#ceph IRC Log

Index

IRC Log for 2010-10-20

Timestamps are in GMT/BST.

[13:06] -solenoid.oftc.net- *** Looking up your hostname...
[13:06] -solenoid.oftc.net- *** Checking Ident
[13:06] -solenoid.oftc.net- *** No Ident response
[13:06] -solenoid.oftc.net- *** Found your hostname
[13:06] * CephLogBot (~PircBot@fubar.widodh.nl) has joined #ceph
[13:48] * Yoric (~David@213.144.210.93) Quit (Quit: Yoric)
[13:49] * Yoric (~David@213.144.210.93) has joined #ceph
[15:40] * sentinel_e86 (~sentinel_@188.226.51.71) Quit (Quit: sh** happened)
[15:42] * sentinel_e86 (~sentinel_@188.226.51.71) has joined #ceph
[15:44] * sentinel_e86 (~sentinel_@188.226.51.71) Quit ()
[15:45] * sentinel_e86 (~sentinel_@188.226.51.71) has joined #ceph
[15:47] * sentinel_e86 (~sentinel_@188.226.51.71) Quit ()
[15:49] * sentinel_e86 (~sentinel_@188.226.51.71) has joined #ceph
[15:51] * sentinel_e86 (~sentinel_@188.226.51.71) Quit ()
[15:52] * sentinel_e86 (~sentinel_@188.226.51.71) has joined #ceph
[16:30] * gregorg_taf (~Greg@78.155.152.6) has joined #ceph
[16:30] * gregorg (~Greg@78.155.152.6) Quit (Read error: Connection reset by peer)
[16:31] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) has joined #ceph
[17:14] * sagelap (~sage@dhcp-63-161.cse.ucsc.edu) has joined #ceph
[17:19] * f4m8_ (~f4m8@lug-owl.de) Quit (Ping timeout: 480 seconds)
[17:24] * f4m8_ (~f4m8@lug-owl.de) has joined #ceph
[17:32] * greglap (~Adium@cpe-76-90-74-194.socal.res.rr.com) Quit (Quit: Leaving.)
[17:48] <jantje_> sagelap: you here?
[17:56] * greglap (~Adium@166.205.137.4) has joined #ceph
[18:07] <sagelap> jantje_: yeah
[18:12] * deksai (~deksai@dsl093-003-018.det1.dsl.speakeasy.net) has joined #ceph
[18:42] * Yoric (~David@213.144.210.93) Quit (Quit: Yoric)
[18:49] * joshd (~joshd@ip-66-33-206-8.dreamhost.com) has joined #ceph
[18:58] * sagelap (~sage@dhcp-63-161.cse.ucsc.edu) Quit (Ping timeout: 480 seconds)
[19:21] * deksai (~deksai@dsl093-003-018.det1.dsl.speakeasy.net) Quit (Ping timeout: 480 seconds)
[19:23] * sjust (~sam@ip-66-33-206-8.dreamhost.com) has joined #ceph
[19:33] * killmsft (~user@66.87.0.167) has joined #ceph
[19:38] <jantje_> sagewk: oh, i just mailed the mailinglist
[19:38] <jantje_> :)
[19:41] * sentinel_e86 (~sentinel_@188.226.51.71) Quit (Quit: sh** happened)
[19:43] * sentinel_e86 (~sentinel_@188.226.51.71) has joined #ceph
[19:58] * killmsft (~user@66.87.0.167) Quit (Ping timeout: 480 seconds)
[20:14] * greglap (~Adium@166.205.137.4) Quit (Ping timeout: 480 seconds)
[20:19] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[20:25] * greglap (~Adium@ip-66-33-206-8.dreamhost.com) has joined #ceph
[20:30] * greglap1 (~Adium@ip-66-33-206-8.dreamhost.com) has joined #ceph
[20:30] * greglap (~Adium@ip-66-33-206-8.dreamhost.com) Quit (Read error: Connection reset by peer)
[20:53] * greglap (~Adium@ip-66-33-206-8.dreamhost.com) has joined #ceph
[20:53] * greglap1 (~Adium@ip-66-33-206-8.dreamhost.com) Quit (Read error: Connection reset by peer)
[21:01] * conan (~conan@190.220.14.114) has joined #ceph
[21:01] <conan> hello everybody
[21:01] <gregaf> hi conan
[21:02] <yehudasa> hey conan!
[21:02] <conan> I'm new to ceph and I was wonderin if it can be tested by making a filesystem on a file
[21:02] <conan> Haven't found anything about it on the wiki
[21:02] <gregaf> not on a file exactly
[21:02] <gregaf> but you can put it and the journal in a directory of an existing fs if you don't want to dedicate a drive/partition to anything
[21:03] <conan> I would like to test it without having to use a partition on my machine
[21:03] <conan> gregaf: hum... but is it not necessary to use btrfs?
[21:04] <gregaf> for generic testing it'll run fine on any FS
[21:04] <gregaf> if you want failure recovery you'll need to use ext3 or similar
[21:04] <gregaf> it'll be way faster on btrfs
[21:05] <gregaf> in certain scenarios
[21:05] <gregaf> but Sage changed the journaling several months ago so it's got less stringent requirements on the underlying FS in exchange for somewhat lower performance
[21:06] <conan> so you're telling me I can test it with my /home partition which is an ext3?
[21:06] <gregaf> yep!
[21:06] <conan> cool :) tough the notice about ceph not being ready for production sites scares me a little bit
[21:07] <conan> are there any chance I loss something?
[21:07] <gregaf> yes
[21:07] <gregaf> what you're doing with it will significantly impact the odds
[21:08] <gregaf> RADOS and rbd are fairly safe; and on Ceph you should use a single MDS
[21:08] <gregaf> doing that will keep you pretty safe, especially barring failures
[21:09] <gregaf> the wiki provides a fair bit of info (http://ceph.newdream.net/wiki/Main_Page)
[21:09] <gregaf> start out with the "Simple test setup" section
[21:09] <conan> thanks gregaf, I guess I need more reading to do
[21:09] <gregaf> np :)
[21:09] <gregaf> thanks for taking a look, let us know if you encounter any problems
[21:11] <conan> sure
[21:12] * cmccabe1 (~cmccabe@dsl081-243-128.sfo1.dsl.speakeasy.net) Quit (Ping timeout: 480 seconds)
[21:13] * deksai (~deksai@dsl093-003-018.det1.dsl.speakeasy.net) has joined #ceph
[21:30] * jantje (~jan@paranoid.nl) has joined #ceph
[21:31] * monrad (~mmk@domitian.tdx.dk) has joined #ceph
[21:34] * jantje_ (~jan@paranoid.nl) Quit (synthon.oftc.net reticulum.oftc.net)
[21:34] * monrad-51468 (~mmk@domitian.tdx.dk) Quit (synthon.oftc.net reticulum.oftc.net)
[21:34] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) Quit (synthon.oftc.net reticulum.oftc.net)
[21:34] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) Quit (synthon.oftc.net reticulum.oftc.net)
[21:34] * pruby (~tim@leibniz.catalyst.net.nz) Quit (synthon.oftc.net reticulum.oftc.net)
[21:34] * DLange (~DLange@dlange.user.oftc.net) Quit (synthon.oftc.net reticulum.oftc.net)
[21:34] * conan (~conan@190.220.14.114) Quit (synthon.oftc.net reticulum.oftc.net)
[21:34] * todinini_ (tuxadero@kudu.in-berlin.de) Quit (synthon.oftc.net reticulum.oftc.net)
[21:35] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) has joined #ceph
[21:35] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) has joined #ceph
[21:35] * pruby (~tim@leibniz.catalyst.net.nz) has joined #ceph
[21:35] * DLange (~DLange@dlange.user.oftc.net) has joined #ceph
[21:35] * conan (~conan@190.220.14.114) has joined #ceph
[21:35] * todinini_ (tuxadero@kudu.in-berlin.de) has joined #ceph
[21:36] * DLange (~DLange@dlange.user.oftc.net) Quit (charon.oftc.net galapagos.oftc.net)
[21:36] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) Quit (charon.oftc.net galapagos.oftc.net)
[21:36] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) Quit (charon.oftc.net galapagos.oftc.net)
[21:36] * pruby (~tim@leibniz.catalyst.net.nz) Quit (charon.oftc.net galapagos.oftc.net)
[21:38] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) has joined #ceph
[21:38] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) has joined #ceph
[21:38] * pruby (~tim@leibniz.catalyst.net.nz) has joined #ceph
[21:38] * DLange (~DLange@dlange.user.oftc.net) has joined #ceph
[21:39] * DLange (~DLange@dlange.user.oftc.net) Quit (synthon.oftc.net galapagos.oftc.net)
[21:39] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) Quit (synthon.oftc.net galapagos.oftc.net)
[21:39] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) Quit (synthon.oftc.net galapagos.oftc.net)
[21:39] * pruby (~tim@leibniz.catalyst.net.nz) Quit (synthon.oftc.net galapagos.oftc.net)
[21:44] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) has joined #ceph
[21:44] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) has joined #ceph
[21:44] * pruby (~tim@leibniz.catalyst.net.nz) has joined #ceph
[21:44] * DLange (~DLange@dlange.user.oftc.net) has joined #ceph
[22:38] <jantje> hi !
[22:38] <gregaf> hi
[22:51] * cmccabe2 (~cmccabe@ip-66-33-206-8.dreamhost.com) has joined #ceph
[22:51] <cmccabe2> hi all. is anyone using sepia2 right now?
[22:51] <jantje> I can't check my work mail right now, but I was wondering if anyone replied yet? :)
[22:52] <jantje> cclien: what's sepia?
[22:52] <cmccabe2> it's a computer we use for testing stuff
[22:52] <yehudasa> cmccabe2: I'm using it
[22:53] <cmccabe2> hmm, are there any sepia machines I can get hold of today?
[22:53] <cmccabe2> yehudasa: or alternately, any machine where I can make a journal on a block device
[22:53] <yehudasa> let me check
[22:54] <jantje> Oh, i see
[22:54] <yehudasa> I think sepia26 is not being used
[22:54] <cmccabe2> k, thanks
[22:54] <gregaf> jantje: I'm looking at an issue that is probably related to yours; if it's not then yours is up next after that one :)
[22:54] <gregaf> no replies yet, though
[22:55] <jantje> gregaf: ok, thanks, let me know if you want me to test thing
[22:55] <jantje> *s
[22:55] <gregaf> will do!
[22:57] <jantje> I always break something .. btrfs, ceph, .. :)
[22:59] <jantje> I'm really going for the throughput, I'm bonding 4 gbit on 3 ceph servers and for now just one client (I should get some more ..., but only after they pass my basic test)
[22:59] <jantje> hehe :)
[22:59] <jantje> toys++
[23:00] * sagelap (~sage@ip-66-33-206-8.dreamhost.com) has joined #ceph
[23:00] * sagelap (~sage@ip-66-33-206-8.dreamhost.com) Quit ()
[23:01] <jantje> too bad you guys are in a different timezone
[23:03] <cmccabe2> jantje: I wrote a simple test of rados. It's in commit 602b4e667f1581455f998d9ae4256ccd2a93ff53
[23:04] <cmccabe2> jantje: in the lost_revert branch (although I should probably add just that test to the testing branch)
[23:04] <cmccabe2> jantje: I don't know if it would be helpful to you or not. It wasn't really designed to test throughput.
[23:05] <jantje> cool, i'll take a look at it
[23:05] <jantje> for now I just looped dd with different blocksizes
[23:06] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) Quit (Quit: Yoric)
[23:08] <jantje> oh, i see you'r bringing down the osd
[23:09] <jantje> I shoud make something similar for all my OSD's and MDSs/MONs
[23:09] <jantje> although it isn't a good idea, I run 6 OSD's and 3MDSs and 3MONs on 3 machines
[23:09] * allsystemsarego (~allsystem@188.27.167.113) Quit (Quit: Leaving)
[23:10] <jantje> that brings me to a question
[23:10] <jantje> is it possible to do something like
[23:10] <jantje> ceph tell osd 1 die or something like that
[23:11] <jantje> start/stop
[23:11] <gregaf> what are you trying to do?
[23:11] <jantje> guess not since the process would do :P
[23:11] <gregaf> you can just kill the process on the machine
[23:11] <jantje> i'd like to easily kill the remote process
[23:11] <cmccabe2> the pid is in a pid file
[23:12] <jantje> ok, good hint
[23:12] <jantje> i'll set something up tomorrow at work
[23:12] <cmccabe2> it is created by the daemon itself I think (not wrapper scripts) so it should always be there
[23:12] <yehudasa> e.g., ssh host kill `ssh host pgrep cosd`
[23:15] <jantje> can I use http to access the git repo ? (didn't try it yet, but we only have http at work..)
[23:15] <jantje> (proxy)
[23:17] <gregaf> not sure, but probably
[23:21] <cmccabe2> try "git clone http://ceph.newdream.net/git/ceph.git"

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