Forum Discussion

Deon's avatar
Deon
Icon for Nimbostratus rankNimbostratus
Mar 03, 2021

LTM health monitor logging shows success but all members still marked down

I am trying to setup a new pool for our on-prem BigIPs and apps that routes to a cloud hosted service hosted by a 3rd party vendor and I am having difficulty getting the https based health monitor to mark as up the one IP that the FQDN DNS name resolves to. I've tried FQDN Node and FQDN Node List types for the pool members and saw no difference. The IP resolves fine and auto-populates into the pool as needed. I have a basic HTTP 1.0 get request of type https fetching the WSDL document from the external service. I then enabled "Monitor Logging" on the FQDN pool member that I have in the pool that matches in a simple string I know is in the WSDL. Tailing the health monitor log file in /var/logs/monitors/ I can see it send and I can see the proper wsdl contents and a 200 response code and then it also says '...(_recv_active_service_ping): Response matched regex...' yet the member is still being marked as down. I did a 'tmsh show /pool <poolname>' from the command line and it is showing 1 member but zero available. I am at a loss as to what to try next to get this to work.

2 Replies

  • What version BigIP?

     

    Do you have node monitors or other monitors assigned to the pool (requiring all monitors to be marked up?

     

    I'd probably recommend opening a Support Ticket so you can provide logs and config for review.

  • Deon's avatar
    Deon
    Icon for Nimbostratus rankNimbostratus

    BigIP 14.1.2.6 here. Yeah I had started thinking along the same lines that this might be related to the node manager and the auto node being down. Default node monitor is icmp. I tried changing the FQDN pool to monitor=none but it remains marked as down. Will keep trying for a bit and then open a support ticket if still no success.