Issue Backup disk usage spiked about a month ago.

[FutureX]

Server operating system version

AlmaLinux 8.8 (Sapphire Caracal)

Plesk version and microupdate number

Plesk Obsidian Version 18.0.52 Update #3

I'm attaching a screen shot of the disk usage. Nothing had changed with this server, no new accounts, etc. but around Apr 26 the disk usage started to spike. I believe it's from the backups?

If I look in the /var/lib/psa/dumps directory there are 53387 files. does that sound normal? I almost want to wipe all the backups and let it start fresh. I have external backups on DropBox.

Just trying to track down what caused the spike in disk usage. I manage all the sites on this server and nothing has really changed.

 

[Peter Debik]

This can have different reasons.

Local backups are stored in many separate files. So if you do local backups, this is an expected behavior. Maybe in addition to the remote backup you have also told Plesk to store remote backups locally?

It is also thinkable that in addition to your default Dropbox backup you have an extra backup configured, either as admin or as a subscriber. Please descend into /var/lib/psa/dumps/clients to see the accounts for that local backups are stored. Maybe this hints at the source.

If you have configured remote backups, you have an option that lets you automatically store a backup locally in case the transport to the remote repository fails, so that you still have a backup even if the remote storage location cannot be reached. In that case you see a local backup entry of the affected backup in your Backup Manager in Plesk. This is the setting in Backup Manager "Settings" tab where you can disable the option:


On scheduled backups you also have an option to not only store the backups in your remote storage location, but also keep a copy in your local storage. This is the option from the "Schedule" button in Backup Manager:



Do any of these match your situation?

 

[FutureX]

That's for the help!

I do have Plesk store backups locally and on remote storage but that's what I've always done.

I also have backups set to run individually for each client, in addition to the full server backup but that's what I've always done.

I do have the option set to store locally if the remote repository fails but the remote repository hasn't failed.

Great suggestions but I'm trying to find what changed around the end of April that caused the disk usage to spike suddenly. I haven't changed anything in terms of backup settings, etc. I wish I had the option of saving to multiple remote storage locations. Then I could do away with the local storage. It would be a little too risky to only have one remote backup.