Retry Mechanism
The CxArchive process consists of the following stages:
Fetch data from the CxSAST database.
Store data in the Main Elastic index.
Register the scan in the Management Elastic index. This only occurs when scans are successfully exported to Elastic.
Export source files. The source files are not exported by default, but by configuration only.
The Retry Mechanism involves the following three failure modes:
If the process fails between stages 1 and 2, no data is stored in Elastic.
If the process fails between stages 2 and 3, partial data will be stored in the Main index, but not in the Management index. The way to identify that a scan is partially in Elastic is by searching for results that are in the Main index, but are not in the Management index.
In the above two failure modes, the Retry Mechanism process attempts to re-process the scan X number of times. If all these attempts fail, the recorded information is removed from both the Main and the Management indexes, and the scan is marked as Failed and is removed from the queue. The number of retries is configured under ExportingSettings → Number of Retries in the appsettings.json file. For more information, see AppSettings customization.
If the process fails on the last stage due to a critical issue, such as a power failure, the system will not be able to mark the scans as Failed and the scans will remain in the queue. Once the service is running again, it will resume the scan export process.