Strictly System Check is a site monitoring tool that allows website administrators who may not have access to professional monitoring tools
the ability to regularly check the status of their site and to be notified if the site goes down or becomes overloaded.
I created this plugin initially for my own use as on one of my sites I regularly import content from XML feeds and I noticed that occasionally
after a large update my site would suffer problems such as:
* Error establishing a database connection, error appearing when I tried to access the site even though the configuration was correct.
* All my articles and other content disappearing from the site.
* A high server load that didn’t reduce quick enough after the import had finished
I soon realised that this was down to one or more of the MyISAM database tables in the MySQL database becoming corrupted which meant that the tables
were out of action, data couldn’t be retrieved and requests to the system were quickly building up. Running a REPAIR statement on the database
always seemed to fix the problem. Therefore as I wasn’t always able to catch this problem when it occurred I thought I would automate a process that would
check the homepage at regular intervals and on finding the database connection error it would then check the database for corrupt tables and
automatically repair them.
This was the primary reason for the plugin and I soon extended it to offer some more features such as
- The option to choose which URL is scanned when the system check is carried out
- The ability to search for a key phrase or piece of text in the source code and to raise a report if it’s not found
- The system also checks the server load and can raise a report if it’s over a specified limit
- The system also checks the database load to ensure that it’s not overloaded and not running too many queries
- The system is not suffering poor performance due to fragmented table indexes
- Reports on the number of connections, queries, reads, writes, aborted connections, slow queries
The site administrator can specify their own threshold limits for the webserver and database loads and a report will be emailed out
whenever these limits are reached or if the site is inaccessible.
The report will return details of the response time it took to load the page, the current server load, the current database load and whether or not
there were connection problems, issues with corrupt tables that needed repairing or fragmented indexes that were optimized.
Whilst not a replacement for professional server monitoring tools it is a nice easy to use plugin that can help notify you when your site is down
as well as rescuing your system from corrupt database tables before you even realise there has been an issue.
- Added extra check for system functions in the configuration test
- Added the option to carry out a test for fragmented tables and then a repair using the OPTIMIZE command
- Added new database reports such as the percentage of queries that are slow, the number of joins without indexes, the number of reads & writes and much more
- Changed the formatting and layout of the admin page
- Added an extra test for MySQL version in the test configuration option
- Added code to inform the user if the plugin gets updated with new options so that they know to re-configure their plugin
- Added nonces and is_admin checks to the admin page
- Added code to handle REPAIRS if the database is overloaded
- Added extra logging for the REPAIRS if they are carried out OR not carried out
- Tested with latest version of WordPress 3.5.2
- Added code to only OPTIMIZE if the server load is above a certain level
- Added code to run DB checks if the page load is too slow
- Ensured the system works with WordPress 3.6
- Updated Readme.txt with information on how to debug any problems
- Added code to allow users to choose an external WP Cron option to run their jobs from
- Added code so if page response is blank its same as a critical error
- Added some options to the error email for people to try if their site is down
- Changed the check config to check for empty responses and 500 status codes
- Added code to show the PHP memory usage at the time of the report running
- Added code to show the output of the “free” command and the free Apache memory
- Added extra fixes that could be done
- Informed users about the great mix between Strictly AutoTags Premium and Strictly TweetBOT PRO
- Added code to fix free memory usage
- Added code to show total memory available
- Added code to show total disk swap memory available
- Added code to show the disk swap usage
- Added extra fix options to increase RAM, reduce disk swapping, reduce bandwidth, reduce load.
- Added code to check for reverse proxy error codes like CloudFlare messages
- Added code to return the current user the system is using
- Added code to ensure that if the server is very quiet but long page loads a warning is sent
- Added warnings about disk swap usage