Disk Space Is Filling Up When Purging Old Events

Overview

You are trying to purge unnecessary records to free up some space, but the drive runs out of free space during this process.

This issue happens because EventsManager creates temporary files (which are removed by the commit deletions job) for records that are marked to be deleted, which may take a significant amount of space if your database holds millions of records.

Solution

Remove unnecessary debug logs:
1. Open the following (default) path:
  C:\Program Files (x86)\GFI\EventsManager13\DebugLogs
2. Check how much data is stored in the folder (to know how much free space you can get).
3. You can safely delete files that you don't need to investigate issues with EventsManager (i.e. you can delete all the files if you are not going to do any troubleshooting).
  Note: Every file has a date when it was created.
Check if you have old databases that you can remove:
1. Navigate to Configuration > Options > File Storage > Configure file storage.
2. From the Name drop-down, check if you have other databases available and select one of them, then click OK.
3. Open Events Browser and check how many records are stored in the selected database and how old are they.
  Note: 2,000,000 events take about 1GB of disk space.
4. If you no longer need events stored in this database, manually remove the DB - open the folder with the databases. The default one is:
  C:\Program Files (x86)\GFI\Database Server 2.1\Databases
5. Remove all files and folders that start with the name of the selected database.
6. Perform the same steps for other available databases (step 2.b).
7. Once done, switch back to the main database which was used initially (steps 2.a-b).
If you still run out of space when purging old events and do not have other files that you can delete, move the current database to a drive or server that has more disk space; EventsManager needs to be able to connect to the server (via a direct path) to purge records.
This action can be temporary, and once the unnecessary records are purged, the DB can be moved back to the initial drive.