DNS resolving issue when migrating to a new cloud provider

Hello,

I have recently migrated to a new cloud provider with the opportunity to upgrade to a new version of miab which I successfully did to v0.61.1

During the transition, I used the backup method explained in the documentation and everything was fine.

When changing the server, the location and the IP addresses have changed, I used to have both ipv6 and ipv4 and now I am limited to an ipv4 one only.

The only issue now I am facing is the domain name resolution with the DNS server. It seems that this step is failing even though I am getting no error on the status page. The only error I have that is related to the DNS is The DANE TLSA record for incoming mail ..... is not correct while all others are with no error.

The glue and reverse DNS are correct as well.

when using dnstracer everything is fine as well.

However, when using dig + trace + all domainname I got the following:

;; AUTHORITY SECTION:
domainname.               86400   IN      NS      ns1.box.domainname.
domainname.               86400   IN      NS      ns2.box.domainname.

;; ADDITIONAL SECTION:
ns1.box.domainname.       86400   IN      A       ipaddress
ns2.box.domainname.       86400   IN      A       ipaddress

couldn't get address for 'ns1.box.domainname': failure
couldn't get address for 'ns2.box.domainname': failure

I also turned off the old cloud VM image so there is no DSN conflict coming from that one.

I am not getting a server fail with dig from my own server that is hosting the miab

I am not able to understand how it is possible that the addition section has the IP addresses but then you get couldnĀ“t get address failure at the end message.

Does anyone have an idea what should I do? or how to debug this issue?

Okay, after hours of debugging, it seems that my cloud provider blocked the 53 port number on UDP and TCP which is what DNS uses to propagate