Ilya, The SysRq is not doing anything as the kernel is hung. Btw, this is a
VirtualBox environment so I used the VBoxManage to send the SysRq commands.
Just to let you know, the system locksup and the only way out is a hard
reset.
On Mon, May 19, 2014 at 12:10 PM, Jay Janardhan <jay.janardhan at kaseya.com>wrote:
> Thanks for the response Ilya. I need to figure out how to use SysRq on my
> Mac. Meanwhile, here is the strace output and ceph version:
>
> *Ceph Version: *ceph version 0.80.1
> (a38fe1169b6d2ac98b427334c12d7cf81f809b74)
>
> *?*
> Note that IP 192.168.56.102:6789 is reachable from the client node. ?
>
> *?192.168.56.102 is a monitor node.*
>
> ?$ ?
> ceph status
>
> cluster df4f503a-04a9-4572-96d3-e31218592cfa
>
> health HEALTH_OK
>
> monmap e1: 1 mons at {ceph-node1=192.168.56.102:6789/0}, election
> epoch 2, quorum 0 ceph-node1
>
> osdmap e102: 3 osds: 3 up, 3 in
>
> pgmap v1604: 192 pgs, 3 pools, 1373 bytes data, 4 objects
>
> 22744 MB used, 202 GB / 236 GB avail
>
> 192 active+clean
>
> *strace output:*
>
> map(NULL, 8392704, PROT_READ|PROT_WRITE|PROT_EXEC,
> MAP_PRIVATE|MAP_ANONYMOUS|MAP_STACK, -1, 0) = 0x7f7899894000
>
> mprotect(0x7f7899894000, 4096, PROT_NONE) = 0
>
> clone(child_stack=0x7f789a093f70,
> flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID,
> parent_tidptr=0x7f789a0949d0, tls=0x7f789a094700,
> child_tidptr=0x7f789a0949d0) = 1629
>
> rt_sigprocmask(SIG_SETMASK, [PIPE], NULL, 8) = 0
>
> open("/etc/ceph/ceph.client.admin.keyring", O_RDONLY) = 3
>
> close(3) = 0
>
> open("/etc/ceph/ceph.client.admin.keyring", O_RDONLY) = 3
>
> fstat(3, {st_mode=S_IFREG|0644, st_size=63, ...}) = 0
>
> read(3, "[client.admin]\n\tkey = AQDb7HRTkB"..., 63) = 63
>
> close(3) = 0
>
> futex(0x7f789cc741a4, FUTEX_WAKE_PRIVATE, 2147483647) = 0
>
> brk(0x17e3000) = 0x17e3000
>
> futex(0x17a4b84, FUTEX_WAKE_OP_PRIVATE, 1, 1, 0x17a4b80, {FUTEX_OP_SET, 0,
> FUTEX_OP_CMP_GT, 1}) = 1
>
> futex(0x17a4b00, FUTEX_WAKE_PRIVATE, 1) = 1
>
> brk(0x17e1000) = 0x17e1000
>
> add_key(0x425208, 0x7fffce5220b0, 0x7fffce521fe0, 0x22, 0xfffffffe) = -1
> ENODEV (No such device)
>
> stat("/sys/bus/rbd", 0x7fffce522230) = -1 ENOENT (No such file or
> directory)
>
> rt_sigaction(SIGINT, {SIG_IGN, [], SA_RESTORER, 0x7f789c6d84c0}, {SIG_DFL,
> [], 0}, 8) = 0
>
> rt_sigaction(SIGQUIT, {SIG_IGN, [], SA_RESTORER, 0x7f789c6d84c0},
> {SIG_DFL, [], 0}, 8) = 0
>
> rt_sigprocmask(SIG_BLOCK, [CHLD], [PIPE], 8) = 0
>
> clone(child_stack=0, flags=CLONE_PARENT_SETTID|SIGCHLD,
> parent_tidptr=0x7fffce522060) = 1630
>
> wait4(1630, [{WIFEXITED(s) && WEXITSTATUS(s) == 1}], 0, NULL) = 1630
>
> rt_sigaction(SIGINT, {SIG_DFL, [], SA_RESTORER, 0x7f789c6d84c0}, NULL, 8)
> = 0
>
> rt_sigaction(SIGQUIT, {SIG_DFL, [], SA_RESTORER, 0x7f789c6d84c0}, NULL, 8)
> = 0
>
> rt_sigprocmask(SIG_SETMASK, [PIPE], NULL, 8) = 0
>
> rt_sigaction(SIGINT, {SIG_IGN, [], SA_RESTORER, 0x7f789c6d84c0}, {SIG_DFL,
> [], SA_RESTORER, 0x7f789c6d84c0}, 8) = 0
>
> rt_sigaction(SIGQUIT, {SIG_IGN, [], SA_RESTORER, 0x7f789c6d84c0},
> {SIG_DFL, [], SA_RESTORER, 0x7f789c6d84c0}, 8) = 0
>
> rt_sigprocmask(SIG_BLOCK, [CHLD], [PIPE], 8) = 0
>
> clone(child_stack=0, flags=CLONE_PARENT_SETTID|SIGCHLD,
> parent_tidptr=0x7fffce522060) = 1633
>
> wait4(1633, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0, NULL) = 1633
>
> rt_sigaction(SIGINT, {SIG_DFL, [], SA_RESTORER, 0x7f789c6d84c0}, NULL, 8)
> = 0
>
> rt_sigaction(SIGQUIT, {SIG_DFL, [], SA_RESTORER, 0x7f789c6d84c0}, NULL, 8)
> = 0
>
> rt_sigprocmask(SIG_SETMASK, [PIPE], NULL, 8) = 0
>
> open("/sys/bus/rbd/add_single_major", O_WRONLY) = -1 ENOENT (No such file
> or directory)
>
> open("/sys/bus/rbd/add", O_WRONLY) = 3
>
> write(3, "192.168.56.102:6789 name=admin,s"..., 87
>
>
> ?
>
>
>
>
> On Mon, May 19, 2014 at 10:16 AM, Ilya Dryomov <ilya.dryomov at inktank.com>wrote:
>
>> On Mon, May 19, 2014 at 5:42 PM, Jay Janardhan <jay.janardhan at kaseya.com>
>> wrote:
>> > (Sorry if this is a duplicate message - email server is acting up this
>> > morning).
>> >
>> >
>> > I'm following quick start guide and have a ceph cluster with three
>> nodes.
>> > When I try to map image to block device my command hangs. This seems
>> like a
>> > kernel hang as the only way I was able to get out is via a hard reset
>> of the
>> > image. The following is my configuration. Any help is greatly
>> appreciated.
>> >
>> > command on the ceph-client node (that hangs):
>> >
>> > $ sudo rbd map foo1 --pool rbd --name client.admin
>>
>> What's your ceph version (ceph --version)? Can you run 'rbd map' under
>> strace,
>> and when it hangs do SysRq+w followed by SysRq+t and send along strace and
>> SysRq outputs?
>>
>> Thanks,
>>
>> Ilya
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ceph.com/pipermail/ceph-users-ceph.com/attachments/20140519/ddd67475/attachment.htm>
