ASR 5500 supports Assume Positive configurations, and this feature is tailored to provide the service to users even when the Online Charging Server (OCS) is unreachable. This Assume Positive configuration allows the users to configure the interim-quota (either volume or time or both together along with the number of retries) that can be used when the charging servers are unreachable or not responding. This feature also lets the user to configure the action to be taken when the interim-quota and retries are exhausted.
In the existing implementation with Assume Positive feature, there are high chances of losing the usage data reported through the CCR-T when the session is being terminated while in Assume Positive mode. This problem is addressed by allowing the DCCA module to write the CCR-T messages (with locally assigned quota details) in the HDD of the chassis.
In cases where the Assume-Positive interim-quota is allocated, and CCR-T is not reported/answered, the CCR-T message is written to a local file, and saved in the HDD. This local file and directory information can be fetched and parsed to account for the lost bytes/usage. The retrieval of the file can be done with the PULL mechanism.
This feature requires a valid license to be installed prior to configuring this feature. Contact your Cisco account representative for more information on the licensing requirements.
In releases prior to 20, failed CCR-T is written to HDD only if the session is in Assume Positive state. In Release 20 and later, the existing behavior is modified such that, even if the sessions are not in Assume Positive state, the failed CCR-Ts are written to HDD for later processing. This enhancement is applicable for all CCR-T failures like Tx/response timeouts, result code errors, diabase errors, etc.
In case of Session Recovery, if a DCCA session which is in pending-terminate state is recovered, then a fresh CCR-T will be initiated. This CCR-T will be written to hard disk if it fails. In case of ICSR, the sessions which are already in terminating state are not recovered.
Once the bearer/session gets terminated, the same in the standby will be deleted and that session will not come up in case of ICSR.
This feature is controlled through the CLI command "diameter hdd " introduced in the Credit Control Group configuration mode. When the CLI configuration is enabled, the DCCA application sends the failed CCR-T messages to the CDR module for storing in the HDD.
Relationships to Other Features
This feature is applicable for sessions that are in Server-Unreachable state. That is, this feature is applicable only when Assume Positive feature is enabled.
This dependency is no longer valid in Release 20 and later. In Release 20, this feature works even if the sessions are not in Assume Positive state.
This feature requires a valid license to be installed prior to configuring this feature. Contact your Cisco account representative for detailed information on specific licensing requirements. For information on installing and verifying licenses, refer to the Managing License Keys section of the Software Management Operations chapter in the System Administration Guide.
The following are the limitations of this feature:
- When an ICSR event occurs unexpectedly before the CCR-T is written, the CCR-T will not written to the HDD and hence the usage will be lost.
- It is expected that the customers requiring this feature should monitor the HDD and periodically pull and delete the files so that the subsequent records can be buffered.
- It is recommended not to configure PUSH mechanism for the diameter-hdd-module.
- Diameter records will not be written to the HDD when CCR-T is not generated during session termination resulting due to certain error result codes in CCA-I/CCA-U.
- If Diameter records should be dumped to the HDD for all session terminations resulting from failed CCR-Us, then it is recommended to configure the failure-handling template CLI command in the Global Configuration mode. In this case, the CCR-T is generated during session termination for all CCR-U failures.
- T bit is set in the HDD records for CCR-T message failures (response/tx timeout and result code errors).