vagrant stuck at "Setting hostname"

[Adirelle]

Since I upgraded Vagrant to 1.2.4 (from 1.2.2), "vagrant up" gets stuck at "Setting hostname". If I interrupt it and try to run a "vagrant reload", it gets stuck at "Attempting graceful shutdown of VM".

Reverting to 1.2.2 fixes this issue.

Context:

• Vagrant 1.2.4 without any plugin,
• VirtualBox 4.2.16,
• Custom Debian Squeeze base box, created with Vagrant 1.0,
• Windows 7 64-bits.

See https://gist.github.com/Adirelle/5e5a0ae063e9f02e23c7 for a debug log.

[arosenhagen]

got the same issue. I solved it by removing config.vm.hostname = "" from the Vagrantfile, then resetting the hostname inside the VM "manually" with chef.

[hectcastro]

The fix appears to be on master: bc7b481

[Adirelle]

I'm not sure how to get Vagrant working from sources on Windows to verify this. Can update it from what the installer installed ?

[suan]

Funny, running the latest version (1.2.7) doesn't fix this for me, even though it looks like the fix should already be in that version.

[colby-swandale]

getting the same issue as well running latest version (1.2.7)

[elijahandrews]

Also getting this issue using a custom Debian Squeeze base box on Vagrant 1.2.7. Reverting to 1.2.3 fixes the issue.

Other info: VirtualBox 4.2.16, Mac OS 10.8.4.

[jperras]

Same issue here.

While watching the INFO log, it gets stuck at the following:

 INFO ssh: Execute: ifdown -a; ifup -a; ifup -a --allow=hotplug (sudo=true)
 INFO subprocess: Starting process: ["/usr/bin/VBoxManage", "showvminfo", "60cfcca5-ae82-4906-afd1-1495dbacbd20", "--machinereadable"]
 INFO subprocess: Starting process: ["/usr/bin/VBoxManage", "showvminfo", "60cfcca5-ae82-4906-afd1-1495dbacbd20", "--machinereadable"]

Vagrant 1.2.7
VirtualBox 4.2.16

[spanneberg]

Observed the same issue on a colleagues machine running VirtualBox 4.2.16 / Vagrant 1.2.7 on a Ubuntu 12.04 LTS 64bit running a Debian Wheezy box (also 64 bit).

[ppp0]

Somehow eth0 is not affected by either ifup -a and ifup --allow=hotplug and once ifdowned, ssh obviously can't talk to the VM anymore. That's why it gets stuck

[njam]

@fpletz The restarting of networking was introduced by you in 65f275a to renew dhcp leases. This seems to let the box not being accessible in some cases.
Is there another way to accomplish what you want without restarting networking? I.e. just calling dhclient?

[fpletz]

@njam Thanks for mentioning me. Looking into this.

Could anybody please post the network configuration in your Vagrantfile and /etc/network/interfaces from the VM? Thanks.

[ppp0]

Vagrantfile

#[...]
config.vm.define :m3 do |m3|
    m3.vm.hostname = 'pppagent2'
    m3.vm.network :private_network, ip: "10.10.11.12"
end
#[...]

This Vagrantfile defines a second NIC for the private network. Ther first NIC ist automatically created (NAT networking)

/etc/network/interfaces

# This file describes the network interfaces available on your system
# and how to activate them. For more information, see interfaces(5).

# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interface
allow-hotplug eth0
iface eth0 inet dhcp
pre-up sleep 2
#VAGRANT-BEGIN
# The contents below are automatically generated by Vagrant. Do not modify.
auto eth1
iface eth1 inet static
      address 10.10.11.12
      netmask 255.255.255.0
#VAGRANT-END

Box is a debian squeeze amd-64

[adamcstephens]

Running 1.2.7, I am experiencing this with debian squeeze amd64 only.

This same config works without problems on centos5/6, debian7 and ubuntu1204.

  config.vm.network :public_network, :bridge => "en0: Wi-Fi (AirPort)"

  ['centos5', 'centos6', 'debian6', 'debian7', 'ubuntu1204'].each do |os|
    ['base', 'test'].each do |on|
      config.vm.define "#{os}#{on}".to_sym do |localconfig|
        localconfig.vm.hostname = "#{os}#{on}"
        localconfig.vm.box = "dev-#{os}"
      end
    end
  end

[mitchellh]

Fixed by #2026. Will be in next version. Sorry guys!

[haseebeqx]

same issue happening again with 2.0.4

[JoannaRMIT]

Same here

[haseebeqx]

for me the problem only happens with harshicorp/precise32 box, may be something related to that

[briancain]

@haseebeqx @JoannaRMIT Please open a new issue if you're encountering a problem, thanks!

[zackshapiro]

Glad I'm not alone

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.