Hi, It seems you OSD is stalling. When you do "ps aux", do you see any processes in the "D" state (beside cosd) And what does "ceph -s" show? Btw, in your ceph.conf, use "host = <hostname>" instead of the IP-Address of the machine. Wido On Wed, 2010-11-10 at 18:45 +0530, srimugunthan dhandapani wrote: > Hi all, > I used v0.22.2 ceph with latest git "ceph-standalone-client kernel" > (2.6.36-rc8) and when i try to mount i get the following error > mount error 5 = Input/output error > the dmesg call trace is as follows: > [ 199.099511] ceph: loaded (mds proto 32) > [ 227.730778] libceph: client4099 fsid 2bf77910-49e4-89cb-8ef0-b74ce321acc4 > [ 227.730999] libceph: mon0 192.168.1.79:6789 session established > [ 360.236062] INFO: task cosd:5004 blocked for more than 120 seconds. > [ 360.236068] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" > disables this message. > [ 360.236073] cosd D f4f7df20 0 5004 1 0x00000000 > [ 360.236080] f4f7df30 00200086 00000002 f4f7df20 f4f7dee8 c01673d0 > ffffffff 00000000 > [ 360.236092] c02383f0 f4f758d0 c085c940 f4f75b58 f4f75b5c 5adef48d > 00000035 c085c940 > [ 360.236104] c085c940 c085c940 00000000 c085c940 5ade69ed 00000035 > f4f758d0 00000001 > [ 360.236115] Call Trace: > [ 360.236128] [<c01673d0>] ? autoremove_wake_function+0x20/0x50 > [ 360.236135] [<c02383f0>] ? blkdev_writepage+0x0/0x20 > [ 360.236143] [<c05a21e5>] rwsem_down_failed_common+0x95/0xf0 > [ 360.236148] [<c05a2272>] rwsem_down_read_failed+0x12/0x20 > [ 360.236153] [<c05a22c7>] call_rwsem_down_read_failed+0x7/0x10 > [ 360.236157] [<c05a16ec>] ? down_read+0x1c/0x20 > [ 360.236163] [<c020f51f>] iterate_supers+0x5f/0xc0 > [ 360.236167] [<c0230400>] ? sync_one_sb+0x0/0x30 > [ 360.236171] [<c0230459>] sys_sync+0x29/0x60 > [ 360.236176] [<c0102fe3>] sysenter_do_call+0x12/0x28 > > my ceph.conf is attached. i was able to mount with the same ceph.conf > file in ceph0.21.1 and kernel 2.6.35 and also with ceph0.21.3 and > 2.6.35 kernel. > I got the same error with v0.22.2 and kernel 2.6.35 . So i updated > the kernel but it didn't help. > In ceph code , in src/cmds.cc and src/cosd.cc i commented out the line > // g_conf.profiler_running = IsHeapProfilerRunning; > for the compilation to pass. > > the steps i used the following commands as root: > > mkcephfs -c /etc/ceph/ceph.conf --allhosts -v > /etc/init.d/ceph -a start > modprobe ceph > mount -t ceph 192.168.1.79:/ /ceph > > Kindly point out what i am missing. > thanks, > mugunthan -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
