A recent ED trip with one of the kids revealed a surprisingly accurate triage wait-time counter. I expected a feel-good dummy number — but nope, it actually worked.
Read On
On my latest visit to the emergency department for treatment of non life threatening injuries my kid sustained I was impressed to see that they had a counter that showed how many people were in the emergency room waiting to be triaged. I was even more impressed that it seemed to be relatively accurate and responsive, I half expected it to be a static number that made everyone feel better about their chances of seeing a nurse before the kids' iPad ran out of battery.
Turns out NSW Health have a website called www.emergencywait.health.nsw.gov.au that reports on wait counts in public hospitals, and the API behind it is laid out in such a way that with Zabbix we can scrape the data for the entire state and report on it.
I present to you, the latest dashboard on my home monitoring server (that gets viewed way more than it should)….
Leveraging Zabbix’ ability to automatically discover hosts (ask the health website for a list of every single hospital) then create new items (ask the health website how many people are waiting at each specific hospital) and create triggers (ask the health website how many beds are available, and alert when waiting is greater than bed count) we can build up a dashboard of the ideal time to seek medical attention, or more realistically which of your local hospitals has the lowest wait time.
If I started tracking the kids with GPS dongles, calculated items would allow me to work out the distance from child to hospital with lowest wait times but they can't even remember to take their library books to school, the chances of them losing a GPS tracker are ironically high.
The flexibility of Zabbix's discovery items reduces the human error elements and can ensure “on demand” hosts, containers, & hardware are automatically monitored when they exist. Automation from these metrics allows other automation to make decision;
Which of my docker heads has the lowest IO wait times and should we shuffle hosts to balance the load.
Which of my network switches has the lowest active port count and should be filled next.
Which hosts under my load balancer have the highest load and is the balancer doing its job correctly.
One of the things I first thought about when HPE announced its plans to acquire Juniper Networks was how will this benefit our customers? We were a long term Juniper Partner with great expertise and a proven track record of supporting our customers across Juniper products and services.
A recent ED trip with one of the kids revealed a surprisingly accurate triage wait-time counter. I expected a feel-good dummy number — but nope, it actually worked.