I have a script thats does basic testing of whether a server is up or down by using the test-connection cmdlet. Recently two servers, both behind firewalls, are now randomly failing - the servers are not having any problems (not rebooting, etc.) but yet my script fails for them 3-5 times per day randomly. The network and FW guys see nothing going on (no flapping, etc.)
I am wondering if my script is too sensitive. However, I do already have a 30 retry delay and it still fails - if it detects a failure it sleeps for 30 seconds and tries again before firing off an email. But even with that it still fails.
Anyways, I am wondering how the following parameters affect test-connection...do they make is more sensitive, less sensitive?
For example if I set my TTL to 255, what exactly does that do in terms of how it affects the cmdlet? Does it make is less sensitive to minor network hiccups if I make that parameter higher, or make it lower, or does it not have any affect? Obviously something is causing minor failures to occur for these two server and I am just looking for a way to make test-connection more tolerant to the hiccups.