How can I check which DNS server am I using (in Linux)? I am using network manager and a wired connection to my university's LAN. (I am trying to find out why my domain doesn't get resolved)
You should be able to get some reasonable information in:
$ cat /etc/resolv.conf
-
34However, please be aware that (on modern Linuxen) the contents of
/etc/nsswitch.conf
dictate what name services are used (DNS, LDAP, etc) and in what order. Sayfgrep hosts: /etc/nsswitch.conf
. If it only references DNS,/etc/resolv.conf
is the right place to look for your nameservers. But chances are you're also using mDNS (aka ZeroConf, aka Avahi, aka Bonjour, etc), etc. In that case, things depend on what you're using. – Alexios Jan 12 '12 at 13:35 -
48This file typically points at 127.0.1.1 on Ubuntu - it's the local DNS cache server, not the actual upstream. – Barry Kelly Mar 8 '16 at 10:24
-
3
-
4And if you have several upstream server configured ? How to know which one is currently used ? – Sylvain Leroux Nov 24 '16 at 23:31
-
6See the answers by @G32RW or @Lonniebiz for a more robust approach under various circumstances, e.g. when you get an answer like
127.0.0.53
– nealmcb Nov 11 '18 at 22:13
Here's how I do it:
( nmcli dev list || nmcli dev show ) 2>/dev/null | grep DNS
This worked previous to the way above:
nm-tool | grep DNS
On Debian, you need to have the network-manager package installed.
-
17This one is usefull if you are using VPN and NetworkManager. Your
/etc/resolv.conf
will point to your machine, withdnsmasq
resolving names as configured by NetworkManager. – Grzegorz Żur May 30 '13 at 11:32 -
7
-
2nm-tool is not available in newer linuxes. for example it is not in the 'network-manager' package of debian 8. – don bright Oct 31 '15 at 15:06
-
2
-
3
On systems running systemd use:
systemd-resolve --status
-
8
-
2Says
Failed to get global data: Unit dbus-org.freedesktop.resolve1.service not found.
– xji Apr 12 '18 at 11:19 -
14This is the new default way to do it in Ubuntu 18.04 Bionic Beaver - get used to it, everybody! – AveryFreeman Apr 24 '18 at 20:32
-
7This is the only solution that worked for me, as the others returned 127.0.0.53 – greuze Oct 25 '18 at 7:50
-
1
I think you can also query DNS and it will show you what server returned the result. Try this:
dig yourserver.somedomain.xyz
And the response should tell you what server(s) returned the result. The output you're interested in will look something like this:
;; Query time: 91 msec
;; SERVER: 172.xxx.xxx.xxx#53(172.xxx.xxx.xxx)
;; WHEN: Tue Apr 02 09:03:41 EDT 2019
;; MSG SIZE rcvd: 207
You can also tell dig
to query a specific DNS server by using dig @server_ip
Just do an, nslookup
. Part of its results include the server that it's using.
In the example below, it shows that the DNS server used is at 8.8.8.8.
$ nslookup google.com
Server: 8.8.8.8
Address: 8.8.8.8#53
Non-authoritative answer:
Name: google.com
Address: 172.217.22.174
-
1
-
-
8
-
9On a recent Ubuntu, this again points to the local cache server 127.0.0.1 as already hinted at in this comment – FriendFX Dec 1 '17 at 1:50
-
In CentOS 7 it quits with error, but it is a vm so I did
nslookup google.com
in the Windows host and I found the nameserver. Add it in/etc/resolv.conf
like:nameserver xx.xx.xx.xx
and restart servicenetwork
, and all is fine. Praise you. – WesternGun May 3 '18 at 15:22
With the new network-manager
command nmcli
, do this:
nmcli --fields ipv4.dns,ipv6.dns con show [connection_name]
On newer versions of network-manager (such as in Ubuntu 16.04), the field names are slightly different:
nmcli --fields ip4.dns,ip6.dns con show [connection_name]
If you don't know the connection name, use:
nmcli -t --fields NAME con show --active
For example, on old versions of nmcli :
$ nmcli --fields ip4.dns,ip6.dns con show 'Wired connection 1'
IP4.DNS[1]: 172.21.0.13
IP4.DNS[2]: 172.21.0.4
-
-
It works fine for me with network-manager 1.0.4 on Ubuntu 15.10. Maybe you have an older version? – Sameer Mar 17 '16 at 5:40
-
The tabular format is pretty bad. I hope to get a column like format similar to Powershell. – CMCDragonkai Mar 14 '17 at 6:42
-
2Returns
Error: invalid field 'ip4.dns'; allowed fields: NAME,UUID,TYPE,TIMESTAMP,TIMESTAMP-REAL,AUTOCONNECT,AUTOCONNECT-PRIORITY,READONLY,DBUS-PATH,ACTIVE,DEVICE,STATE,ACTIVE-PATH.
– FriendFX Dec 1 '17 at 1:53 -
to get the first DNS SERVER (IP only) :
cat /etc/resolv.conf |grep -i '^nameserver'|head -n1|cut -d ' ' -f2
cat
will output DNS configgrep
filters only nameserverhead
will keep only the first row/instancecut
take the ip part of the row (second column with ' ' as separator)
To put DNS ip in an environment variable, you could use as follow:
export THEDNSSERVER=$(cat /etc/resolv.conf |grep -i '^nameserver'|head -n1|cut -d ' ' -f2)
-
3
-
To lighten the pipeline even more, capture groups with Perl regexp is very neat, and grep takes a file argument:
grep -Pom 1 '^nameserver \K\S+' /etc/resolv.conf
. Just wrote up Capture groups with grep perl regular expression – sshow Jul 28 '17 at 9:04 -
There's both IPv4 and IPv6 DNS servers. Also, secondary servers exist for a reason. Which one this command returns? Is it advised to take the first one, and just ignore the others? – André Werlang Jan 16 '20 at 2:53
If you are using network manager probably you get all network parameters from your dhcp server at your university.
If you don't want use your shell to check your dns settings (as described by hesse and Alexios), you can see them from the panel "Network information".
You can reach this panel by pressing right mouse button on network manager icon and selecting "Connection Information" from the menu.
Using resolvectl
$ resolvectl status | grep -1 'DNS Server'
DNSSEC supported: no
Current DNS Server: 1.1.1.1
DNS Servers: 1.1.1.1
1.0.0.1
For compatibility, systemd-resolve
is a symbolic link to resolvectl
on many distros as for Ubuntu 18.10:
$ type -a systemd-resolve
systemd-resolve is /usr/bin/systemd-resolve
$ ll /usr/bin/systemd-resolve
lrwxrwxrwx 1 root root 10 nov. 15 21:42 /usr/bin/systemd-resolve -> resolvectl
$ type -a resolvectl
resolvectl is /usr/bin/resolvectl
$ file /usr/bin/resolvectl
/usr/bin/resolvectl: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 3.2.0, BuildID[sha1]=09e488e849e3b988dd2ac93b024bbba18bb71814, stripped
I have Fedora 25 and also had similar slow response on command line to sudo commands.
nmcli dev show | grep DNS
showed that only one of my 3 adapters (two active) had DNS entries. By adding DNS entries to the one active card that didn't have an entry - presto! All is good and response time is immediate.
In Ubuntu >= 15
nmcli device show <interfacename> | grep IP4.DNS
Replace <interfacename>
with yours.
In Ubuntu <= 14
The command
nmcli dev list iface <interfacename> | grep IP4
Replace <interfacename>
with yours.
Examples
nmcli device show eth0 | grep IP4.DNS
Or
nmcli dev list iface eth0 | grep IP4
This will list all DNS servers(If you use more than one).
-
-
-
1On debian i get an error--- $ nmcli dev list iface eth0 Error: 'dev' command 'list' is not valid. – don bright Oct 31 '15 at 15:00
-
-
On a recent Fedora 33, you can just use
# resolvectl dns
Global:
Link 2 (enp0s31f6):
Link 3 (wlp4s0): 1.1.1.1 8.8.8.8
Link 4 (virbr0):
Link 5 (virbr0-nic):
To check if systemd-resolved
is active, do cat /etc/resolv.conf
It will tell you who is controlling the file.
If you want more details, you can use resolvectl status
resolvectl statistics
, you can also flush caches ...
To troubleshoot, you can use journalctl -u systemd-resolved -f -o cat | grep Looking
after setting the level to DEBUG: sudo resolvectl log-level debug
On systems where systemd-resolved
is NOT installed :
$ host -v something.unknown | awk -F "[ #]" '/Received /{print$5}' | uniq
192.168.1.1
On systems where NetworkManager
is running :
$ ( nmcli -f IP4.DNS,IP6.DNS dev list || nmcli -f IP4.DNS,IP6.DNS dev show ) 2>/dev/null | awk '/DNS/{print$NF}'
192.168.1.1