I want to get fast dns resolution with dnsmasq and keep the default systemd-resolved.
Looking for an elegant way to do this
Ask Ubuntu is a question and answer site for Ubuntu users and developers. It only takes a minute to sign up.Sign up to join this community
I wanted to get fast dns resolution with dnsmasq and keep the default systemd-resolved/NetworkManager setup untouched for future use. Yes the huge dns caching of dnsmasq can improve browsing speed. Yes the goal was to keep the default featured dns setup of 18.04
1 - With sudo
apt-get -y install dnsmasq
2 - With sudo
tee -a /etc/dnsmasq.conf << ENDdm interface=lo bind-interfaces listen-address=127.0.0.1 # DNS server from OpenDns. Use yours... server=22.214.171.124 server=126.96.36.199 ENDdm systemctl restart dnsmasq systemctl enable dnsmasq
3 - With USER, configure NetworkManager
# Get NM first active profile name NetManProfile=$(nmcli -t connection show --active | cut -f 01 -d ':') # remove, if exists, current dns servers nmcli con mod "$NetManProfile" ipv4.dns "" # set 'manual' dns server nmcli con mod "$NetManProfile" ipv4.ignore-auto-dns yes # set dnsmasq as manually set dns server nmcli con mod "$NetManProfile" ipv4.dns 127.0.0.1 # i also disabled ip6, do what u want nmcli con mod "$NetManProfile" ipv6.method ignore # reconnect to take effect nmcli connection down "$NetManProfile" nmcli connection up "$NetManProfile"
4 - Check verify
netstat -antup Proto Recv-Q Send-Q Adresse locale Adresse distante Etat PID/Program name tcp 0 0 127.0.0.1:53 0.0.0.0:* LISTEN 1036/dnsmasq tcp 0 0 127.0.0.53:53 0.0.0.0:* LISTEN 766/systemd-resolve cat /run/systemd/resolve/resolv.conf nameserver 127.0.0.1
I have a specific use case that works great. I run dnsmasq on my LAN router (an Ubuntu server machine, with no systemd-resolved), and let the LAN machines behind the router default to vanilla systemd-resolved DNS resolution. It's all possible and works elegantly, with a few tweaks to dnsmasq:
# Make clients that request IPs use this box for DNS dhcp-option=option:router,192.168.0.1 domain=mydomain.lan local=/mydomain.lan/ expand-hosts
Now I can stand up a gazillion Ubuntu VMs inside my LAN and never have to fiddle with DNS any more - it just works.
The tweaks are required because systemd-resolved does not allow you to use "single-label" host names (with no dot in them), unlike dnsmasq and "classic DNS". Once you get dnsmasq to automatically extend LAN host names into FQDNs, everything is happy. This took me a LONG time to figure out, btw. These systemd-resolved issues 1 2 helped me crack the problem.
I tried to find a reasonable solution and looks that there are different approaches.
I wanted to stay at most within the distribution layout while keeping all business requirements fulfilled. This is what I collected around and tested to work on clean Ubuntu 18.04 and KDE Neon flavour:
# Install required package and reconfigure service plans (i.e. disablesystemd-resolved, enable dnsmasq sudo apt-get install dnsmasq sudo systemctl disable systemd-resolved sudo systemctl stop systemd-resolved sudo systemctl enable dnsmasq # These two lines should work on most environments, but .. :-) - so I kept them commented out for less experienced users # Just add or change 'dns=dnsmasq' to your NetworkManager.conf to the section [main] # and yes, the sed expression can be better :-) #sudo cp /etc/NetworkManager/NetworkManager.conf /etc/NetworkManager/NetworkManager.conf.backup #sudo bash -c 'cat /etc/NetworkManager/NetworkManager.conf.backup |sed -e "s/^dns=.*//"| sed -e "s/\[main\]/\[main\]\ndns=dnsmasq/" >/etc/NetworkManager/NetworkManager.conf' # Restart NetworkManager to make the change above applied sudo systemctl restart NetworkManager # This removes the systemd resolv.conf link only if it has NetworkManager replacement :-) ls /var/run/NetworkManager/resolv.conf && sudo rm /etc/resolv.conf # And add NetworkManager's resolv.conf available for the system resolver sudo ln -s /var/run/NetworkManager/resolv.conf /etc/resolv.conf
(please note that the only general difference with the above answers is that the NetworkManager handle the dnsmasq DNS server assignments automatically
As you know, Docker copy host /etc/resolv.conf file to containers but removing any local nameserver.
My solution to this problem is to keep using systemd-resolvd and NetworkManager but add dnsmasq and use it to "forward" Docker containers DNS queries to systemd-resolvd.
Step by step guide:
sudo rm /etc/resolv.conf sudo touch /etc/resolv.conf
[main] # NetworkManager will push the DNS configuration to systemd-resolved dns=systemd-resolved # NetworkManager won’t ever write anything to /etc/resolv.conf rc-manager=unmanaged
sudo apt-get -y install dnsmasq
# Use interface docker0 interface=docker0 # Explicitly specify the address to listen on listen-address=172.17.0.1 # Looks like docker0 interface is not available when dnsmasq service starts so it fails. This option makes dynamically created interfaces work in the same way as the default. bind-dynamic # Set systemd-resolved DNS server server=127.0.0.53
# systemd-resolvd name server nameserver 127.0.0.53 # docker host ip nameserver 172.17.0.1
sudo service network-manager restart sudo service dnsmasq restart sudo service docker restart
For more info see my post (in spanish) https://rubensa.wordpress.com/2020/02/07/docker-no-usa-los-mismos-dns-que-el-host/
IMHO, if your going to be running dnsmasq, you should statically assign your ip address instead of getting it from dhcp. This way you can just disable systemd-resolved all together.
sudo apt-get install dnsmasq
sudo systemctl disable systemd-resolved
sudo systemctl stop systemd-resolved
Manually assign your ip address, gateway, and assign the ip address to your machine as DNS.
configure /etc/dnsmasq.conf (really...RTFM --> man dnsmasq.conf)
sudo systemctl enable dnsmasq
sudo systemctl status dnsmasq
point dhcp on your dhcp server to your shiny new dnsmasq server (..if yumpto)