This is usually caused by a dispatching priority problem with the Xmanager
The operating system dispatching priority assigned to the Xmanager address space
can be of concern when collection is active.
Selected collection components executing within the Xmanager address space require
CPU cycles to be obtained at a comparable level to the application workload being generated.
We recommend that you assign Xmanager a dispatching priority equal to or higher
than the dispatching priority of the DB2 system services or database services address spaces.
If your site has address spaces that use DB2 and are assigned dispatching priorities higher
than DB2, then the dispatching priority assigned to Xmanager should be equal to or
higher than their dispatching priority.
Also, Xmanager needs a high dispatching priority to keep up with SQL trace data
generated when you use the CA Detector SQL trace facility.
If the Xmanager dispatching priority is lower than the address space
generating the activity, SQL trace data can be lost and reporting
can become inaccurate.
If this occurs, messages indicating that trace data is being lost appear on the syslog.