Server relocations

Last change on 2021-10-12 • Created on 2020-03-25 • ID: RO-ACBAD

Introduction

The Hetzner Online data center parks are among the most innovative data centers in Germany. We want to continue to innovate; this is an important goal to the company. And we want to provide products that help as many of our customers as possible. Sometimes, that means we need to move or relocate some of our customers' older servers so we can make space for newer models. Whenever we do this, we inform customers in advance via email. If you have received one of these emails from us, please read this article carefully. The information in it can help the relocation go more smoothly. If you have any questions, please send us a support request on Robot. Note: You may also see the words phrase "server move" or "server migration" to mean "server relocation".

Server relocations

Keeping your data safe is our top priority. That's why, whenever we migrate servers, we are very careful. However, even when we do not expect there to be any data loss during a server migration, we strongly urge you to make a backup in advance. Please also check the condition of your drives and of your software/hardware RAID. If you notice any defective hardware, please let our team know before the migration. They will be happy to replace the defective hardware.

We have some useful articles here on Hetzner Docs if you need some help with these aspects of server administration:

Below you will find additional information about the relocation process.

  • Your server(s) listed in the email will not be accessible during the relocation.
  • Please shut down your server(s) yourself via a time-control (at-job) by the start of the relocation at the latest You can find instructions on how to do this and other tips below.
  • Once normal operations have resumed at the new location, we will let you know that your server(s) are functioning normally.
  • If you have any questions, please write a support request on Robot, and our support team will be happy to help you.

Tips

If you have installed a new kernel since the last time you restarted your server, and you are using the lilo bootloader, then you need to update your lilo configuration and run the lilo command to rewrite the boot sector of your drive.

Instructions for the time-controlled execution of jobs with at

Do the following steps to implement a time-controlled execution of an at-job. You will need an SSH connection with your server for this.

  • Please use the date command to make sure your server is using the correct system time. If it is not, you can adjust it using the command ntpdate. (Note: You need to install ntpdate or alternatively netdate):
    ntpdate ptbtime1.ptb.de && hwclock -w
  • In order to receive command information on the time-controlled execution of at commands, please use manpage from at at the command line of your server:
    man at

Please check to see if an at service is running on your server by searching atd in the process list:

    ps ax | grep atd
  • Start the at service. (If the service is not active, you will need to install it):
    /etc/init.d/atd start
  • Please create a new time-controlled at job for your system to execute on the relocation date/time, for example, June 25, 2019 at 09:15 AM:
    echo "/sbin/shutdown -h now"  |  at  09:15  25.06.2019
  • Use the following command to make sure your at job is ready:
    atq
Table of Contents