Forum Discussion

ERLomboy_27803's avatar
ERLomboy_27803
Icon for Nimbostratus rankNimbostratus
Jan 16, 2014

How to setup monitoring on member nodes

Hi guys! Been learning our F5 environment and got a request.

 

Our App team requested to setup the below;

 

Type of VIP (Public or Private): Private

 

Member(s) name and IP address: webserver1_200.201.202.3, webserver2_200.201.202.4

 

Application Port of Member(s): 8086

 

Client Application Port: 80

 

Load Balancing Algorithm (Ex. Round Robin, Least Connections): RoundRobin

 

Health Check: http://webserver1:8086/ResearchDocumentAccess.svc/restful/healthcheck http://webserver2:8086/ResearchDocumentAccess.svc/restful/healthcheck

 

200 is okay, else is not.

 

Can someone advise how will I properly setup monitoring for these two nodes?

 

I've created the nodes and pool, also the private VIP. Just need to setup the health check.

 

I tried to create the monitoring with the ff send string: GET /http://webserver1:8086/ResearchDocumentAccess.svc/restful/healthcheck \r\n

 

with a receive string of 200. The nodes are showing offline (red diamond)

 

Do I need to specify alias service port to 8086? How can I validate the health I created are working as expected?

 

Thanks in advance!

 

3 Replies

  • Looks like your send string is not configured correctly. You may need something along the line of: GET /ResearchDocumentAccess.svc/restful/healthcheck HTTP/1.1\r\nHost: webserver1\r\nConnection: Close\r\n\r\n

     

    Check out the section for creating a custom HTTP monitor: http://support.f5.com/kb/en-us/products/big-ip_ltm/manuals/product/ltm-monitors-reference-11-2-0/2.html

     

    • ERLomboy_27803's avatar
      ERLomboy_27803
      Icon for Nimbostratus rankNimbostratus
      Hi Tim, Thanks for this. What about the alias service port, do I need to change it? the link didn't quite explain this for http, only for https. :)
  • The alias is used if you want the LTM to monitor a service port that is different than what is configured in the pool. The default value should be fine unless you have a need to expand on your monitoring requirements.