When I backup using Virtualmin, it does that with a single tar file per domain, per day etc. You can customise this using the date format. I can the restore sections, web dir, email files etc. as per restore options. These backups are not designed for you to unzip and restore individual folders, the system will restore for you.
When I backup using the file system backup, I do this when above isn’t specific enough or I want more control of random folders I backup. Again, using date format I do per day/date/ per domain etc. You can use this backup service to zip up your public_html if you don’t want to use the backup service as above.
He split the public_html folder into multiple sections.
Actually, I can combine all public_html folders into one folder, but this is terrible for large files.
Folders like “etc, cgi” are separate yes, no problem.
But I want public_html to be one piece.
If one day I want to leave Virtualmin, how will another panel merge these disparate folders?
Doesn’t make sense for a universal backup. I want all files to stay in one place in public_html.
On standard Virtualmin installs, virtual servers are placed inside the /home directory thus:
/home/xserver
/home/yserver
Now, inside each virtual server is a public_html directory thus:
/home/xserver/public_html
And inside this public_html is content that the owner of the virtual server / user uploads.
In the vid that you recorded, I see a few numbered directories at the beginning of the recording. What is the relative path to those directories in your archive? We need to know this before we can tell you anything definitive.
The only wild idea I have at this point is that your archive is so large that the archiving utility is somehow splitting one directory into multiple numbered directories to manage the archive internally, but that is an explanation in the absence of any other reasonable explanation.
If you try to use Virtualmin backup on a known small sized virtual server / domain, do you still see numbered directories or is everything in public_html as you would expect?
Thank you for your interest.
I have now made a backup of my server using the Virtualmin “Scheduled Backup” option. The size of the public_html archive file of this server is “40.69 MiB.”
Then I downloaded this backup as “.tar.gz”.
Now I want to show you a gif of extracting “.tar.gz” and what happened next.
Please watch this with care:
1- I downloaded the backup directly via FTP.
2- I started extracting with “.tar.gz”.
See how it then shreds a directory that isn’t even 100 MB?
If this isn’t a problem unique to my server, it’s terrible.
Hello Joe,
Maybe I should explain the problem more clearly.
Different servers have different public_html from each other, that’s true.
My problem is this:
When I make a backup of a single site, the result is still the same.
The fragmented public_html I posted above is a backup of a single site.
When you back up your home directory it is going to back up every virtual server you have in that directory separately.
Go to the Webmin tab and look under tools and hit the file manager. Then go to the home directory. That will show you the entire file structure that you’re actually backing up and then you’ll understand why things are where they are.
I don’t know how to explain my problem more clearly.
I only choose 1 website when backing up. So 1 server. There is a simple Wordpress installed.
It shatters its public_html folder.
I `m talking about this.
There is not more than one site involved.
There isn’t a huge amount of data. (100MB)
It even shreds the backup of 1 site.
That’s my problem.
I have no experience with 7-zip, but I can say that Virtualmin backups are standard tarballs made with GNU tar (or whatever tar is on your system, but all of our supported Linux distros use GNU tar). It’s Open Source, and the most popular tar implementation in the world. We literally could not make it more standard or more accessible across any platform than using GNU tar.
