For more information regarding the security incident at F5, the actions we are taking to address it, and our ongoing efforts to protect our customers, click here.

Forum Discussion

John_124726's avatar
John_124726
Icon for Nimbostratus rankNimbostratus
Feb 19, 2014

Using GTM to monitor instance in Google cloud, GCE (Google Cloud Engine)

We have 2 instances of our app running in Google cloud Engine(GCE). The app is accessed using SSL. It is currently accessible through the Internet. We have F5 GTMs at a different locations. Our requirement is to avail the GTMs to provide topology based response to the clients. However when I setup the configuration as below, it keeps the server/wideip/pool keep bouncing up and down.

 

a) Setup a server object as a generic host b) Setup monitor as gateway icmp c) Add virtual server with port 443

 

The server object comes up, stays up for a minute or so and then fails. I see this repeatedly occurring in the log message. Appreciate any help or direction.

 

availability
BIG-IP DNS
cloud
deployment

2 Replies

  • PeteWhite's avatar
    PeteWhite
    Icon for Employee rankEmployee
    Feb 20, 2014

    I would agree that you should setup the server as a generic host but not that you use gateway icmp. Better at the node level to use basic icmp monitor and at the VS level an https monitor specific to the application. Have you tried doing tcpdump to see what is happening?

     

  • Cory_50405's avatar
    Cory_50405
    Icon for Noctilucent rankNoctilucent
    Feb 20, 2014

    If all of your GTMs are doing the health checking against the application, it could be that one or more of your GTMs can't reach it for whatever reason and mark it down. Then another GTM that can reach it marks it right back up, hence the bouncing.

     

    Check /var/log/gtm and look at the reason the health monitor is failing. Take note to which units mark the pool member down, and which mark it up.