The Cisco CallManager service can crash because the service does not have enough resources such as CPU or memory to function. Generally, the CPU utilization in the server is 100 percent during this time. This document discusses the problems that cause high CPU utilization and their solutions.
Note: CPU spikes on a Cisco CallManager server can be normal depending on what service/application runs at the time. But, sustained CPU spikes are not good and can cause the CallManager service to terminate or issues like delayed dialtone, and so forth.
There are no specific requirements for this document.
The information in this document is based on Cisco CallManager 3.x and 4.x.
The information in this document was created from the devices in a specific lab environment. All of the devices used in this document started with a cleared (default) configuration. If your network is live, make sure that you understand the potential impact of any command.
Refer to Cisco Technical Tips Conventions for more information on document conventions.
The InsertCDR.exe application spikes and then results in high CPU utilization in the Cisco CallManager server.
This issue arises when the Cisco Call Detail Record (CDR) Insert service is activated and the CDR Enabled flag is not set to True.
Complete these steps in order to solve this issue.
Go to the Cisco CallManager Administration page.
Choose Service Parameter and select Cisco CallManager Service.
Change the CDR Enabled Flag value to True and restart the Cisco CallManager service.
The sql.exe application consumes high CPU which reduces performance of the Cisco CallManager Publisher server.
This issue can be caused by any third party billing application installed in the Cisco CallManager Publisher server. In order to resolve this, stop the application and restart the Cisco CallManager server.
This problem might arise when you use a short duration WAV file that loops for Music on Hold (MoH). The CPU utilization is caused by the constant looping that needs to take place when you play a short duration WAV file.
In order to resolve this issue, use a longer duration WAV file that loops less often. This decreases the CPU utilization.
In the Cisco CallManager 4.x cluster, aupair.exe causes high CPU utilization on the Cisco CallManager subscriber server. The Cisco Database Layer Monitor service (aupair.exe) monitors aspects of the database layer as well as CDRs. The database layer comprises a set of dynamic link libraries (DLLs) that provides a common access point for applications that need to access the database to add, retrieve, and change data.
Complete these steps in order to resolve this issue:
Restart the CDR Insert service on the publisher.
Restart the Database Layer Monitor service and the RIS Collector service on both the publisher and subscriber.
Use DBLHelper to reestablish a broken Cisco CallManager Cluster SQL Subscription.
Refer to Using DBLHelper to Reestablish a Broken Cisco CallManager Cluster SQL Subscription for more information about DBLHelper.
Sometimes, the performance monitoring service such as Realtime Information Server Data Collection (RISDC) can result in a CPU spike.
Complete these steps in order to restart the RISDC service and resolve this issue.
Go to the Cisco CallManager Administration page and choose Application > Cisco CallManager Serviceability.
Go to Tools > Control Center and select the appropriate CallManager server from the list of servers.
Select the Cisco RIS Data Collector service and click Restart.
Sometimes the Cisco CallManager response becomes very slow when a user answers calls, transfers calls, puts a call on hold, and even pulls a dial tone to the IP phone. When this happens, the tcdsrv.exe service causes high CPU which is responsible for telephony call dispatching.
Tcdsrv.exe is a service used for Telephony call dispatching and made for use with Attendant Console. If the CallManager does not use Attendant Console, stop the Telephony call dispatching service from the Control Center in the Cisco CallManager Serviceability page.