Global Enqueue Services Deadlock causes node restart

The error message in the log is as follows:
Sun Sep 02 01:29:47 2012
Global Enqueue Services Deadlock detected. More info in file
/U01/app/11.1.0/diag/rdbms/zzbrac2/zzbrac21/trace/zzbrac21_lmd0_11802.trc.
Sun Sep 02 01:30:30 2012
Thread 1 advanced to log sequence 16189 (LGWR switch)
Current log #7 seq #16189 mem #0: + DATADG/zzbrac2/onlinelog/group_7.660.791891943
Current log #7 seq #16189 mem #1: + DATADG/zzbrac2/onlinelog/group_7.1693.791891943
Sun Sep 02 01:41:28 2012
Alter system set service_names = 'sys $ SYS. KUPC $ s_1_20902010315.zzbrac2 ', 'zzbrac2' SCOPE = memory sid = 'zzbrac21 ';
Alter system set service_names = 'zzbrac2' SCOPE = memory sid = 'zzbrac21 ';
Sun Sep 02 01:59:27 2012
Clearing Resource Manager plan via parameter
Sun Sep 02 01:59:35 2012
Thread 1 advanced to log sequence 16190 (LGWR switch)
Current log #8 seq #16190 mem #0: + DATADG/zzbrac2/onlinelog/group_8.3050.791891983
Current log #8 seq #16190 mem #1: + DATADG/zzbrac2/onlinelog/group_8.2893.791891985
Thread 1 advanced to log sequence 16191 (LGWR switch)
Current log #1 seq #16191 mem #0: + DATADG/zzbrac2/onlinelog/group_1.264.734243259
Current log #1 seq #16191 mem #1: + DATADG/zzbrac2/onlinelog/group_1.265.733693261
Sun Sep 02 02:04:12 2012
LMON (ospid: 11800) has not called a wait for 87 secs.
Sun Sep 02 02:04:43 2012
LMON (ospid: 11800) has not called a wait for 118 secs.
ERROR: LMON is not healthy and has no heartbeat.
ERROR: LMD0 (ospid: 11802) is terminating the instance.
LMD0 (ospid: 11802): terminating the instance due to error 482
Sun Sep 02 02:04:43 2012
System state dump is made for local instance
System State dumped to trace file/u01/app/11.1.0/diag/rdbms/zzbrac2/zzbrac21/trace/zzbrac21_diag_11788.trc
Instance terminated by LMD0, pid = 11802
 
Cause analysis:
At 01:29:47, The Global queue deadlock (caused by applications)
02 LMON (Global queue service monitoring process) cannot be called
LMON is unhealthy and has no heartbeat
LMD0 (Global queue service daemon) restarts the instance
 
It is preliminarily determined that the application causes a deadlock, resulting in LMON hang. LMD0 restarts the instance.