Overview
Wikipedia defines Uptime as a measure of the time a computer system has been “up” and running, but did you know there are different kinds of uptime? Did you know that many web hosts use this common lack of knowledge against you?
Types of Uptime
- Network lifetime: This is the type of uptime many hosts use to incorporate loopholes in to their advertisements and guarantees. Based on this measurement, a network can have 9 hours of downtime each year and still fulfill an advertised 99.9% guarantee! How would you feel if your site was down for 9 hours every year?!
- Set time frame: Most commonly measured annually, monthly, weekly or daily, this is the method we use in our uptime guarantee to assure a set, recorded number each month.
- Rolling period of time: This type of measurement is counted from the last X amount of days, usually the last 30, and is a constant measurement – never ending.
A live-updating uptime graph for our beloved "lambda" server
We use Pingdom
PhireFast uses and recommends a great, paid service called Pingdom. Their global network of monitoring servers measures and reports on each of our hosting servers. If you are interested in a constantly-updated uptime graph for the server your site is on, let us know!
When shopping for hosting
Find out if the host clearly defines the type of uptime their guarantees are based on. Do they back-up their advertised features in their Service Level Agreement? We do.
For the “techies”: Finding uptime
Windows
How: From command prompt, run systeminfo | find “Time:”
Example output: “System Up Time: 8 Days, 8 Hours, 30 Minutes, 17 Seconds
Unix/Apple OS
How: From a terminal session, run uptime
Example output (taken from PhireFast’s very own “lambda” server): 22:51:47 up 34 days, 6:39, 1 user, load average: 4.51, 4.04, 4.21
