Fresh Install Ubuntu 14.04 Server Slow Boot due to failure to connect 169.254.169.254

Question

Have just installed Ubuntu Server 14.04 and have no VM instances operational. Boot time takes an eternity and on pressing F1 during loading this is what I see, the continued connect attempts and failures are endless any straightforward fixes would be appreciated (Note have replaced http with LINK in the log mentions to enable this posting):-

Cloud-init v. 0.7.5 running 'init-local' at Tue, 06 May 2014 19:42:27 +0000. Up 359.24 seconds.
Cloud-init v. 0.7.5 running 'init' at Tue, 06 May 2014 19:42:31 +0000. Up 363.43 seconds.
ci-info: +++++++++++++++++++++++Net device info++++++++++++++++++++++++
ci-info: +--------+-------+-----------+-----------+-------------------+
ci-info: | Device |   Up  |  Address  |    Mask   |     Hw-Address    |
ci-info: +--------+-------+-----------+-----------+-------------------+
ci-info: |   lo   |  True | 127.0.0.1 | 255.0.0.0 |         .         |
ci-info: |  em1   | False |     .     |     .     | 9c:b6:54:02:0d:ab |
ci-info: +--------+-------+-----------+-----------+-------------------+
ci-info: !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!Route info failed!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2014-05-06 20:42:33,252 - url_helper.py[WARNING]: Calling 'LINK://169.254.169.254/2009-04-04/meta-data/instance-id' failed [0/120s]: request error [HTTPConnectionPool(host='169.254.169.254', port=80): Max retries exceeded with url: /2009-04-04/meta-data/instance-id (Caused by : [Errno 101] Network is unreachable)]
2014-05-06 20:42:34,255 - url_helper.py[WARNING]: Calling 'LINK://169.254.169.254/2009-04-04/meta-data/instance-id' failed [1/120s]: request error [HTTPConnectionPool(host='169.254.169.254', port=80): Max retries exceeded with url: /2009-04-04/meta-data/instance-id (Caused by : [Errno 101] Network is unreachable)]
2014-05-06 20:42:35,259 - url_helper.py[WARNING]: Calling 'LINK://169.254.169.254/2009-04-04/meta-data/instance-id' failed [2/120s]: request error [HTTPConnectionPool(host='169.254.169.254', port=80): Max retries exceeded with url: /2009-04-04/meta-data/instance-id (Caused by : [Errno 101] Network is unreachable)]

Answer 1

You must be running the cloud version of Ubuntu server. cloud-init is attempting to contact 169.254.169.254, a link-local IP address that cloud providers listen on to provide metadata about the VM being initialized.

You can prevent cloud-init from querying any remote servers by setting the datasources_list setting to None.

In practice, this means changing the file /etc/cloud/cloud.cfg.d/90_dpkg.cfg from:

# to update this file, run dpkg-reconfigure cloud-init
datasource_list: [ NoCloud, ConfigDrive, OpenNebula, Azure, AltCloud, OVF, MAAS, GCE, OpenStack, CloudSigma, Ec2, CloudStack, None ]

to:

# to update this file, run dpkg-reconfigure cloud-init
datasource_list: [ None ]

If possible, you'll want to run dpkg-reconfigure cloud-init to make the change.

Answer 2

Yes, and in our case, we had to use [ OpenStack ] as opposed to [ None ] in our cloud image's /etc/cloud/cloud.cfg.d/90_dpkg.cfg file, as that was the acceptable Datasource type for our cloud OS. This issue manifested when we started using our existing icehouse friendly images in a juno cloud. No other OS had this issue except our 14.04 image.
We also based our 14.04 image on the Canonical Cloud version.

Answer 3

In my case (new install using Ubuntu Juju) the problem was that neutron was misconfigured with the wrong ip address for the metadata server.

This is good: http://techbackground.blogspot.ie/2013/06/metadata-via-dhcp-namespace.html

That pointed me at: root@neutron:/etc/neutron# cat metadata_agent.ini

    ...
  nova_metadata_ip = 192.168.100.151
  nova_metadata_port = 8775
    ...

I fixed the nova_metadata_ip adn restarted the neutron metadata service and that solved the issue.