Cloning a server across architectures

Posted on

Cloning a server across architectures – Managing your servers can streamline the performance of your team by allowing them to complete complex tasks faster. Plus, it can enable them to detect problems early on before they get out of hand and compromise your business. As a result, the risk of experiencing operational setbacks is drastically lower.

But the only way to make the most of your server management is to perform it correctly. And to help you do so, this article will share nine tips on improving your server management and fix some problem about linux, debian, central-processing-unit, architecture, .

I have this Linux (Debian) server that I want to duplicate, with all its services and their settings. Actually, it is to replace the existing server with newer hardware. I’d like the process to be as painless as possible.

So I was thinking, that I could do a fresh install of the same version of OS (with the appropriate binary format, of course) to the newer hardware. Then install the additional software packages (I have a list of all the apt-get install commands from the old machine that I can turn into a script of sorts. Then copy the /var, /etc, /home and /root directories over from the old machine to the new one.

My question is, has anyone tried it? What hiccups/pitfalls were encountered during the process? Was it even successful, or am I just planning an exercise in futility?

I am willing to go through if no one has tried it, and it should work “in theory”, so any other ideas are also welcome.

Solution :

Remember that when you do the command “apt-get install” on your new server, that would pull new version packages from repository, may get you in trouble of conflict packages.

I have another idea that I tried several years ago: You plug your new harddisk into current system, create partitions with the same layout of your Debian system (number of partitions, the last partition could be bigger due to differ from harddisk size). Then you boot into rescue usb, use ‘dd’ command to make a binary clone of each partition to new harddisk. After this phase, you would have the same data on new harddisk, except the bootloader. The last thing you should do is, plug your new harddisk into your new system, install a bootloader by yourself (via rescue usb), and boot into the OS.

Leave a Reply

Your email address will not be published.