CDR Repository Manager

This section does not apply to IM and Presence Service.

Use the CDR Management Configuration window to set the amount of disk space to allocate to call detail record (CDR) and call management record (CMR) files, configure the number of days to preserve files before deletion, and configure up to three billing application server destinations for CDRs. The CDR Repository Manager service repeatedly attempts to deliver CDR and CMR files to the billing servers that you configure in the CDR Management Configuration window until it delivers the files successfully, until you change or delete the billing application server on the CDR Management Configuration window, or until the files fall outside the preservation window and are deleted.

Note

To access the Enterprise Parameters Configuration window, open Cisco Unified Communications Manager Administration and choose System > Enterprise Parameters. The CDR File Time Interval parameter specifies the time interval for collecting CDR data. For example, if this value is set to 1, each file will contain 1 minute of CDR data (CDRs and CMRs, if enabled). The external billing server and CAR database will not receive the data in each file until the interval has expired, so consider how quickly you want access to the CDR data when you decide what interval to set for this parameter. For example, setting this parameter to 60 means that each file will contain 60 minutes worth of data, but that data will not be available until the 60-minute period has elapsed, and the records are written to the CAR database. and the CDR files are sent to the configured billing servers. The default value equals 1. The minimum value specifies 1, and the maximum value specifies 1440. The unit of measure for this required field represents a minute.

Both the CDR Agent and the CDR Repository Manager process files with an interval that is independent of the CDR File Time Interval. The CDR Repository Manager sends all existing CDR files to the billing application servers, sleeps for 6 seconds before checking the new files to send, and continues that 6-second interval. If the destination (the external billing application servers) does not respond, the system attempts the process again by using a doubled length of the sleep interval (12 seconds). Each delivery failure results in double the sleep time (6, 12, 24, 48,and so on, seconds) until 2 minutes occurs, then stays at 2-minute intervals until successful delivery occurs. After successful delivery, the 6-second interval automatically resumes.

Users cannot configure the 6-second processing time, with the sleep time interval doubling in case of failure. Users can configure only the CDR File Time Interval enterprise parameter. No alert gets sent after the first file delivery failure. By default, the system generates the CDRFileDeliveryFailed alert after the second delivery failure of the Cisco CDR Repository Manager service to deliver files to any billing application server. You can configure the alert to send you an e-mail or to page you. For information on configuring alerts, see the "Working with Alerts" chapter in the Cisco Unified Real-Time Monitoring Tool Administration Guide.

The system generates the CDRFileDeliveryFailureContinues syslog alarm upon subsequent failures to deliver the files to the billing application servers.

The CDR Agent behaves in almost the same manner. First, it sends all the existing CDR files to the publisher. If no additional files to send exist, the CDR Agent sleeps for 6 seconds before checking for new files. Each delivery failure results in the immediate change of the sleep interval to 1 minute, then says at 1-minute intervals until successful delivery. After the first successful delivery of files, the 6-second interval resumes.

The system sends no alert after the first file delivery failure by the CDR Agent. By default, the system generates the CDRAgentSendFileFailed alert after the second delivery failure of the CDR Agent. You can configure the alert to send you an email or to page you. For information on configuring alerts, see the "Working with Alerts" chapter in the Cisco Unified Real-Time Monitoring Tool Administration Guide

The system generates the CDRAgentSendFileFailedContinues syslog alarm upon subsequent failures to deliver the files.

If you need to start or restart the file transfer timer, you can restart the Cisco CDR Repository Manager or CDR Agent process by going to the Cisco Unified Serviceability window and selecting Tools > Control Center > Network Services.

When you enable the file deletion based on high water mark parameter, the CDR repository manager service monitors the amount of disk space that CDR and CMR files use. If disk usage exceeds the high water mark that you configure, the system purges the CDR and CMR files that have been successfully delivered to all destinations and loaded into the CAR database (if CAR is activated) until the disk space reaches the low water mark or the system deletes all successfully delivered files. If disk usage still exceeds the high water mark after the system deletes all successfully delivered files, it does not delete any more files, unless the disk usage still exceeds the disk allocation that you configure. If the disk usage still exceeds the disk allocation that you configure, the system purges files beginning with the oldest, regardless of whether the files fall within the preservation window or have been successfully delivered, until the disk usage falls below the high water mark.

Note

Regardless of whether you enable the deletion of files based on the high water mark parameter, if disk usage exceeds the disk allocation that you configure, the CDR repository manager service deletes CDR and CMR files, beginning with the oldest files, until disk utilization falls below the high water mark.

The Cisco Log Partition Monitoring Tool service monitors the disk usage of CDR and CMR flat files that have not been delivered to the CDR repository manager.

Cisco Unified Communications Manager only: If the disk usage of the log partition on a server exceeds the configured limit and the service has deleted all other log and trace files, the log partition monitor service deletes CDR/CMR files on the subsequent nodes that have not been delivered to the CDR repository manager.

For more information about log partition monitoring, see the Cisco Unified Real-Time Monitoring Tool Administration Guide.