We use cookies to understand how the site is used and to improve it. You can accept them, or carry on with them switched off.
Also known as: DNS failure, name resolution error
The client cannot translate the provider's hostname into an address, so no connection is even attempted. Causes include a typo in the server name, a resolver outage at the access provider, and DNS filtering on the local network. Providers publish round-robin records covering several machines, so a partial failure can also hand back an address that is out of service. Testing against another resolver confirms it.
Everything downstream of resolution is untested when this happens. No packet reached the provider, no port was tried, no certificate was seen, so nothing in the server settings beyond the hostname itself has been exercised. A reader troubleshooting from this error should resist changing ports, toggling SSL or re-entering credentials, because none of them are implicated and none of them can help.
These two are constantly confused. A refusal requires the hostname to have resolved successfully; a resolution failure means it never did. The log wording separates them: a message naming the hostname, or containing phrasing about a host not being found or a name not being resolved, is DNS, while a message naming an address and a port number is a connection failure. Timing is less reliable here, since a resolver can answer instantly from cache or hang for several seconds waiting on something unresponsive.
Responsibility splits roughly three ways. A typo in the server name belongs to the reader and is worth ruling out first because it costs nothing. A resolver outage belongs to the access provider, not the news provider, and typically breaks every site on the connection at once, which makes it easy to confirm in a browser. Filtering is the third: some networks, routers with content controls, and some public resolvers block news-related hostnames outright, returning no answer or an address that leads nowhere. The news provider is at fault only in the uncommon case of a lapsed domain or a broken zone, which would affect every customer simultaneously.
Providers publish several addresses under one hostname and hand them out in rotation. A client can therefore receive the address of a machine that is out of service while another client resolving the same name receives a healthy one. The symptom is intermittent: connections succeed and fail alternately with no configuration change between attempts. Restarting the client, or waiting for the cached record to expire, usually lands on a working address. Hard-coding a single address to sidestep this trades the problem for a certificate hostname mismatch and the loss of automatic failover, so it is not a sound permanent arrangement.
Browse all terms in Errors & Troubleshooting, or see the full Usenet glossary.