Networking / Stage 9 of 14 · step 7 of 8

Internal names do not resolve

~15 min

Not running Starting usually ~20 s

Situation

After the server moved, internal names stopped resolving: getent hosts app.internal prints nothing, and ping app.internal says Temporary failure in name resolution. The *.internal names on this network are served by a DNS server running on this very machine.

Task

The name app.internal resolves: getent hosts app.internal prints the address 10.0.0.10.
The address comes from the DNS server, not from a line in /etc/hosts.

There is no internet on the machine, and everything you need is installed. The app.internal service itself is not running in this lab — only the name is checked.

Lab goals

0 of 2

getent hosts app.internal prints 10.0.0.10 — not checked yet
The address comes from the DNS server, not from /etc/hosts — not checked yet

Hints

0 of 3 open · they open one at a time

Hint 1

Show the solution after 3 failed checks

Network · picked for this lab

DNS and name resolution

Cheat sheet

$_ — insert into the terminal

getent hosts example.com how the system actually resolves it (for apps)
resolvectl status DNS servers and search domains
dig +short example.com a quick answer (IP only)
dig example.com MX a specific record type
dig @8.8.8.8 example.com check via a public DNS server
dig +trace example.com where delegation breaks
resolvectl flush-caches clear the cache
local overrides go in /etc/hosts; permanent ones in netplan's nameservers

/etc/hosts — local overrides (usually take priority over DNS):

127.0.0.1     localhost
192.168.1.10  myserver myserver.local

bash

getent hosts myserver             # check the system sees the override
resolvectl flush-caches           # clear systemd-resolved's cache
resolvectl statistics             # cache hits/misses

dig / host / nslookup

bash

dig example.com                   # a detailed DNS query (package dnsutils/bind9-dnsutils)
dig +short example.com            # answer only (IP) — handy in scripts
dig +short example.com AAAA       # the IPv6 address
dig example.com MX                # a record type (A, AAAA, MX, TXT, NS, SOA, CNAME…)
dig @8.8.8.8 example.com          # ask a SPECIFIC server (bypassing the local one)
dig +trace example.com           # walk delegation from the root (where it breaks)
dig -x 1.2.3.4                    # a reverse lookup (PTR)
dig +noall +answer example.com   # only the answer section
host example.com                  # short and simple
host -t MX example.com           # a record type
nslookup example.com             # the classic (also interactive)

IPs ping, but names don't

DNS isn't working. Work down the ladder: 1. resolvectl status — does the interface have a DNS server (Current DNS Server)? 2. getent hosts example.com — does it resolve through the system at all? 3. dig @8.8.8.8 example.com — does a public DNS server answer (if yes, the problem is local/resolved)?

A name resolves to the wrong IP

An entry in /etc/hosts overrides DNS — check it (getent hosts name).
A stale cache: resolvectl flush-caches.
Multiple A records/round-robin — that's normal (load balancing).

resolv.conf gets overwritten after a reboot

Don't edit /etc/resolv.conf directly (it's a symlink to resolved). Configure DNS in netplan (nameservers:) or /etc/systemd/resolved.conf, then run sudo netplan apply / systemctl restart systemd-resolved.

Internal names don't resolve, external ones do

There's no search domain or the right DNS for the internal zone. Check resolvectl status (Search Domains) and add the domain/server in netplan.

`dig` works, but the application "can't find" the host

dig goes to DNS bypassing NSS. Check with getent hosts/resolvectl query instead — the cause is in /etc/hosts, nsswitch.conf, or resolved's cache.