Help with "Could not resolve hostname" errors but DNS is fine

[lnlyssg]

I've just started getting a ton of messages like "Could not resolve hostname news.eweka.nl: Error -3 - Try again" for every server I have defined but I can't see why - I'm running from Docker and I can ping the addresses fine from the container e.g.

I have also run a test connection OK from the servers page so I'm struggling to work out what the issue could be.

[luckedea]

@lnlyssg Can you clarify:

1. Does this happen only on eweka or on other news servers too? How consistent is this?
2. What's the connection limit for that news server configuration?
3. Can you give us all the environment versions - docker, docker-compose, host system, nzbget version?

[lnlyssg]

All my other containers, and the Synology itself, were all working fine so I was reluctant to reboot completely but if it happens again I will do so.

[noodlers]

I'm experiencing the same issue, except with the binary install on a Windows 10 system (version 23.0).

This happens for all configured servers. All configured servers "test" successful, however Messages tab filled with continuous "Could not resolve hostname: Error 11001 - No such host is known" error messages.

Servers can be pinged from host system, no DNS issues on host system with other applications.

No firewall changes have occurred.

Tried:

• Reloading from web interface
• Stopping and restarting from web interface
• Restarting host system multiple times

I'm not entirely sure what might have changed to cause this, but it has been ongoing for over a week without "fixing itself".

[luckedea]

Do you have any software that may impact this setup? VPN with SplitTunneling, AV software or some custom DNS tools (SmartDNS? SecureDNS?)

[noodlers]

No VPN software, or custom DNS tools that might influence the communication. Windows Defender is the only AV tool, and it has been tested both on and off with no change.

After more investigation I noticed all my servers were set to IPv6 in the configuration. Changing these to IPv4 immediately worked. I don't know why this is--the servers can all be pinged via IPv6 and I've confirmed IPv6 connectivity via other IPv6-only services. Also confirmed IPv6 DNS resolves.

[luckedea]

This is a good find which can explain many things. I'll do more tests about it, @phnzb please assist as well.

[paralovosca]

Same issue noticed today as well. Using nslookup within docker exec -ti nzbget /bin/sh works fine. Yet, nzbget shows [ERROR] Could not resolve hostname for every server configured.

[paralovosca]

It's an intermittent issue but not sure what causes it. :(