Simulating Oracle server & disk failures Oracle Database Tips by Donald Burleson

When performing Oracle disaster recovery exercises it is sometime necessary to simulate a catastrophic server failure.  This is a necessary part of the due diligence testing required for every Oracle DAB to ensure that their database is recoverable from a major hardware of instance failure.

WARNING - These command WILL ruin your server environment and they SHOULD NEVER be executed without taking a full backup.

• Instance failure - You can implode your Oracle database with high volume transactions to find out exactly where the database will be overcome with transaction loads.  Here are excerpts from an actual example of a stress test.  See the book "Oracle Benchmarking" for complete details on Oracle instance stress testing.
   
• Disk Failure - You can easily simulate a catastrophic disk failure by forcing your OS to delete the Oracle data files.

su - root
cd $ORACLE_BASE
rm -rf *

• Server failure - You can overload your server with transaction by submitting a recursive program.

root> cat /tmp/crash_server.ksh

#!/bin/ksh
nohup /tmp/crash_server.ksh >/dev/null 2>&1 &
nohup /tmp/crash_server.ksh >/dev/null 2>&1 &

root> nohup /tmp/crash_server.ksh >/dev/null 2>&1 &

 

If you like Oracle tuning, see the book "Oracle Tuning: The Definitive Reference", with 950 pages of tuning tips and scripts.  You can buy it direct from the publisher for 30%-off and get instant access to the code depot of Oracle tuning scripts.