I am trying to figure out exactly how the 'Connection History: Percentage of refused connections' monitor obtains it's values for Aborted_connects and Connections to determine why we have been receiving alerts. I have looked through the documentation and forums, but am unable to locate any explanation for this.
My understanding is the global status variables Aborted_connects and Connections are cummulative from the time the server is started. I also checked to make sure there haven't been any FLUSH STATUS commands issued since server startup. The current values are: Aborted_connects=95, Connections=12010210, and Uptime=18846734. Yet, over the past 36 hours we have received 4 critical alerts saying this monitor is at anywhere between 33.33% and 100.00%.
Can someone please describe how this monitor determines the current values for these status variables, or better yet, why the alert is being triggered when the status variables on the server do not seem to reflect the value being returned by the monitor?
Thank you,
Steven Libby