Default HTTP monitor v10.2

 

1. What is the health monitor doing when you leave the send string empty ? Is this simply a socket based check ? Which would give us not much more than a standard TCP monitor?

 

 

I've never tested it, but I don't see a reason to do this. I don't think an HTTP monitor could succeed without a send string. I assume bigd would establish a TCP connection and send no data.

 

 

 

2. What is the check for a success status when using the default “out-of-the-box” http health monitor ? (get /\r\n)Does it expect a HTTP 2xx return code?

 

 

 

If there isn't a receive string configured any data the server responds with will be considered a success. Again, I don't suggest doing this--you should always configure a receive string for an HTTP monitor.

 

 

SOL2167 is a good place to start for building HTTP/S monitors:

 

 

sol2167: Constructing HTTP requests for use with the HTTP or HTTPS application health monitor

 

http://support.f5.com/kb/en-us/solutions/public/2000/100/sol2167.html

 

 

Aaron

If there is no SEND string then the default monitor will just send a 'GET /' . The server is marked up if ANY content is received (it will not check the details of that content).

 

Notice this must be HTTP content. It is possible to receive just an HTTP Header response, but no actual content (Header: content-length: 0) and this will result in the monitor marking the pool/node as down.