CDR Data
Call detail records (CDRs) detail the called number, the number that places the call, the date and time that the call starts, the time that the call connects, and the time that the call ends. Call management records (CMRs), or diagnostic records, detail the jitter, lost packets, the amount of data sent and received during the call, and latency. CDR data comprises CDRs and CMRs collectively. A single call can result in the generation of several CDRs and CMRs. Unified Communications Manager records information regarding each call in CDRs and CMRs. CDRs and CMRs, known collectively as CDR data, serve as the basic information source for CAR.
The Cisco CDR Agent service transfers CDR and CMR files that Unified Communications Manager generates from the local host to the CDR repository node, where the CDR Repository Manager service runs over a SFTP connection. If the SFTP connection fails, the Cisco CDR Agent services continue to make connection attempts to the CDR repository node until a connection is made. The Cisco CDR Agent service sends any accumulated CDR files when the connection to the CDR Repository node resumes. The CDR Repository Manager service maintains the CDR and CMR files, allocates the amount of disk space for use by CMRs and CDRs, sends the files to up to three configured destinations, and tracks the delivery result for each destination. CAR accesses the CDR/CMR files in the directory structure that the CDR Repository Manager service creates.
The high and low water mark settings that you configure specify percentages of the total disk space that are allocated for the CDR repository. Although the preserved folder under the CDR repository folder contributes to the high and low water mark percentages, Log Partition Monitoring never deletes the folder if the high water mark gets reached. If the high water mark gets reached, the CDR Repository Manager deletes processed CDR files until the low water mark is reached or all processed files are deleted, whichever comes first. If all processed CDR files are deleted but the low water mark has not been reached, the deletion stops. The CDRHighWaterMarkExceeded alarm gets generated until the system reaches the maximum disk allocation. If the maximum disk allocation gets reached, the system deletes undelivered files, and files within the preservation duration, starting with the oldest files, until disk utilization falls below the high water mark. If you receive the CDRMaximumDiskSpaceExceeded alarm repeatedly for this scenario, either increase the disk allocation or lower the number of preservation days.
Note |
When the disk allocation usage exceeds the configured high water mark threshold value, LMP also purges the CDR and CMR data that are exported at the following path automatically: /var/log/active/tomcat/logs/car/carreports/reports/ondemand/temp. Customers or any third-party applications should ensure to retrieve the exported files immediately to avoid losing their buffered historical data. |
Information on these alarms is found in the CDR Repository Alarm Catalog (CDRRepAlarmCatalog). The following table displays the alarms and alerts in this catalog.
To configure these alarms, go to
.
Name |
Severity |
Description |
---|---|---|
CDRFileDeliveryFailed |
ERROR_ALARM |
SFTP delivery of CDR files to the outside billing server failed. |
CDRAgentSendFileFailed |
ERROR_ALARM |
The CDR Agent cannot send CDR files from the Cisco Unified CM node to the CDR Repository node within the Unified Communications Manager cluster. |
CDRHWMExceeded |
WARNING_ALARM |
The high water mark (HWM) for CDR files was reached; some successfully delivered CDR files have been deleted. |
CDRMaximumDiskSpaceExceeded |
CRITICAL_ALARM |
The CDR files disk usage exceeded the maximum disk allocation. Some undelivered files may have been deleted to bring disk usage down. |
CDRFileDeliveryFailureContinues |
ERROR_ALARM |
SFTP delivery of CDR files failed on retries. |
CDRAgentSendFileFailureContinues |
ERROR_ALARM |
The CDR Agent cannot send CDR files from the Cisco Unified CM node to the CDR Repository node on retries. |
For additional information on these alarms and recommended action, see the alarm definitions at
.For more information on CDR services and alarms, see the Cisco Unified Serviceability Administration Guide.