#ceph IRC Log


IRC Log for 2010-08-26

Timestamps are in GMT/BST.

[1:31] * Osso (osso@AMontsouris-755-1-5-251.w86-212.abo.wanadoo.fr) Quit (Quit: Osso)
[4:50] * cowbar (3af5b6381a@dagron.dreamhost.com) Quit (Quit: bbiaf)
[6:05] * cowbar (bd2ac46502@dagron.dreamhost.com) has joined #ceph
[6:50] * f4m8_ is now known as f4m8
[7:07] <MarkN> Just had a kernel oops on the client mounting the ceph file system: http://pastebin.com/TuLTi0AS all the cluster nodes seem to be ok.
[7:08] <MarkN> kernel was 2.6.35 latest stable ceph. cheers
[7:25] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) Quit (Quit: WeeChat 0.2.6)
[7:25] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) has joined #ceph
[7:37] <kblin> sagewk: ah, good catch, the mkcephfs call failed
[7:38] <kblin> the /data dir seems to have some corruption that I blame on unstable btrfs support
[7:38] <kblin> but if I configure a different directory, the mkcpehfs call works and I can start cmon just fine
[7:47] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[9:00] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) Quit (Quit: Yoric)
[10:00] * allsystemsarego (~allsystem@ has joined #ceph
[10:37] * lidongyang (~lidongyan@ Quit (Read error: Connection reset by peer)
[10:44] * Osso (osso@AMontsouris-755-1-5-251.w86-212.abo.wanadoo.fr) has joined #ceph
[11:54] * MarkN_home (~nathan@c220-239-170-76.randw3.nsw.optusnet.com.au) has joined #ceph
[12:02] * lidongyang (~lidongyan@ has joined #ceph
[12:08] * allsystemsarego (~allsystem@ Quit (Quit: Leaving)
[12:08] * allsystemsarego (~allsystem@ has joined #ceph
[12:13] * Yoric (~David@ has joined #ceph
[13:08] <MarkN_home> anyone seen clients oops-ing on the stat system call ?
[13:12] * eternaleye_ (~eternaley@ has joined #ceph
[13:18] * eternaleye (~eternaley@ Quit (Ping timeout: 480 seconds)
[13:49] <todinini> wido: you here?
[13:57] * Yoric (~David@ Quit (Quit: Yoric)
[14:03] <todinini> wido: your libvrit packages doesn't support hvm. libvir: Domain Config error : unknown OS type hvm
[14:04] <todinini> is it a build problem or a runtime problem?
[14:11] * MarkN_home (~nathan@c220-239-170-76.randw3.nsw.optusnet.com.au) Quit (Quit: Leaving.)
[15:28] * Yoric (~David@ has joined #ceph
[15:45] * f4m8 is now known as f4m8_
[16:10] * johnl (~johnl@host81-142-82-205.in-addr.btopenworld.com) has joined #ceph
[17:22] <wido> todinini: i'm running with HVM
[17:22] <wido> are you sure kvm_amd or kvm_intel is loaded?
[17:22] <wido> and the support is enabled in your BIOS
[17:26] <wido> todinini: http://tracker.newdream.net/issues/381 can you verify that?
[17:47] * johnl (~johnl@host81-142-82-205.in-addr.btopenworld.com) Quit (Ping timeout: 480 seconds)
[18:01] * gregphone (~gregphone@ has joined #ceph
[18:17] <todinini> wido: the problem was, that my kvm/qemu binarys where in /bin/ the must be in /usr/bin
[18:20] <todinini> wido: shouldn't that be delta as well? rbd snap create --snap=alpha001 alpha
[18:21] <todinini> root@node43:~# rbd create --size 10240 delta
[18:21] <todinini> rbd image 'delta': size 10240 MB in 2560 objects
[18:21] <todinini> order 22 (4096 KB objects)
[18:21] <todinini> adding rbd image to directory...
[18:21] <todinini> creating rbd image...
[18:21] <todinini> done.
[18:21] <todinini> root@node43:~# rbd --pool rbd ls
[18:21] <todinini> delta
[18:21] <todinini> foo
[18:21] <todinini> myimage
[18:21] <todinini> root@node43:~# rbd snap create --snap=alpha001 alpha
[18:21] <todinini> list_snaps failed: No such file or directory
[18:24] <todinini> rbd.snap_add execution failed failed: No such file or directory
[18:24] <todinini> failed to create snapshot: No such file or directory
[18:24] <todinini> root@node43:~# rbd snap create --snap=alpha001 delta
[18:24] <todinini> root@node43:~#
[18:33] * Yoric (~David@ Quit (Quit: Yoric)
[18:54] * gregphone (~gregphone@ Quit (Ping timeout: 480 seconds)
[19:04] * sagewk (~sage@ip-66-33-206-8.dreamhost.com) Quit (Quit: Leaving.)
[19:13] <wido> todinini: that's weird, same operations are not working for me
[19:14] * sagewk (~sage@ip-66-33-206-8.dreamhost.com) has joined #ceph
[19:17] <wido> sagewk: is there a possibility I get permission to change my own issues?
[19:17] <sagewk> markn: you mean latest master from ceph-client.git? do you by chance have a commit id?
[19:18] <sagewk> wido: oh yeah, i'll fix that now
[19:18] <gregaf> wido: should be set now
[19:22] <wido> gregaf: I don't see a difference? tried to log out and in, didn't make a difference. For example, can't exit issue #81
[19:22] <wido> 381
[19:23] <gregaf> hmm, maybe that's not actually something you can do in Redmine
[19:23] <gregaf> just realized I don't see a way to edit the text either
[19:24] <wido> hmm, ok. that's to bad then
[19:25] <wido> sometimes I make a typo and would like to correct them
[19:25] <gregaf> ahaha
[19:26] <gregaf> click the "update" button and then the very small 'More" hyperlink at the top of that section
[19:26] <gregaf> wido: you can tell we have a lot of experience with Redmine here ;)
[19:28] <wido> gregaf: yes, i can see that. There is no "More" link in my case. I'm used to Trac though
[19:28] <wido> SVN + Trac
[19:29] <gregaf> hmm, you're set as a Reporter, which grants the "Edit issues" privilege
[19:29] * rklahn (~rklahn@ has joined #ceph
[19:31] <wido> gregaf: i think that's only for the "Ceph" project, not for the sub-projects, might not be recursive?
[19:31] <wido> i can edit #380, which is in the Ceph project
[19:32] <gregaf> ah
[19:32] <gregaf> try now
[19:33] <wido> yes, works now :)
[19:33] <wido> tnx
[19:34] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[20:53] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) Quit (Quit: Yoric)
[20:55] * eternaleye_ is now known as eternaleye
[20:58] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[21:56] <todinini> wido: in your issue your volume name is diffrent, maybe that is the problem?
[21:58] * johnl (~johnl@cpc3-brad19-2-0-cust563.barn.cable.virginmedia.com) has joined #ceph
[22:23] * eternaleye (~eternaley@ Quit (Quit: eternaleye)
[22:23] * eternaleye (~eternaley@ has joined #ceph
[23:00] * johnl (~johnl@cpc3-brad19-2-0-cust563.barn.cable.virginmedia.com) Quit (Ping timeout: 480 seconds)
[23:03] * Osso (osso@AMontsouris-755-1-5-251.w86-212.abo.wanadoo.fr) Quit (Quit: Osso)
[23:04] <kblin> haha, cephmkfs seems to corrupt my btrfs partitions, leaving me with directories I can't delete
[23:04] <kblin> er mkcephfs, of course
[23:05] <gregaf> pretty sure it's creating subvolumes
[23:05] <gregaf> which you can delete with a slightly more recent version of btrfs/the btrfs tools than is available in debian
[23:06] <kblin> ah
[23:06] <kblin> ubuntu maverick, here, but I guess they couldn't grab a working version from debian then :)
[23:07] <kblin> thanks for the info
[23:07] <gregaf> yeah
[23:08] <gregaf> I hit it every time I test and vstart tries to clean up the OSD directory :x
[23:08] <kblin> I'll follow up on that tomorrow, getting a kernel.org kernel and looking for more up-to-date btrfs-tools
[23:13] * Osso (osso@AMontsouris-755-1-5-251.w86-212.abo.wanadoo.fr) has joined #ceph
[23:51] <MarkN> sagewk: just got up now, I meant it was running .21.0, should i update to latest unstable and see how i go ?
[23:51] <gregaf> MarkN: which version of the kernel client?
[23:52] <gregaf> just the one that came baked into 2.6.35?
[23:52] <MarkN> yeah, all standard
[23:54] <MarkN> I only started seeing these when I added more clients to the fs
[23:54] * johnl (~johnl@cpc3-brad19-2-0-cust563.barn.cable.virginmedia.com) has joined #ceph

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