Possible optimization of the load balancer how-to with Hetzner

[Nexulo]

I have seen at https://coolify.io/docs/knowledge-base/how-to/hetzner-loadbalancing that the use of a load balancer is explained with Hetzner. Regardless of these instructions, I already have a similar setup. In the past few days, however, there has somehow been a problem at Hetzner:

(Support message from Hetzner, translated from german to english)
We have identified a bug that caused some load balancer targets to be unreachable over cloud networks (this also affects vSwitch-coupled dedicated servers).

This only affected routing setups where the target service was unaware of the 1450 byte MTU of the private network interface and when sending large network packets. Only very few load balancers were affected by this problem.

Now the question arises as whether the MTU should be set to 1450-byte MTU for Hetzner servers. I am not an expert in networking, MTU etc., but is it possible that this should be adjusted somewhere so that the load balancer does not fail if this problem occurs again with Hetzner? As I understand the message from Hetzner Support, my load balancer would not have failed if this MTU setting had been correct.

It's more about optimizing the How-To and also about my understanding to not have this problem again in the future.
I am aware that the problem was apparently mainly by Hetzner, but it would be good if the load balancer would not fail again if this problem reoccurs and the MTU is correctly adjusted.