Ignition Network Interface Binding

Ignition
1

Hi All, I have some questions about the network interface binding in Ignition, especially around redundancy.
So, Assume we have 4 subnets on the gateway, all subnets will have clients connects to the ignition gateway. There is a lot legacy reasons for this setup and we are currently stuck with it. Here is what my test results are:

  1. If i use subnet 1 as gateway redundancy network communication, it seems none of the clients on subnet 2-4 would be able to connect to gateway anymore.
  2. If i then turn off auto detect http address, and place all the subnet 1-4 in the address field, the redundancy on subnet 2-4 would not work, it will report redundant connection unknown.
    At the moment, i am using a work around where i don’t specify an ip address in the redundancy interface binding field, but use server name. Then I can use client’s host file to resolve it into the corresponding subnet. Do any one have similar situation and was there a better choice in gateway settings under this situation?
    Thanks
2

IIRC, using the server name with subnet-dependent name resolution is the right answer. It doesn’t have to be handled in a client’s hosts file, though. Non-amateur DNS servers listening on multiple subnets can deliver the correct behavior. (Eg. the --localise-queries option in dnsmasq)

3

Thanks pturmel, I have looked at the windows equivalent of dnsmasq, which seems to be maraDNS, I will see if i can get more resources to put a DNS or a pair of DNS into the environment.

4

My condolences. /: