The app_log is growing fast

If your [dbo].[sqlwatch_app_log] table is growing fast, there a few things you can try.

Stop logging INFO messages.

Before version 2.4 INFO messages were being logged by default. INFO are informational messages, similar to highly verbose logging. This was changes in version 2.4 and INFO are not logged by default, only WARNING and ERROR.

You can changes this behaviour in the global config:

  UPDATE [dbo].[sqlwatch_config]
  SET [config_value] = 0 -- 1 log INFO, 0 do not log INFO
  WHERE [config_id] = 7

Reduce retention period.

Before version 2.4 the default retention period for the app_log table was 30 days. This changed to 7 days in the 2.4. You can set retention period for the app_log table in the global config table:

  UPDATE [dbo].[sqlwatch_config]
  SET [config_value] = 7 --retention days. 
  WHERE [config_id] = 1

Please note that before the version 2.4 the retention was not batched up, meaning that the delete was executed in one transactions. If your app_log table is large and you are reducing retention significantly in version < 2.4 it may take a while and may also blow the transaction log. If you are on < 2.4 you may want to truncate the table.

Truncate it.

Before version 2.4 the table's retention was not batched up meaning large deletions would be executed in a single transaction. This could take a long time and could also blow the transaction log. It is perfectly safe to truncate the app_log table.

truncate table [dbo].[sqlwatch_app_log]