Archive Mirroring Guide

To ensure the longevity of the archive, you're encouraged to create your own local (or hosted, if needed) mirrors of both the archive content, and the website itself.

Both are supported by the unreal-archive tool.

Content Mirror

Before you begin, you will need the following:

1. Java 11 JRE
   • for Linux, install OpenJDK 11 JRE using your package manager
   • for windows, I would recommend Azul's Zulu Community Edition JRE
2. Download the unreal-archive tool
   • the latest can always be downloaded here
   • on Linux, you may download the unreal-archive binary, and make it executable via chmod +x unreal-archive
   • for Windows, download unreal-archive-exec.jar
3. Several hundred GB free disk space and data transfer capacity

To create a local mirror of all content (maps, skins, mods, etc) within the archive, run the unreal-archive tool with the following parameters:

Linux terminal

$ ./unreal-archive mirror /local/mirror/output --concurrency=5

Windows command prompt

\> java -jar unreal-archive-exec.jar c:\local\mirror\output --concurrency=5

A temporary copy of the archive metadata will be downloaded and extracted, loaded, and then begin downloading all the actual files to the specified /local/mirror/output directory.

If you want to tune the concurrency, adjust the --concurrency option's value. The default is 3.

The entire process will take a while, depending on the size of the content repository, available bandwidth, etc., but once completed, you'll have a complete local copy of everything the archive contains.

Website Mirror

The option exists to create a complete mirror of the archive website, with the intention that this can be easily re-generated and re-hosted at any point, should the original website go down, or if you simply want to create an offline copy of the website for your own use.

Before you begin, you will need the following (see above for more detail):

1. Java 11 JRE
2. Download the unreal-archive tool
3. Several GB free disk space

To create a local mirror of the website, run the unreal-archive tool with the following parameters:

Linux terminal

$ ./unreal-archive www /local/website/path --local-images=true --content-path=/path/to/unreal-archive-data

Windows command prompt

\> java -jar unreal-archive-exec.jar www C:\local\website\path --local-images=true 

This will generate a copy of the website, within the directory specified by /local/website/path. The option --local-images=true will also download all images needed by the website, to the output directory, rather than linking to them remotely.

You may then open the index.html file within the output directory to begin browsing the site.