Details
-
New Feature
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
23.08.11, 24.02.7, 25.01.4
-
None
-
None
-
None
Description
Summary:
Currently customers can access failed login attempts via warnings messages in maxscale.log and need to enable log_info to get see successful client sessions connected. For audit reasons, similar to mariadb audit log, customers want a simplified log access tracking separate from the main maxscale.log which can be too verbose with log_info enabled and would require cleaning up to prepare a report. Mariadb audit logged is specific to a single node, but maxscale stands in front of a cluster with the opportunity to generate a single report for all access. A separate optional flag per monitor would make sense. Starting with failed auth and successful auth, timestamp, and from what originating IP on what listeners connecting to in a comma delimited list .
Sample:
[timestamp],[username],[host],[listener],[service],CONNECT, extra info ?
|
[timestamp],[username],[host],[listener][service],DISCONNECT,
|
[timestamp],[username],[host],[listener][service],FAILED_CONNECT,error returned ?
|