[ILUG] logrotate hogs resources

[ccostelloe at flogas.ie]

I am no Linux expert, but generally I can coax it into doing what I want, but I don't know where to start with this one.

A 64 MB, 500 MHZ, lightly-loaded RedHat 7.3 server (bind, Apache), is unresponsive and has constant disk activity.  top shows logrotate as the culprit, ~ 90% CPU, 50 MB memory, and the majority of the minutes since the last boot-up.  Nothing else is much bigger than 1%.  It's not the only 7.3 server set up, but the only one with this problem.

I tried man logrotate, apparently it is configured via a file specified on it's command line, but I have no idea of what in the myriad of files in RedHat 7.3 is firing it up (my feeble guesses did not help).  I initially suspected logrotate was getting messed up by having to rotate the message log, but it is relatively modest in size.

Can anyone please give me a clue where to start?

Ciaran