Quote:
Originally Posted by
filosophizer
[...]I just want to be monitor how the log file is increasing and at what point ? During startup or shutdown ... etc..[...]
That depends. During normal operation the logs will stay small. Start and stop information does not turn out to be a problem for a cluster that is configured correctly. However there are occasions when the logs will grow fast and without limit. To give one example: when a cluster detects problems the clstrmgr.debug file might grow with 1MB per second. If it grows in /var and you don't realise in time that /var is flooded the affected node may be doomed.
Another example: hacmp.out is a seven day rolling log. With current HACMP versions it is rotated on midnight usually. Furthermore new hacmp.log files might turn up on a cluster (re-)start. That leads to that the log files are of different size. When the oldest hacmp.out file is deleted it might appear as if /var has more free space again. However that does not mean anything as the next hacmp.out file might become even bigger than the one that was deleted.