[spry-jd]

The backup file obtained from the Plesk Backup utility is a multi-part mime-encoded file. Each component within the file is for a specific purpose or location. There are individual areas for the configuration files, databases, SSL certificates, mail, and websites. The backup file is useful if you wish to restore an entire site or server. However if you need only a specific directory or a selection of files, the restore functionality with Plesk does not allow this. Instead you will need to follow the instructions below to separate the components and then extract the information you need.


The process is to use a program like ripMime (project page: http://freshmeat.net/projects/ripmime/) to extract out the different mime-encoded components of the backup file.

To download this program into the current directory (I would suggest doing this in /root) use the command

wget http://freshmeat.net/redir/ripmime/9...1.4.0.5.tar.gz

and then extract it with the command

tar -zxvf ripmime-1.4.0.5.tar.gz

Once the program is downloaded and installed, you can then run the command

/path_to_file/ripmime /path_to/backup_file /directory_for_extracted_files

to extract the individual mime-encoded components into the directory you specified.

Once the files are extracted, you'll be able to use the 'tar' command to extract all or only specific files from the backup files present. You'll need to identify which file contains the information you are looking for, so you will need to run the command

tar -ztvf /directory_for_extracted_files | grep directory_or_filename

Once you know which file contains the information you are searching for, you can then extract the directory or file with the command

tar -zxvpf /directory_for_extracted_files/filename_containing_information directory_or_filename

Once the directory or file is extracted, you can then move it to the proper location. The permissions and file ownership should be correct as long as you use the 'mv' command since we preserved ownership and permissions with the 'p' flag in the 'tar' command.

[sourcez]

This was a great help, thanks! I was able to decode a PLESK backup for a domain (after some trial and error!)

In addition - you basically need a MIME decoder. There are also some available for windows like this one (apologies if link is not allowed)
miken.com/uud/
That also worked a charm on my laptop.

Thanks again for the great guide!

[HarleyToday]

Another option is to use Winzip. I needed to restore multiple Plesk domains after the included Plesk restore program failed to restore all of the site content. Winzip opened the full backup file and allowed all files needed to be extracted. My process was to first extract the files to a secondary storage device and then copy them to affected domains using the IDs and passwords so that all permissions would be correct.

The full back up file does contains all files required to recreate a site from scratch, including ID's, passwords, site properties for each domain.

[nathan3011]

Sorry about this guys but just looking for a little more help if you could?

I'm just wondering if the easiest way is to use Winzip or is it the first way because i tried it using winzip and couldn't get it to work??
Thanks in advance

[Patrick@Spry]

Hello,

Tar is the easiest way to extract and zip on a linux server. It works very similar to pkzip back in the windows 3.1 days. But keep in mind that winzip doesn't exist for linux as it is a windows application.


--Patrick

[gelsner]

I use the Windows XP Pro operating system and have tried lots of versions of WinZip and cannot get it to open the *.gz file from PLESK.

Which version of Winzip worked for you?

(I just tried the most recent one today...I think...Winzip120).

Your help will be greatly appreciated!!

[Patrick@Spry]

Hello,

You shouldn't need to unzip plesk on your local machine as it's not a gui application. If you really need to unzip a .gz file, you'll need an application that can do it like winrar or izarc.