ingress TCP connection state

[Miloš Kozák]

Hello,

I have got haproxy installed from CentOS 6 repository, and I use TCP mode for a 
special messaging protocol. Occasionally, the ingress TCP connection dies, but 
the TCP connection towards the application server survives. There is a not very 
short time period when application server thinks that the TCP connection is 
alive and sends out messages towards the mobile client. However, those messages 
get never delivered! My configuration is very simple:

defaults
        log     global
        mode    http
        option  httplog
        option  dontlognull
        retries 3
        redispatch
        maxconn 2000
        contimeout      5000
        clitimeout      50000
        srvtimeout      50000 

listen  relay
        bind 0.0.0.0:2268
        mode tcp
        server s1 10.85.45.39:2268

Is there a way how to act based on ingress TCP connection state? I am 
considering to use keepalive in order to detect the connection loss faster. Do 
you think this is a viable approach?

Thanks for any help,
Miloš

smime.p7s
Description: S/MIME cryptographic signature