APPLIES TO:
Oracle Database – Enterprise Edition – Version 11.2.0.3 and later
Information in this document applies to any platform.
SYMPTOMS
The following symptoms are known to be attributed to this issue :-
a) Database is 2 or more node RAC
b) The following error or similar is seen in the database alert log(s) :-
ORA-00600: internal error code, arguments: [ORA_NPI_ERROR], [603], [ORA-00603: ORACLE server session terminated by fatal error
ORA-00569: Failed to acquire global enqueue.
Process ID:
Session ID: 105 Serial number: 1
c) The database is managed by ASM instance(s) and ORA-04031 errors are observed in the ASM alert log(s) for example on the LMD0 process :-
ORA-04031: unable to allocate 3768 bytes of shared memory (“sharedpool”,”unknown object”,”sga heap(1,0)”,”ges enqueues”)
If the same ORA-600 error is seen on a non-RAC/ASM managed database this article might still be appropriate, if the same ORA-600 error is
seen on a non-ASM managed database this article does not offer a solution. The ORA-04031 error might be on a different process and might
have a different failed allocation request.
CAUSE
The issue is correlated to :-
Bug:16502516 – ORA-600 ARGS [ORA_NPI_ERROR], ALONG WITH ORA-00603 & ORA-00569
This BUG is published but is in ‘Suspended’ status informing that the ORA-04031 errors must be resolved on the ASM side.
SOLUTION
Increase the SGA within the ASM instance(s). The amount of increase is subject to the machine environment so this article cannot give detailled
recommendations on how much of an increase will be needed.