Maintenance and Monitoring

It is necessary from time to time to purge old log files and check the system for errors. It is also recommended to monitor Data Source Solutions DSS for runtime errors. All of that can be done via alerts and log rotation.

Alerts

DSS can send alerts about errors and warnings that happen during data movement and replication. Alerts can be created and managed via both UI and CLI.

In the DSS User Interface, the Alerts tab on the System page displays the list of all alerts configured on a DSS hub. Here you can create new alerts and manage the existing ones.

In the Command Line Interface, DSS alerts can be created and managed using the dssalertconfig command.

The command dssalertmanager is responsible for executing all the alerts configured on the hub server. For more information about scheduling alerts in DSS, see Scheduling Alerts.

Log Rotation

Log rotation is the process of archiving old log files. This is achieved in DSS through the dssmaintlogrotate command. It moves the log files from the hub's log directory DSS_CONFIG/hubs/hub/logs/ to the archive directory DSS_CONFIG/hubs/hub/logarchives/.

Since DSS 6.1.0/3, log file rotation is an automated process.

Automated Log Rotation

Log rotation is a scheduled task that is seamlessly managed by the DSS hub server. This scheduled task calls dssmaintlogrotate, which rotates (archives and reopens) log files at regular intervals. The default interval is 24 hours and can be configured using the hub server property Log_Rotate_Interval. The log files subject to rotation include both the hub server log files (archived to DSS_CONFIG/logarchives/) and the log files specific to individual hubs (archived to DSS_CONFIG/hubs/hub/logarchives/).

The name of each archive directory is a UTC timestamp. The contents of each archive directory is a number of archived files with .gz suffixes.

Automated log rotation occurs for LIVE hubs only.