Start of change

6027-3941 [W] Excessive timer drift issue keeps getting detected (number nodes; above threshold of number). Resigning the cluster manager role (amIGroupLeader number, resignCCMgrRunning number).

Explanation

GPFS has detected an unusually large difference in the rate of clock ticks (as returned by the times() system call) between the cluster node and the cc-mgr node, and the rate difference between the cc-mgr node and different cluster nodes still remains.

User response

Check the error log for hardware or device driver problems that might keep causing timer interrupts to be lost or the tick rate to change dramatically.
End of change