Cluster Time Synch Fails; Crs Servers Respond Slowly In Ha Environment; Multilple Failovers With High Cpu Usage; Voip Monitor Problems - Cisco Unified Contact Center Express Servicing And Troubleshooting Manual

VoIP Monitor Problems

Cluster time synch fails

Symptom
Unified Communications Manager.
Error Message The Network Time Protocol Service on the CRS machine exits with a
message in the event log stating that the time difference is greater than the
threshold.
Possible Cause
Recommended Action
•
•

CRS Servers respond slowly in HA environment

Symptom
response time, the CRS system is unusable. No calls can get to agents.
Error Message LDAP error 51 in MADM logs. SDL and SQL errors on the Subscriber.
Possible Cause
Recommended Action

Multilple failovers with high CPU usage

Symptom
servers. When the CRS server boots up, the primary server begins to spike the CPU usage causing the
primary server to be unusable, and it fails over to the secondary server. The secondary server then starts
the same behavior and eventually fails back over to the primary server.
Error Message None.
Possible Cause
Recommended Action
VoIP Monitor Problems
This section contains the following troubleshooting tips on VoIP Monitor Problems
•
•
Cisco Customer Response Solutions Servicing and Troubleshooting Guide, Release 5.0(1)
8-76
The cluster time on the CRS machines (on the same cluster) is not synchronized with the
The NTP service has a built-in threshold of 1000 seconds. If the time difference between
the NTP client (CRS machines) and the server (Cisco Unified Communications Manager) is more
than this threshold, the NTP service does not correct the time.
Do the following:
Manually set the clock on each CRS machine in the cluster. The Time on the CRS machines should
be approximately the same as the time on the Unified Communications Manager machine.
Restart the Network Time Protocol service on the CRS machines, using the Service Control
Manager.
CRS servers in a high availabilitly (HA) environment respond very slowly. Because of the slow
Publisher and Subscriber databases are no synchronized.
Synch up the Publisher and Subscriber databases.
Multiple failovers occur in a high availability environment and there is high CPU usage on both
The problem can be caused by errors in the customer Java steps.
Check for custom Java steps in the scripting and remove them.
VoIP monitor does does not work correctly, page 8-77
CRS fails to start, page 8-77
Chapter 8 Troubleshooting Tips