Zone transfer failure

[Andreas Brandino]

Hello all,

I wonder if anyone can help me to find the cause of the problem I am
currently having.
All servers are running on Debian and BIND 9.10.3-P4-Debian.

I have a master server and 4 slaves.
The zone is transfered from the master [ns1] to all slaves [ns3,ns4,ns5 and
ns6].
I am also using TSIG with a different key for each server.
Moreover, the zone file refers to the internal view.

When I change the myzone.com, I always update the serial and I reload the
zone.

The problem:
ns3 and ns4 never get the updated zone file automatically.
On the other hand, ns4 and ns5 always get the updated zone file immediately.

If I initialize the transfer manually from ns3 and ns4, I get no errors.

Here is the config:

NS1 config: (IP 1.1.1.1 - master DNS)

        zone "myzone.com" {
                type master;
                file    "/etc/bind/master/myzone.com.INSIDE";
                allow-transfer { key ns1ns3_key; key ns1ns4_key; key
ns1ns5_key; key ns1ns6_key; };
                also-notify {
                        3.3.3.3 port 53 key ns1ns3_key;
                        4.4.4.4 port 53 key ns1ns4_key;
                        5.5.5.5 port 53 key ns1ns5_key;
                        6.6.6.6 port 53 key ns1ns6_key;
                };
                notify explicit;
                notify-source 1.1.1.1 ;
                };


NS3 config: (IP 3.3.3.3 - transfer fails)

       zone " myzone .com" {
                file    "/etc/bind/master/myzone.com.INSIDE";
                type slave;
                allow-update { key ns1ns3_key; };
                masters { 1.1.1.1; };
                allow-notify { 1.1.1.1; };
                notify yes;
                request-ixfr no;
                };

NS5 config: (IP 5.5.5.5, successful transfer)

zone "myzone.com" {
                file    "/etc/bind/master/myzone.com.INSIDE";
                type slave;
                allow-update { key ns1ns5_key; };
                masters { 1.1.1.1; };
                notify yes;
                request-ixfr no;
                };

Do you see any errors in the above configuration that could cause this
problem?

Best Regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/bind-users/attachments/20181017/c15b0e12/attachment.html>