Ceph

Hello,

I try to set up a little test ceph cluster, based on the testing branch.
I encounter a problem using the cfuse client.

My system is in a given state, where when I try to mv a file. The cfuse app crashed:

:$ cd /mnt/ceph/
:/mnt/ceph$ ll
total 1
-rw-r--r-- 1 root        root          0 2010-05-21 11:02 test.txt

I mv one file

:/mnt/ceph$ sudo mv test.txt test.ttt
mv: cannot move `test.txt' to `test.ttt': Software caused connection abort

Then the mount point is not accessible anymore since cfuse crashed:

$ ll /mnt/
total 776
d????????? ? ?    ?           ?                ? ceph

Here is the core dump of cfuse:

Program terminated with signal 11, Segmentation fault.
#0  0x000000000046fe3a in Client::encode_dentry_release(Dentry*, MClientRequest*, int, int, int) ()
(gdb) bt
#0  0x000000000046fe3a in Client::encode_dentry_release(Dentry*, MClientRequest*, int, int, int) ()
#1  0x000000000047013b in Client::encode_cap_releases(MetaRequest*, MClientRequest*, int) ()
#2  0x000000000048009a in Client::send_request(MetaRequest*, int) ()
#3  0x0000000000483f01 in Client::make_request(MetaRequest*, int, int, Inode**, int, ceph::buffer::list*) ()
#4  0x00000000004904b2 in Client::_rename(Inode*, char const*, Inode*, char const*, int, int) ()
#5  0x0000000000490860 in Client::ll_rename(vinodeno_t, char const*, vinodeno_t, char const*, int, int) ()
#6  0x000000000045b959 in ?? ()
#7  0x00007f7272f0b5aa in fuse_session_loop () from /lib/libfuse.so.2
#8  0x000000000045d153 in ceph_fuse_ll_main(Client*, int, char const**) ()
#9  0x0000000000443c4d in main ()
(gdb) info threads 
  12 Thread 2227  0x00007f7272ce885c in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
  11 Thread 2228  0x00007f7271ef6f53 in poll () from /lib/libc.so.6
  10 Thread 2210  0x00007f7272ce885c in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
  9 Thread 2211  0x00007f7271ef6f53 in poll () from /lib/libc.so.6
  8 Thread 2205  0x00007f7271ef6f53 in poll () from /lib/libc.so.6
  7 Thread 2208  0x00007f7272ce8bc9 in pthread_cond_timedwait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
  6 Thread 2203  0x00007f7272ce8bc9 in pthread_cond_timedwait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
  5 Thread 2206  0x00007f7272ce8bc9 in pthread_cond_timedwait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
  4 Thread 2202  0x00007f7272ceb464 in __lll_lock_wait () from /lib/libpthread.so.0
  3 Thread 2204  0x00007f7272ce885c in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
  2 Thread 2201  0x00007f7272ce8bc9 in pthread_cond_timedwait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
* 1 Thread 2200  0x000000000046fe3a in Client::encode_dentry_release(Dentry*, MClientRequest*, int, int, int) ()

Notive that I have the following message when mounting ceph :

10.05.21 10:58:19.253804 7f5069bc2720 -- :/2665 messenger.start BUG: there are 1 already started that will now die!  call messenger.start() sooner.