Empty /etc/resolver.conf

[Lachlan Donald]

When I've booted my VM, I see an empty /etc/resolver.conf and I can't resolve DNS addresses inside my VM.

This is the bug: https://bugs.busybox.net/show_bug.cgi?id=6788

This seems to be the fix: boot2docker/boot2docker#129 (comment)

[YungSang]

@lox Thank you for letting me know.
It seems fixed already.. Hmm

[Lachlan Donald]

Hrm, odd. I wonder why my /etc/resolver.conf is empty.

[YungSang]

I've known that issue, because a few users have reported to me on Twitter.
But I've never seen that on my side and never tried the fix.
Let me investigate for a while.

For now, please put the following at the top of Vagrantfile and try it.

  config.vm.provision :shell do |s|
    s.inline = <<-EOT
      sudo /sbin/udhcpc
      cat /etc/resolv.conf
    EOT
  end

Thanks.

[YungSang]

@lox BTW, which version of yungsang/boot2docker do you use?

[Lachlan Donald]

0.9.0, with the latest VirtualBox and Docker 1.6.2.

[Lachlan Donald]

That works nicely, thank you.

[YungSang]

Thanks. That's nice to hear.
Anyway I will comment here when I get something.

[YungSang]

VM created by init does not have DNS set up correctly · Issue #102 · boot2docker/boot2docker-cli
boot2docker/boot2docker-cli#102

jpetazzo/pipework
https://github.com/jpetazzo/pipework

If you use VirtualBox, you will have to update your VM network settings. Open the settings panel for the VM, go the the "Network" tab, pull down the "Advanced" settings. Here, the "Adapter Type" should be pcnet (the full name is something like "PCnet-FAST III"), instead of the default e1000 (Intel PRO/1000). Also, "Promiscuous Mode" should be set to "Allow All". If you don't do that, bridged containers won't work, because the virtual NIC will filter out all packets with a different MAC address. If you are running VirtualBox in headless mode, the command line equivalent of the above is modifyvm --nicpromisc1 allow-all. If you are using Vagrant, you can add the following to the config for the same effect:

config.vm.provider "virtualbox" do |v|
  v.customize ['modifyvm', :id, '--nicpromisc1', 'allow-all']
end

[YungSang]

Ooops, I closed accidentally. Re-opened.

[YungSang]

https://bugs.busybox.net/show_bug.cgi?id=6788

However, using scripts which rewrite resolv.conf
only based on DHCP data is bound to be problematic
in a lot of non-trivial networking setups.
You can't fix the script to work well for everybody.

Please read http://busybox.net/~vda/no_ifup.txt

[YungSang]

In conclusion, I think the best solution for now is

  config.vm.provision :shell do |s|
    s.inline = <<-EOT
      sudo /sbin/udhcpc
      cat /etc/resolv.conf
    EOT
  end

[YungSang]

Final version

  # Fix busybox/udhcpc issue
  config.vm.provision :shell do |s|
    s.inline = <<-EOT
      if ! grep -qs ^nameserver /etc/resolv.conf; then
        sudo /sbin/udhcpc
      fi
      cat /etc/resolv.conf
    EOT
  end

[YungSang]

Wrapup

# -*- mode: ruby -*-
# vi: set ft=ruby :

# Vagrantfile API/syntax version. Don't touch unless you know what you're doing!
VAGRANTFILE_API_VERSION = "2"

Vagrant.configure(VAGRANTFILE_API_VERSION) do |config|
  config.vm.box = "yungsang/boot2docker"

  config.vm.network "private_network", ip: "192.168.33.10"

  config.vm.synced_folder ".", "/vagrant", type: "nfs"

  # Fix busybox/udhcpc issue
  config.vm.provision :shell do |s|
    s.inline = <<-EOT
      if ! grep -qs ^nameserver /etc/resolv.conf; then
        sudo /sbin/udhcpc
      fi
      cat /etc/resolv.conf
    EOT
  end

  # Adjust datetime after suspend and resume
  config.vm.provision :shell do |s|
    s.inline = <<-EOT
      sudo /usr/local/bin/ntpclient -s -h pool.ntp.org
      date
    EOT
  end
end

[YungSang]

boot2docker/boot2docker#468
Confirmed, it won't need sudo /usr/local/bin/ntpclient -s -h pool.ntp.org any more on resume