (2) No such file or directory
[18:47] <sagewk> could be.. dan might know more (he's not in yet)
[18:47] <sagewk> could be that it's not a format 2 image?
[18:47] <elder> I looked for him (here) and didn't see him, just thought I'd ask you.
[18:47] <elder> It is a format 2 image.
[18:48] <elder> Well, let me verify, but I created it with "--new-format"
[18:48] <elder> Which is a bad name, by the way.
[18:48] <elder> Oh, and the rados pool shows objects with the expected names for format 2.
[18:48] <elder> And it has an image id, which is format 2 only.
[18:49] * senner1 (~Wildcard@68-113-228-89.dhcp.stpt.wi.charter.com) Quit (Read error: Operation timed out)
[18:49] <sagewk> what command are you using?
[18:50] <elder> rbd snap create --snap i1s1 image1
[18:53] <elder> I'd like to look in the osd log to see more, but I'm not sure what to look for.
[18:53] <elder> I believe I set it up to run with lots of logging.
[18:55] <sagewk> elder: dunno, works fine for me
[18:55] <elder> I think I may have found something.
[18:55] <sagewk> $ ./rbd create foo --size 10000 --new-format
[18:55] <sagewk> $ ./rbd snap create --snap i1s1 foo
[19:00] <elder> Nope, those exact commands did not produce what I wanted. But I'm not using ./rbd, I'm using one that I downloaded from some back alley somewhere.
[19:00] <elder> How can I be sure I'm using a good version of the user space tools?
[19:00] * maelfius (~mdrnstm@pool-71-160-33-115.lsanca.fios.verizon.net) Quit (Quit: Leaving.)
[19:01] <elder> Ooh, I intended to add the ceph repository to my /etc/apt/sources.list but it's not there.
[19:01] <elder> Do you have the right line for that?
[19:02] <elder> (Or anyone else, for that matter)
[19:03] * mengesb (~bmenges@servepath-gw3.servepath.com) has joined #ceph
[19:12] * dabeowulf (dabeowulf@free.blinkenshell.org) has joined #ceph
[19:13] * nhorman (~nhorman@hmsreliant.think-freely.org) Quit (Ping timeout: 480 seconds)
[19:13] * deepsa (~deepsa@115.242.27.38) Quit (Ping timeout: 480 seconds)
[19:15] * jbd_ (~jbd_@34322hpv162162.ikoula.com) has left #ceph
[19:21] * maelfius (~mdrnstm@pool-71-160-33-115.lsanca.fios.verizon.net) has joined #ceph
[19:27] <SpamapS> FYI, a new ceph with radosgw built is uploaded to quantal and awaiting archive-admin review (because of new deps pulled in and new binary packages)
[19:29] * nhorman (~nhorman@hmsreliant.think-freely.org) has joined #ceph
[19:33] * chutzpah (~chutz@100.42.98.5) has joined #ceph
[19:34] * dmick (~dmick@38.122.20.226) has joined #ceph
[19:45] * EmilienM (~EmilienM@78.251.173.55) has joined #ceph
[19:45] <elder> dmick, do you know what I add to my sources.list file so I can pull in the latest built images for ceph user space?
[19:48] <Tv_> elder: http://ceph.com/docs/master/install/debian/#add-development-testing-packages
[19:49] <elder> Ahah!
[19:49] <elder> I knew I got that from somewhere.
[19:49] <joao> sagewk, I suppose you were talking about 'mon_recovery.py' ?
[19:49] <sagewk> yeah
[19:49] <joao> cool
[19:49] <Tv_> ooh broken commands
[19:49] <elder> I think I managed to get it though. I was pointing at the wrong URL.
[19:49] <elder> Plus I needed to import the public key.
[19:50] <sagewk> mikeryan: looks like your suite is running
[19:51] <mikeryan> yep, i launched it off a plana machine accidentally
[19:52] <mikeryan> it silently fails
[19:52] * stxShadow (~Jens@ip-78-94-238-69.unitymediagroup.de) has joined #ceph
[19:53] <sagewk> mikeryan: it's only silent if you fail to 'echo $?' :)
[19:53] <Tv_> putting non-zero exit statuses in my prompt was one of the best customizations ever
[19:56] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[20:36] <elder> Tv_, is there an easy fix to this:
[20:36] <elder> The following packages have unmet dependencies:
[20:36] <elder> ceph : Depends: libc6 (>= 2.15) but 2.11.3-3 is to be installed
[20:36] <elder> ...
[20:36] <Tv_> elder: are you trying to install ubuntu 12.04 packages on ubuntu 11.10?
[20:37] <elder> I'm on a UML image that I built using ceph src/script/build_debian_image.sh
[20:37] <elder> I'd be happy to rebuild the image using the Ubuntu package repository if that's what I need to do (and if I can figure out how...)
[20:37] <Tv_> elder: that's debian squeeze
[20:38] <Tv_> elder: it seems we do have packages for squeeze, perhaps..
[20:39] <Tv_> oh taht looks like it's only releases
[20:39] * Cube (~Adium@12.248.40.138) Quit (Ping timeout: 480 seconds)
[20:39] * stxShadow (~Jens@ip-78-94-238-69.unitymediagroup.de) has left #ceph
[20:41] * Leseb (~Leseb@193.172.124.196) Quit (Quit: Leseb)
[20:42] * Cube (~Adium@12.248.40.138) has joined #ceph
[20:46] <elder> I may be able to alter the debootstrap command to point at ubuntu precise. I'm going to give that a shot Tv_
[20:46] <Tv_> elder: i'm assuming you need this to be uml for kernel development?
[20:48] <elder> Yes.
[20:48] <elder> Much, much quicker.
[20:48] * stxShadow1 (~Jens@jump.filoo.de) has joined #ceph
[20:55] <nhm> god, sueprmicro has some shitty java utility for remote management too.
[20:56] <nhm> Apparently they've removed fan control speed from the bios to make it easier to access through their web/java shit.
[20:56] <nhm> how convenient.
[20:57] * andreask (~andreas@chello062178013131.5.11.vie.surfer.at) has joined #ceph
[20:58] * tnt_ (~tnt@110.179-67-87.adsl-dyn.isp.belgacom.be) has joined #ceph
[21:00] * tnt (~tnt@89.40-67-87.adsl-dyn.isp.belgacom.be) Quit (Ping timeout: 480 seconds)
[21:06] * stxShadow (~Jens@ip-78-94-238-69.unitymediagroup.de) has joined #ceph
[21:06] * stxShadow (~Jens@ip-78-94-238-69.unitymediagroup.de) Quit (Read error: Connection reset by peer)
[21:09] * Leseb (~Leseb@5ED01FAC.cm-7-1a.dynamic.ziggo.nl) has joined #ceph
[21:09] <Cube> nhm: I need to setup the 10gb interfaces on the burnupi nodes I am using. Do you remember how yours were configured?
[21:13] * stxShadow1 (~Jens@jump.filoo.de) Quit (Ping timeout: 480 seconds)
[22:00] <nhm> Cube: I just configured the interface in /etc/network/interfaces
[22:01] <Cube> Yeah, but is it on a different vlan?
[22:11] * lightspeed (~lightspee@fw-carp-wan.ext.lspeed.org) Quit (Ping timeout: 480 seconds)
[22:16] * lightspeed (~lightspee@2001:8b0:16e:1:216:eaff:fe59:4a3c) has joined #ceph
[22:27] <nhm> Cube: sorry, my wireless keeps dropping down here. I think the crazy fan speeds are causing interference. Anyway, Dan is probably the right guy to talk to about this.
[22:27] <nhm> I got it working once and haven't touched it since. :)
[22:28] <Cube> sounds good, thanks
[22:29] <Cube> think I may have got it working
[22:29] * bshah (~bshah@sproxy2.fna.fujitsu.com) Quit (Ping timeout: 480 seconds)
[22:31] <Tv_> Cube: 10gig is a different vlan from the 1gig "front"
[22:32] <Tv_> Cube: there are specific ip addresses you should use, based on what host you are on
[22:32] <Cube> that makes sense, im on burnupi56-59
[22:32] <Tv_> Cube: dan or me can help, Dan's been there last so he may have it fresher in his memory, but email him or me a list of your machines and you'll get back what ips and netmasks to use
[22:33] <Cube> thanks
[22:33] <Tv_> ooh actually,
[22:33] <Tv_> host burnupi56.back.sepia.ceph.com
[22:33] <Tv_> burnupi56.back.sepia.ceph.com has address 10.214.149.18
[22:33] <Tv_> all you need is the netmask, which is...
[22:35] <Tv_> it's /20 aka 255.255.240.0
[22:35] <Tv_> Cube: ask dns for the ip addresses
[22:35] <Tv_> Cube: and please rather ask for help setting it up and make a mess of the network
[22:36] <Tv_> err
[22:36] <Tv_> Cube: and please rather ask for help setting it up than make a mess of the network
[22:36] * fghaas (~florian@91-119-204-193.dynamic.xdsl-line.inode.at) has joined #ceph
[22:37] <Cube> Tv_ : okay, I have a couple questions, we can move this over to jabber
[22:38] <fghaas> hey guys -- anyone here who could let me know the argonout equivalent of "ceph class list"?
[22:38] <fghaas> argonaut (/me curses typos)
[22:40] <fghaas> doc is pretty sparse on classes, http://ceph.com/docs/master/architecture/ only lists them as a "todo" item :)
[22:47] * Leseb_ (~Leseb@5ED01FAC.cm-7-1a.dynamic.ziggo.nl) has joined #ceph
[22:47] * Leseb (~Leseb@5ED01FAC.cm-7-1a.dynamic.ziggo.nl) Quit (Read error: Connection reset by peer)
[22:47] * Leseb_ is now known as Leseb
[22:50] * nhorman (~nhorman@hmsreliant.think-freely.org) Quit (Quit: Leaving)
[22:56] * aliguori (~anthony@cpe-70-123-140-180.austin.res.rr.com) Quit (Remote host closed the connection)
[23:03] * danieagle (~Daniel@177.43.213.15) has joined #ceph
[23:04] <dmick> fghaas: I found the commit responsible: 7c04f81ca16d11fc5a592992a4462b34ccb199dc
[23:04] <dmick> git show shows some text
[23:04] * fghaas checks
[23:05] <dmick> not knowing for certain what class list did, I'd guess it's now just "ls <path>", where path si something like /usr/lib/rados-classes on the osd machines
[23:05] <dmick> there is still ceph-clsinfo which can introspect the lib files
[23:05] <dmick> (in a very simple way)
[23:06] <fghaas> as per http://ceph.com/docs/master/control/, "ceph class list" would have listed classed that are actually loaded
[23:06] <dmick> ah
[23:06] <dmick> I've used pmap on the osd process to get that info. not exactly succinct, but useful
[23:06] <nhm> So now lsi has some kind of crazy windows like gui for their raid controller bioses that's significantly worse than the old one. I don't know how they can put out such terrible software.
[23:07] <elder> nhm, you were using the word "shitty" a lot before.
[23:07] * tnt_ (~tnt@110.179-67-87.adsl-dyn.isp.belgacom.be) Quit (Ping timeout: 480 seconds)
[23:07] <dmick> shitty would be an improvement
[23:07] <elder> But then again, you were referring to SuperMicro.
[23:07] <fghaas> dmick: that's indeed a bit kludgy :)
[23:08] <dmick> there might be an osd command for it, not suer
[23:08] <fghaas> I naively tried "ceph osd class list", but that doesn't work
[23:08] <dmick> nice try tho
[23:08] <nhm> elder: I'm sure my profanity will only increase with time.
[23:09] <elder> I anxiously await it.
[23:11] <fghaas> dmick: but the only thing this would have listed would have been whatever in /usr/lib64/rados-classes is loaded by the osds, correct?
[23:11] <dmick> fghaas: I don't see anything. It seems like it would have to have been in ClassHandler, but I don't see any enumeration there
[23:11] <dmick> fghaas: and yes, I think so
[23:12] <dmick> pmap $(pgrep osd) | grep libcls isn't horrible. I guess there wasn't much perceived need?...
[23:15] <dmick> anyway, looks easy enough to add as an osd command if there's a good reason
[23:20] <didders> How do I add an osd?
[23:20] <fghaas> didders: http://ceph.com/wiki/OSD_cluster_expansion/contraction
[23:22] <fghaas> dmick: no, I'm just trying to wrap my head around classes a little bit better
[23:22] * aliguori (~anthony@cpe-70-123-140-180.austin.res.rr.com) has joined #ceph
[23:24] <rturk> fghaas: I think that John is working on new command reference docs right now
[23:24] <rturk> although our doc work on object classes is a bit in the future
[23:24] <rturk> let me check with him, maybe he has some details on ceph class list
[23:25] <dmick> rturk: it's gone, no details to be had
[23:25] <dmick> functionality was removed
[23:25] <rturk> oh :)
[23:25] <rturk> didders: also docs on adding OSDs at http://ceph.com/docs/master/ops/manage/grow/
[23:26] <dmick> apparently once upon a time the monitors coordinated class loading; now it's up to the OSD
[23:26] * The_Bishop (~bishop@e179002234.adsl.alicedsl.de) has joined #ceph
[23:26] <rturk> although it's largely the same content???needs to be filled in a bit :)
[23:26] * EmilienM (~EmilienM@78.251.173.55) has left #ceph
[23:27] <fghaas> dmick: well yeah the mon class loading is gone, that much "ceph class list" tells you pretty verbatim :) but classes as such aren't going anywhere, are they?
[23:30] <fghaas> hey ross btw :)
[23:38] * nhmhome (~nh@184-97-251-210.mpls.qwest.net) Quit (Ping timeout: 480 seconds)
[23:38] * nhm (~nh@184-97-251-210.mpls.qwest.net) Quit (Ping timeout: 480 seconds)
[23:39] * The_Bishop (~bishop@e179002234.adsl.alicedsl.de) Quit (Quit: Wer zum Teufel ist dieser Peer? Wenn ich den erwische dann werde ich ihm mal die Verbindung resetten!)
[23:39] * The_Bishop (~bishop@e179002234.adsl.alicedsl.de) has joined #ceph
[23:46] <didders> I am trying to add osd... I alter the cruI am trying to add osd... I alter the crushmap and try to inject it but ceph status only show 3 osd? http://dpaste.com/790258
[23:46] <didders> s/crul/crushmap
[23:47] <fghaas> did you increase maxosd?
[23:50] <didders> no..
[23:53] <fghaas> didders, the wiki page does tell you to :)
[23:54] <dmick> fghaas: no, classes are in heavy use
[23:55] <dmick> they're loaded, by the osd as always; it's just that the monitor isn't involved in helping, apparently

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