[Ocfs2-users] null pointer dereference

[Pawel]

Hi,
After upgrading ocfs2 my cluster is instable.

At least ones per week I can see:
kernel panic: Null pointer dereference  at 00048
o2dlm_blocking_ast_wrapper + 0x8/0x20 [ocfs2_stack_o2cb]
stack:
dlm_do_local_bast [ocfs2_dlm]
dlm_lookup_lockers [ocfs2_dlm]
dlm_proxy_ast_handler
add_timer
..

After that sometimes deadlock happens on another nodes. Entire cluster 
restart solve the issue.
I see in log:
(dlm_thread,7227,3):dlm_send_proxy_ast_msg:484 ERROR: 
ECB9442E19A94EAC896641BFADD55E4B: res M0000000000000001f411c900000000, 
error -107 send AST to node 4
(dlm_thread,7227,3):dlm_flush_asts:605 ERROR: status = -107
o2net: No connection established with node 4 after 10.0 seconds, giving up.
o2net: No connection established with node 4 after 10.0 seconds, giving up.
o2net: No connection established with node 4 after 10.0 seconds, giving up.
(dlm_thread,7227,4):dlm_send_proxy_ast_msg:484 ERROR: 
ECB9442E19A94EAC896641BFADD55E4B: res M0000000000000001f411c900000000, 
error -107 send AST to node 4
(dlm_thread,7227,4):dlm_flush_asts:605 ERROR: status = -107
o2cb: o2dlm has evicted node 4 from domain ECB9442E19A94EAC896641BFADD55E4B
o2cb: o2dlm has evicted node 4 from domain ECB9442E19A94EAC896641BFADD55E4B
o2dlm: Begin recovery on domain ECB9442E19A94EAC896641BFADD55E4B for node 4
o2dlm: Node 5 (he) is the Recovery Master for the dead node 4 in domain 
ECB9442E19A94EAC896641BFADD55E4B
o2dlm: End recovery on domain ECB9442E19A94EAC896641BFADD55E4B


Additionaly ~4 times per day I see:

ocfs2_check_dir_for_entry:2119 ERROR: status = -17
ocfs2_mknod:459 ERROR: status = -17
ocfs2_create:629 ERROR: status = -17


I currently use kernel 3.4.2
my filesystem has been created with:
-N 8-b 4096 -C 32768 --fs-features 
backup-super,strict-journal-super,sparse,extended-slotmap,inline-data,metaecc,xattr,indexed-dirs,refcount,discontig-bg,unwritten,usrquota,grpquota

Could you tell me what could make my system instable? Which feature ?

Thanks for any  help

Pawel