As ShakeCast continues to operate it downloads and processes ShakeMaps published for events. These events take up space in the filesystem as well as in the database. Part of a regular server maintenance routine is to purge past events which will free up disk space as well as database table space. One method has been described to manually determine which events can be removed and how to remove them. Following these instructions this task can now be automated.
- Download the purge_events.pl script by clicking here.
- Place the purge_events.pl script into the ShakeCast\sc\bin directory
- Determine the frequency of the task to run (every week - recommended or every month) and the data retention timeframe(days worth of event data to keep, e.g. bi-monthly - approx. 62 days, quarterly - approx. 93 days, semi-anualy - approx. 186 days )
- Schedule the script with the operating system's task running subsystem(Windows Scheduled Task, Linux/Unix Cron facility)
- The Task and its User must have permission to read files in the ShakeCast\sc\lib, ShakeCast\sc\conf, and ShakeCast\sc\bin directories and the ability to modify and remove files and directories in the ShakeCast\sc\data directory.