How to install RDO (OpenStack) in Hetzner?

Question

I have tried to install OpenStack on Hetzner (dedicated server), but it does not finish the installation correctly.

I have followed this official guide:

RDO Installation

The same installation was done on computers at home and I didn't have any problems.

$ sudo yum install -y centos-release-openstack-pike
$ sudo yum update -y
$ sudo yum install -y openstack-packstack
$ sudo packstack --allinone

$ sudo systemctl disable firewalld
$ sudo systemctl stop firewalld
$ sudo systemctl disable NetworkManager
$ sudo systemctl stop NetworkManager
$ sudo systemctl enable network
$ sudo systemctl start network

$ sudo yum install -y centos-release-openstack-pike
yum-config-manager --enable openstack-pike
$ sudo yum update -y

$ sudo yum install -y openstack-packstack
$ sudo packstack --allinone

The problem is sure to be that Hetzner offers single floating IPs or RIPE blocks directly routed to your machine.

Someone has had this problem who can provide a little help please.

I add here error information:

Applying 94.130.220.61_controller.pp
Testing if puppet apply is finished: 94.130.220.61_controller.pp  [ / ]
94.130.220.61_controller.pp:                      [ ERROR ]
Applying Puppet manifests                         [ ERROR ]

ERROR : Error appeared during Puppet run: 94.130.220.61_controller.pp
Error: /Stage[main]/Nova::Db::Sync/Exec[nova-db-sync]: Failed to call refresh: Command exceeded timeout
You will find full trace in log /var/tmp/packstack/20180129-090526-jGw32P/manifests/94.130.220.61_controller.pp.log
Please check log file /var/tmp/packstack/20180129-090526-jGw32P/openstack-setup.log for more information
Additional information:
 * A new answerfile was created in: /root/packstack-answers-20180129-090528.txt
 * Time synchronization installation was skipped. Please note that unsynchronized time on server instances might be problem for some OpenStack components.
 * File /root/keystonerc_admin has been created on OpenStack client host 94.130.220.61. To use the command line tools you need to source the file.
 * To access the OpenStack Dashboard browse to http://94.130.220.61/dashboard .
Please, find your login credentials stored in the keystonerc_admin in your home directory.

Pastebin of log error

Thank you

Answer 1

Had the same issue, fixed it by changing the timeout value for puppet.

/usr/share/openstack-puppet/modules/nova/manifests/db/sync.pp

class nova::db::sync(
  $extra_params    = undef,
  $db_sync_timeout = 300,
)

Change $db_sync_timeout from 300 to 3600

I had to make the same changes for neutron /usr/share/openstack-puppet/modules/neutron/manifests/db/sync.pp

Answer 2

Had the same issue on Centos7, Packstack Train. Hardware HP DL580 with 40CPU with 80Threaads, 8 x SSD in RAID0. I made small ansible task if anyone deploys Packstack with ansible

    - name: Fixup for issue at "https://serverfault.com/questions/894327/how-to-install-rdo-openstack-in-hetzner"
      become: yes
      become_user: root
      lineinfile:
        path: "/usr/share/{{ item }}"
        regexp: ^  \$db_sync_timeout = \d+,$
        line:   "  $db_sync_timeout = 0,"
        state: present
        backup: no
      loop:
        - "openstack-puppet/modules/nova/manifests/db/online_data_migrations.pp"
        - "openstack-puppet/modules/nova/manifests/db/sync.pp"
        - "openstack-puppet/modules/nova/manifests/db/sync_api.pp"
        - "openstack-puppet/modules/neutron/manifests/db/sync.pp"

Answer 3

Make sure you have installed NTP service and it works correctly. As you can see in the packstack result:

* Time synchronization installation was skipped. Please note that unsynchronized time on server instances might be problem for some OpenStack components.