Hi Rick,
On occasion, some request ends up hanging all current COM instances of my app. When this happens, my active customers are usually pretty quick to let me know something is wrong. I hop on to the handler status page, verify it's not healthy, flip over to file mode, unload all instances and then flip back to COM mode. I'm wondering if there's a better way for me to know something is unhappy and proactively notify me of the problem as opposed to my clients.
TIA
Great question, yes... Rick created a tool for doing this many years ago. I have used it to monitor my servers and found it to be an invaluable resource. The product is called Web Monitor, here is the link: WebMonitor
It has the ability to send you a message when a site does not respond. It can also be used to run a script to shut down a stuck server and then restart it. You can set time intervals of when to check the status of the site and a lot more.
I thin this is an awesome solution and suggest you try it for yourself.
Best, Harvey
Thanks for chiming in, Harvey. I thought I had maxed put my West Wind toolkit... Will check into that.
As Jim mentioned, Web Monitor is a good tool for this and I use this myself to let me know of problems with my servers. When servers go down it'll send email and text messages (I use email aliases) and it can let you know when servers go down and come back up.
And - one of the more useful features and one of the main reason I built this for my self so many years ago: You can have it fire an executable or batch file when a server fails. I typically restart the application pool or completely do an IIS Reset when this happens.
It's a really old application, and it runs as a Windows Service with a UI front end, but it simple and it works.
+++ Rick ---