server time change, fault process startup failure

Version Mapping

NMS	OS	Database	NE
All	Windows	-	-

Symptom

After performing alarm maintenance operations, a user restarts the fault process. However, the process always stays in the starting state, which lasts more than 1 hour. As a result, the U2000 cannot monitor alarms.

Problem Identification

A user has changed the server time, leading to the failure to start the fault process.

Cause Analysis

                               Step 1     Check whether the database can be connected.

Frontline engineers confirm that the database can be connected.

                               Step 2     Check the database status in the Enterprise Manager.

The FaultDB database is normal.

                               Step 3     Obtain the following logs of the fault process generated that day:

server\log\iMAPFault_*.tar.gz

server\log\iMAPFault_*log

server\log\ iMAPMdp_p2_*tar.gz

server\log\ iMAPMdp_p2_*log

The timestamp of these logs is 20140515, not the current day. The reason is that a user changed the server time in the morning before performing alarm maintenance operations and restarting the fault process.

The fault process starts up successfully after the server time is changed back.

Root cause: The fault process depends on other U2000 processes. If the startup time of these processes is ahead of the server time after the change, the startup time will mismatch. As a result, the fault process cannot start up.

----End

Solution

                               Step 1     Change the server time to the current time.

                               Step 2     Restart the U2000.

----End

Suggestion and Summary

None