Hostnames getting resolved to internal ip inside network

I have a very weird problem.

My local cloud provider is migrationg to a new location and we are setting up our infrastructure there. They use openStack.

After I installed 2 instances one with cpanel and the other with dnsonly, I notice that when i resolve the hostname of the other instance, the hostname get resolved to the internal ip address, but i never used the internal ip address anywhere. Somehow, the DNS response seems to get changed by something.

Lets call then server1 and server2. When i'm in server1, and try to dig server2, i get the following response:

----
[root@server1 ~]# dig +noedns server2 @8.8.8.8

; <<>> DiG 9.11.36-RedHat-9.11.36-14.el8_10 <<>> +noedns server2 @8.8.8.8
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 38557
;; flags: qr rd ad; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0
;; WARNING: recursion requested but not available

;; QUESTION SECTION:
;server2. IN A

;; ANSWER SECTION:
server2. 3600 IN A 10.0.0.233

;; Query time: 1 msec
;; SERVER: 8.8.8.8#53(8.8.8.8)
;; WHEN: Wed Jun 26 08:30:35 -03 2024
;; MSG SIZE rcvd: 72
----

When i ask the same question in a server outside their cloud, i get the correct response:

----
[root@localhost ~]# dig server2 @8.8.8.8

; <<>> DiG 9.11.36-RedHat-9.11.36-14.el8_10 <<>> server2 @8.8.8.8
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 52205
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 512
;; QUESTION SECTION:
;server2. IN A

;; ANSWER SECTION:
server2. 600 IN A 201.xxx.xxx.xxx

;; Query time: 146 msec
;; SERVER: 8.8.8.8#53(8.8.8.8)
;; WHEN: Wed Jun 26 07:25:46 EDT 2024
;; MSG SIZE rcvd: 64
----

Some thing i noticed:
- The first query only takes 1ms, so it must be getting resolved locally, however the hostname in question is not on /etc/hosts
- Not only the ip address, but the TTL is also changed from 600 to 3600.
- If i dont use the +noedns flag, i get a gibberish response that i never saw before.
- The same issue happens if i search server1 from server2..

I already tried my local cloud provider support, but they dont recognize the problem.

What could be the culprit here?

Thanks!