What to check on home network

I'm starting to build a very basic home monitoring system for when I am away. Using an Arduino and a WiShield I have it setup to send me an email twice a day when I am away via a PHP page. There will also be an LCD screen.
I will eventually have 3 sensors. Right now I have just 1, a Maxbotix EZ1 Sonar Rangefinder that is setup to take a distance reading. If anything is in front of it then it triggers a reaction because of the distance. I will get 2 more of these soon. I may eventually attach some kind of sensor to some of my windows if I have enough pins left.

What I want to do is a system check, basically a check to make sure everything is operating normally. This would then email me the results of the system check. Beyond just checking the sensors and getting an accurate reading, what else could I check for?
I suppose I could ping a web page and check the response to make sure the WiShield is working properly.
How can you check that an Arduino is functioning properly in this case? (I know getting a correct sensor reading tells me it is working, I'm looking for other ideas)
Any other suggestions?
Thanks.

Two questions for you to think about. If the network does not ping how will you know? Also if the arduino dies how is it going to let you know? I would develop an alarm system (sounds like what you are planning) that is standalone and communicates problems to the arduino that handles the network and communications. Make sure before you start coding you have a complete understanding of what you are monitoring, what the input will be and what you plan to do when you get that input. Draw it out in a flow chart or outline form.

Thanks Franklin. These are good things to think about. I have already made a flowchart.
I'm not sure how I would know if the Arduino dies.

I'm not sure how I would know if the Arduino dies.

Make a PHP page which contacts your arduino and make it a cron job (google 'php cron job' if you need to). If it can't contact your arduino then flag an error and email you.

If your PHP server is not in your house then open the router's admin page to the internet (make sure you've got a decent connection) and get your PHP script to see if the log-in page exists. If it doesn't your router or home internet connection are down.

Finally, you could get an IP webcam, stick that in front of your LCD display. Upload the picture to your PHP server and have it attach the latest image. If you've got the time on your LCD display you'll be able to tell if it's working or stalled.

All good suggestions CowJam.
While I've never run a Cron Job, it seems easy enough. My server is not in my house. So does going from the internet where my server is to the Arduino on my wireless network inside my house mean I have to do port forwarding?

Yes, you will need to port forward for the server to talk to your arduino.

You don't need to do for the arduino to speak to your server.

It's really easy, you tell your router what the public port is (that's what your server will attempt to reach the arduino on - 80 is web), and the private port and IP address (which is the address of your arduino).

If your ISP doesn't use static IP addresses, you'll need to do some jiggery pokery with DynDNS (or similar) as well.

It's all the

jiggery pokery

business I'm afraid of when it comes to port forwarding.

If your router is a decent brand it will likely support a dynamic IP service. Have a nose around, different brands support different services (my dlink is on a dlinkddns.com address which is a bit crap. My last router had cowjam.doesntexist.com but an ex kept it when I left).

If you're port forwarding to the arduino then there's very little danger. It's not like you can jack it unless you've programmed it to.