Skip to Content
Former Member
Feb 16, 2012 at 07:17 PM

Monitoring DB2 transaction logs for high utilization


In a AIX/DB2 environment, we have 60 primary logs (no secondary) setup in our ECC instance, and they are approx 900MB each. The file system is more than large enough for the files (900MB x 60 = 54GB).

The problem is that sometimes processes run poorly and do not make frequent enough commits. I realize that the solution is to optimize those programs, but we need to be alerted when the log utilization increases past a threshhold.

The built in CCMS monitor CCMS_DB6_SPACE_FSY_LDI_pfcl

(DB2 Universal Database for NT/UNIX\Space management\database related file systems\log directory)

Only monitors the actual file system grown of the log_dir path. But, once those 60 logs are written, the actual filesystem capacity is static and never changes, even though the actual log utilization does. I can't find a CCMS monitor that actually shows that.