Forum Discussion

Moinul_Rony's avatar
Moinul_Rony
Icon for Altostratus rankAltostratus
Mar 17, 2014

Having trouble creating a basic HTTP monitor

Hi,

 

I am trying to set up a HTTP monitor for a HTTP page. I keep getting the node has been marked down by a monitor.

 

GET /some_page/page.html HTTP/1.0 \r\n

 

I can hit the site and can run the same GET command from LTM shell and get page returned. Is there anyway to find out where it is failing ?

 

5 Replies

  • Also tested with the following command [admin@LTM-01:Active:Changes Pending] ~ echo -ne "GET / HTTP/1.1\r\nConnection: Close\r\nHost: \r\n\r\n" | nc 192.168.149.210 80 | grep "Matching strings" And I can see the matching strings
  • Also is there any way of doing a dry run of a monitor so that it does not mark down the node.
  • Hi Moinul!

     

    Would it be possible to post the strings you want to use? There are some "protected" strings that you should avoid.

     

    You can make a "dry run" by adding a monitor you know will work and setting the Availability requirement to at least 1 monitor. Then it won't matter if one fails.

     

    /Patrik

     

  • Two things:

    1. I'd add an additional \r\n to the end of the request string

      GET /some_page/page.html HTTP/1.0\r\n\r\n        
      
    2. Assuming the backend traffic isn't encrypted, you could do a tcpdump capture and look at the payload with the -Xs0 command

      tcpdump -lnni 0.0 -Xs0 host x.x.x.x
      
  • Can you rep your command line test using http 1.0 instead of 1.1? Your monitor is 1.0 but your get test in your question is 1.1. Some Web servers don't respond well with 1.0 anymore.