#ceph IRC Log


IRC Log for 2011-08-24

Timestamps are in GMT/BST.

[0:01] * aliguori (~anthony@ Quit (Ping timeout: 480 seconds)
[0:21] * bchrisman (~Adium@ has joined #ceph
[0:35] <Tv> $ virsh start migraine
[0:35] <Tv> error: Failed to start domain migraine
[0:35] <Tv> error: Timed out during operation: cannot acquire state change lock
[0:35] <Tv> i get ETIMEDOUT when trying to use rbd via libvirt
[0:36] <Tv> still not clear why
[1:01] <Tv> Supported formats: raw cow qcow vdi vmdk cloop dmg bochs vpc vvfat qcow2 qed parallels nbd blkdebug sheepdog blkverify host_cdrom host_floppy host_device file tftp ftps ftp https http
[1:01] <Tv> there we go, i need to rebuild kvm
[1:13] * bchrisman (~Adium@ Quit (Quit: Leaving.)
[2:02] <Tv> anyone have the right sources.list for autobuilt ceph debs?
[2:02] <Tv> ahh wiki saves the day
[2:19] * huangjun (~root@ has joined #ceph
[2:32] * root (~root@ has joined #ceph
[2:46] * huangjun (~root@ Quit (Quit: Lost terminal)
[3:12] * cmccabe (~cmccabe@ has left #ceph
[3:13] * Dantman (~dantman@S01060023eba7eb01.vc.shawcable.net) Quit (Ping timeout: 480 seconds)
[3:36] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) has joined #ceph
[4:06] * jojy (~jojyvargh@70-35-37-146.static.wiline.com) Quit (Quit: jojy)
[4:09] * zyh (~zyh@ has joined #ceph
[4:11] * zyh (~zyh@ Quit ()
[4:18] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[4:50] * jojy (~jojyvargh@75-54-231-2.lightspeed.sntcca.sbcglobal.net) has joined #ceph
[4:50] * jojy (~jojyvargh@75-54-231-2.lightspeed.sntcca.sbcglobal.net) Quit ()
[4:59] * sage1 (~sage@dsl092-035-022.lax1.dsl.speakeasy.net) Quit (Quit: Leaving.)
[5:04] * sage (~sage@dsl092-035-022.lax1.dsl.speakeasy.net) has joined #ceph
[5:56] * ghaskins_ (~ghaskins@66-189-113-47.dhcp.oxfr.ma.charter.com) has joined #ceph
[6:03] * ghaskins (~ghaskins@66-189-113-47.dhcp.oxfr.ma.charter.com) Quit (Ping timeout: 480 seconds)
[7:12] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[7:46] * jim (~chatzilla@c-71-202-13-33.hsd1.ca.comcast.net) Quit (Read error: Connection reset by peer)
[7:46] * jim (~chatzilla@c-71-202-13-33.hsd1.ca.comcast.net) has joined #ceph
[7:49] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) has joined #ceph
[8:10] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[8:13] * root is now known as huangjun
[8:54] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) has joined #ceph
[9:33] * MarkN (~nathan@ Quit (Quit: Leaving.)
[11:19] * kikougirl (~kikougirl@212-198-248-35.rev.numericable.fr) has joined #ceph
[11:21] * kikougirl (~kikougirl@212-198-248-35.rev.numericable.fr) Quit ()
[11:32] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) Quit (Ping timeout: 480 seconds)
[13:29] * huangjun (~root@ Quit (Quit: Lost terminal)
[14:45] * aliguori (~anthony@cpe-70-123-132-139.austin.res.rr.com) has joined #ceph
[14:45] * Dantman (~dantman@S0106001731dfdb56.vs.shawcable.net) has joined #ceph
[17:01] * ajm (adam@adam.gs) has left #ceph
[17:49] * bchrisman (~Adium@c-98-207-207-62.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[17:57] * yehuda_hm (~yehuda@99-48-179-68.lightspeed.irvnca.sbcglobal.net) Quit (Ping timeout: 480 seconds)
[18:27] * gregorg (~Greg@ Quit (Ping timeout: 480 seconds)
[18:37] * jojy (~jojyvargh@70-35-37-146.static.wiline.com) has joined #ceph
[19:03] * bchrisman (~Adium@70-35-37-146.static.wiline.com) has joined #ceph
[19:25] * cp (~cp@c-76-102-0-145.hsd1.ca.comcast.net) has joined #ceph
[19:40] * cmccabe (~cmccabe@c-24-23-254-199.hsd1.ca.comcast.net) has joined #ceph
[20:03] <Tv> naturally i randomly pick the 2nd machine to run the libvirt test on to be one that does not do VT
[20:04] <Tv> i wish sepia cluster was multiple clusters, divided by whatever the physical hardware is
[20:06] <gregaf> aren't the sepia machines AMD nodes?
[20:06] <Tv> not all
[20:06] <Tv> hence me kvetching
[20:07] <gregaf> well I dunno how you map Intel processors to VT anyway, so no hardware clustering is going to solve that problem for you :x
[20:08] <Tv> intel & amd have fairly similar virtualization support, i just need a pair that's similar with each other, doesn't matter which kind
[20:08] * cmccabe (~cmccabe@c-24-23-254-199.hsd1.ca.comcast.net) Quit (Quit: Leaving.)
[20:09] <gregaf> I guess that's a requirement for KVM live switching?
[20:09] <Tv> i actually don't know if it'd let me go intel->amd
[20:09] <Tv> but in this case, the 2nd box had no cpu support for virtualization at all
[20:11] <gregaf> well that's not a strict requirement for KVM, is it? I assume it's just because it needs a different live setup so the transfer won't work
[20:12] <Tv> err, non-hardware virtualization is reaaaally slow
[20:12] <gregaf> Tv: hmm, the result from a greenlet is just whatever the function it calls returns, right?
[20:12] <Tv> gregaf: yup
[20:12] <gregaf> cool
[20:13] <Tv> hey *, free up sepia nodes if you have extras, i need to find one that's a recent enough AMD
[20:13] <gregaf> also, USE THE COMPUTER: http://www.youtube.com/watch?v=M1lVwpPLSx0 :D
[20:14] <Tv> ahh there we go, that looks good
[20:20] <sagewk> tv: still need nodes?
[20:20] <Tv> all good now it seems
[20:20] <sagewk> cool
[21:02] * cp (~cp@c-76-102-0-145.hsd1.ca.comcast.net) Quit (Quit: cp)
[21:06] * cp (~cp@c-76-102-0-145.hsd1.ca.comcast.net) has joined #ceph
[21:07] * cp (~cp@c-76-102-0-145.hsd1.ca.comcast.net) Quit ()
[21:26] <wido> hi
[21:27] <wido> currently the init script exits when one of the start actions fails
[21:27] <wido> for example, one btrfs filesystem won't mount, the init script exits and the remaining OSDs dont get started
[21:27] <wido> shouldn't it continue to start the rest?
[21:34] <sagewk> i could, altho it needs to still fail in the end
[21:37] <Tv> "could not open disk image rbd:rbd/migraine:conf=/home/ubuntu/ceph.conf: Permission denied"
[21:37] <Tv> harrumph
[21:37] <gregaf> cephx?
[21:38] <Tv> i copied over ceph.conf and ceph.keyring
[21:38] <Tv> i can run rbd on the command line on the other box
[21:39] <Tv> actually, i don't even know which host that error came from
[21:39] <Tv> bleh
[21:40] <Tv> ohhh "error reading conf file /home/ubuntu/ceph.conf"
[21:40] <Tv> how can that be
[21:41] <Tv> Aug 24 12:39:05 sepia48 kernel: [ 3339.600629] type=1400 audit(1314214745.956:17): apparmor="DENIED" operation="open" parent=1 profile="libvirt-8ab6d9ec-5f91-2745-7bc8-8ba7f4469183" name="/home/ubuntu/ceph.conf" pid=2920 comm="kvm" requested_mask="r" denied_mask="r" fsuid=104 ouid=1000
[21:41] <Tv> bleh
[21:41] <Tv> turning off apparmor ;)
[21:56] <wido> sagewk: Yeah, it should still exit with non-zero, but at least your remaining OSDs get started
[21:57] <sagewk> yeah, that'd definitely be better
[21:59] <Tv> sagewk: it seems i need to patch libvirt to support migrates on rbd
[21:59] <Tv> currently, it thinks it's a filename
[22:00] <gregaf> libvirt doesn't support shared hierarchy migrates?
[22:00] <Tv> it assumes that string is a valid filename
[22:00] <Tv> at least a parent of which it can stat
[22:01] <Tv> then it checks whether it's on nfs, etc
[22:01] <Tv> virStorageFileIsSharedFSType for the ugly details
[22:01] <gregaf> ugh
[22:02] <Tv> i'll add some ugly "if it begins with "rbd:", say yes" code and see what blows up next, but i need to do other things in the meanwhile
[22:53] * cmccabe (~cmccabe@c-24-23-254-199.hsd1.ca.comcast.net) has joined #ceph
[22:56] * cp (~cp@c-76-102-0-145.hsd1.ca.comcast.net) has joined #ceph
[23:03] <bchrisman> gregaf: there a commit id for that assertion fix? wanna make sure I'm runnign with it.
[23:03] <gregaf> *looks back through logs*
[23:03] <gregaf> I think it's 4a17d71c8e369c47c7fa110de1569190cd555657
[23:03] <gregaf> yeah
[23:03] <bchrisman> okie.. will check for that.
[23:04] <gregaf> headline "mdcache: handle unreadable null dentries"

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