Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> A Real Stinker to Solve - EXPERTS APPLY WITHIN (DB Crashing)
Hi,
I have an interesting problem which has been happening for over a year on my DB instance. Oracle support has given up trying to solve it, so I have nothing to lose posting here.
I run a 9207 DB Instance on a Solaris 9 unix server (Sparc 64-bit). This is utilised by a Weblogic instance running various Java applications. Every few days the database shuts itself down and requires a manual restart.
Something is killing the LGWR process but no other logs are being generated other than what I have attached below.
I've ran truss against various unix processes (LGWR,CKPT,PMON), RDA, Statspack, init.ora changes etc. and nothing has shown up (all submitted to Metalink).
There is no pattern as to when the system crashes and I've managed to replicate the scenario on another unix server so it should not be a server failure etc.
Oracle are now asking me to apply patch 9208 but they admit that the problem is a bug and they have no idea what is causing the error. The frequency of the crash seems to be increasing also and I'm under pressure to solve.
Wondering where I can start next LOL.
Thanks in advance
John
LOGS
Alert log before and after crash / restart
Node name: ncluxora1 Release: 5.9 Version: Generic_118558-25 Machine: sun4u
/oracle/admin/dump/oranc1_arc1_20727.trc
Oracle9i Enterprise Edition Release 9.2.0.7.0 - 64bit Production
With the Partitioning, OLAP and Oracle Data Mining options
JServer Release 9.2.0.7.0 - Production
ORACLE_HOME = /oracle/ora92
System name: SunOS
Node name: ncluxora1 Release: 5.9 Version: Generic_118558-25 Machine: sun4u
Node name: ncluxora1 Release: 5.9 Version: Generic_118558-25 Machine: sun4u