Forum Discussion

brad9iner_11512's avatar
brad9iner_11512
Icon for Altostratus rankAltostratus
Mar 03, 2017

Pool members not getting flagged down

I have a weird situation happening with my F5(BIG-IP 11.5.4 Build 2.45.291 Engineering Hotfix HF2). I have a pool with 2 members: 

node1:6760
node2:6760
I have a monitor setup on the pool: 
Type: HTTP
Parent: http_head_f5
Send String: HEAD /openid-test/meta/health HTTP/1.0\r\n\r\n
Receive String: .*20*.

A little back story my Send string used to have 'localhost' in front of the /openid-test which was causing 400s in my access logs but the nodes were still green/available in the pool. I removed the 'localhost' and am now getting 200s as expected in the access logs. However I am concerned as to why my nodes were still available even when I was seeing 400s? I even tested by changing up the Receive String to a random number to see if it would put my nodes into a down state but they still remained green. Am I missing something on how the monitor handles flagging nodes up or down?

Thanks, Brad

application delivery
BIG-IP
cloud
LTM

1 Reply

  • Amine_Kadimi's avatar
    Amine_Kadimi
    Icon for MVP rankMVP
    Mar 03, 2017

    The monitor will parse the returned result trying to find the string even if the http result is other than 200 as long as the server responds to the monitor. So your members are green because the received headers contain the receive string.

     

    You can set the receive string to "200 OK" if you need the members to go down upon 404 result.