Backtrace for SQL errors and other error monitoring

[KT Walrus]

I'd like to see the current call backtrace shown in the new SQL error log to increase the chance of identifying exactly what the cause of the problem is.

Also, would be nice to include a simple bash or php script that monitors the various error log files for a change and sends an alert email to an address (presumably to the admin). Such a script could be run as a cron job and coded to send out notification emails only every X hours for new changes.

This sample script could monitor other aspects of a boards health such as high CPU load or big total execution time for a board page load (from the PHP server's point of view - so DB queries that all of a sudden get slow can be identified).

Indeed, you could include both a local server monitor sample script (written in PHP to execute at intervals via cron) and a remote server monitor sample script that maybe checks a special new board URL for a status update that would return any slow or abnormal board execution that it can detect and sends an alert if something is out of the ordinary. If you wrote them in PHP, they could probably execute on a wide variety of computers (even my home Mac to monitor my board).

The remote server monitor could report when the website did not respond possibly indicating a problem with the site or the remote server's internet connection.