Dynamic DNS + XMLRPC SYNC
-
All I could find was a 6-year old post about this so I thought I'd ask again since this "bit" me today:
I have a 25.11+ cluster of 1537's with 2 WANs (both static IP) and a DynDNS bound to the Failover GW Group so our VPN server IP floats to WAN2 during outages. The GW Group is set to use the CARP VIP address, not the Interface Addresses.
I noticed that the DynDNS XML is not synced to the backup HA node.
I feel the decision to not sync that config no longer makes sense. I understand there is a possible footgun here if the user forgets to switch the Gateway Group to the VIP instead of the Interface IP (and they end up fighting each other) but a warning or some input validation would fix that.
Can #7292 be reconsidered?
-
The problem you run into there is the secondary could trigger updates when you don't want them to happen.
Making that behave properly would require adding a lot more logic to DDNS updates to know when they're using a CARP VIP (directly or via gateway groups) and to not update unless the node has control of that VIP. Then you'd also practically have to force using a CARP VIP in every possible context of a DDNS entry that's eligible to sync.
While it's technically possible, there are so many pitfalls there it becomes impractical. Trying to sort all that out is a logistical nightmare and has the potential to introduce problems that affect unrelated use cases.
tl;dr: Extremely high effort, extremely high potential for problems, very few requests for the behavior.
Remember: Upvote with the 👍 button for any user/post you find to be helpful, informative, or deserving of recognition!
Need help fast? Netgate Global Support!
Do not Chat/PM for help!
-
What is the recommended method of ensuring high availability of a service running on or behind an HA cluster then? Require running the DynDNS client on a separate system (not the firewall itself?)
