Your configuration files seem okay. New version 0.2.6 doesn't modify any of zone ID fetching code, so I'm not sure why it breaks after upgrading from 0.2.5.

I would suggest you pin the version to 0.2.5 for now, and let's see if anyone else encounters the same issue.

I have just tested a container on another Docker host (same Version and System-Config) with version DDNS 0.2.6 using the current configuration including IPv4 and IPv6 parameters. There is no 'fetch' error there.

Unfortunately, this does not give me a new idea where the cause might lie.

Edit:
An obvious difference is that the production host has significantly more network interfaces, while the test system has only a few. Could it be that version 0.2.6 is possibly addressing the wrong interface and therefore causing the Fetch ZoneID error?

Edit: 2
Gives a debug log that I can enable?

Gives a debug log that I can enable?

There is no option for additional debugging information, as this is a fairly simple utility. However, you might try manually running curl commands this script uses inside and outside the container to see what causes the issue.

Thank you very much for your help. I executed the command / # curl "https://dns.hetzner.com/api/v1/zones" -H "Auth-API-Token: WjcswEAFaASdsdasdasd" on the host and in the Docker container.

1. On the Docker host, all information could be retrieved (2450 characters).
2. In a Docker container with version 0.2.5, all information could be retrieved (2450 characters).
3. In a Docker container with version 0.2.6, I receive the error curl: (6) Could not resolve host: dns.hetzner.com

An nslookup within a container running version 0.2.6 works:

/ # nslookup dns.hetzner.com

Non-authoritative answer:
Name:   dns.hetzner.com
Address: 159.69.255.131

Non-authoritative answer:
Name:   dns.hetzner.com
Address: 2a01:4f8:c010:c083::1

If the name can be resolved in the same container, then it should also be resolvable by the script :-/

When I specify the DNS parameters as follows, fetching the ZoneIDs works during the runtime of the script with version 0.2.6.

The script seems not to be querying the correct DNS servers during runtime.

# xyz.de DDNS  
  xyz:
    image: filiparag/hetzner_ddns:0.2.6
    container_name: 'hetznerddns-xyz'
    dns:
     - 8.8.8.8
     - 1.1.1.1
    environment:
      - TZ=Europe/Berlin
    restart: unless-stopped
    volumes:
      - /var/lib/docker/volumes/hetznerddns-xyz-prod/_data/hetzner_ddns.conf:/etc/hetzner_ddns.conf:ro #ReadOnly
    network_mode: host

Just tested with straight docker_compose.yml file from project, working fine here.
@Crash1602 you may have a Docker problem.

you may have a Docker problem.

@BallonLac That is possible, and I do not want to rule it out. Do you have any idea where I could start with the troubleshooting? At the moment, I have none, and I don't understand the difference between the working nslookup and the curl command during runtime. I would have assumed that the same DNS server would be accessed during runtime :-/

Thanks for your help.

@Crash1602 you may dig about "network_mode: host" option: https://docs.docker.com/reference/compose-file/services/#network_mode.
What gives commenting it out?

#network_mode. What gives commenting it out?

@BallonLac That was a good idea. I test it and commented out the host mode, assigning my custom bridge network instead. The retrieval works, and no error is generated. I just checked the Docker repository, and it seems that the bridge mode is also the recommended option, which avoids any further issues.