Power failure messages may only be successfully logged if you have some form of UPS, as they will be logged only
after the power has failed.
In the /etc/inittab file, there are some lines line this (this is how they appear on SuSE):
Looking at my /etc/init.d/powerfail script, I see these lines:
Quote:
POWERFAIL='THE POWER IS FAILED! SYSTEM GOING DOWN! PLEASE LOG OFF NOW!'
POWERFAILNOW='THE POWER IS FAILED! LOW BATTERY - EMERGENCY SYSTEM SHUTDOWN!'
POWERISBACK='THE POWER IS BACK'
|
So I'd imaging that these are the strings to look for. They may vary completely on Red Hat.
You could also check if /var is mounted as a synchronous filesystem (sync option in /etc/fstab). If it's not then the message, if logged at all, is unlikely to make it out of the write-cache for the disk.
Another suggestion: try passing the -i option to grep for case-insensitive matches.