Question Plesk Backup Manager Gateway Timeout

[gedanggoreng]

Hello,

Today I try to click "Backup Manager" on different customer panel users at the same server, and I found different commands executed.

On customer A its execute :

/usr/local/psa/admin/bin/pmm-ras --get-ftp-dump-list --lightweight-mode --dump-storage=ftps://[email protected]/ --type=domain --guid=61ee760e-fac0-4933-a529-702716541dd4 --name=domainA.com --session-path=/var/log/plesk/PMM


On customer B its execute :

/usr/local/psa/admin/bin/pmm-ras --get-dump-list --dump-storage=/mnt/rbackup/server50/ --type=domain --guid=0009f54f-ae1e-44e6-8cb2-aaab0aea392a --id=308 --name=domainB.net --owner-guid=d2beab4a-1b8a-47bb-bcab-8d9d028a662a --session-path=/var/log/plesk/PMM



On customer A the result always "Gateway Timeout".
On customer B the result always a success.

The user has the same service plan.

/mnt/rbackup/server50/ - is our backup server, mounted with nfs.

Our server :
Plesk Obsidian
Version 18.0.21 Update #5

When i manually execute this command (customer A without ftps) its always success:

/usr/local/psa/admin/bin/pmm-ras --get-dump-list --dump-storage=/mnt/rbackup/server50/ --type=domain --guid=61ee760e-fac0-4933-a529-702716541dd4 --name=domainA.com --session-path=/var/log/plesk/PM


Question:

Is the user/customer possible to change this?
Is this normal? If yes, where I can configure or fix this?

Thank you.

 

[Bitpalast]

The 504 is a result from the much longer time it takes to access the .tar-files on an FTP repository compared to accessing them directly on a local storage space (which is the main difference in your two examples here). It is not really much of a problem, although it makes the user feel as the server completely crashed. That is not the case. It simply takes too long to extract all files so that an sw-cp-server (the Plesk web server for the panel) timeout occurs while the software is still busy extracting files.

I remember that there are several FAQ entries on this. The problem can easily be solved by increasing the timeout of the Plesk web server.
I'd first start with this solution and adapt it to your needs:

Unable to download or access huge backup file in Plesk interface: 504 Gateway Time-out

Applicable to: Plesk for Linux Symptoms Unable to download or access huge backup file (more than 20 GB or with multiple incremental backups) in Tools & Settings > Backup Manager > Backu...

support.plesk.com

It does not exactly refer to your case, but the solution is still what you are looking for, because the underlying reason is the same.

There are several other reasons for a 504, you can find them all in the FAQs.

 

[gedanggoreng]

I follow the instruction to :
Change options :
fastcgi_read_timeout 1200
fastcgi_send_timeout 1200
max_execution_time = 1200

Restart service :
service sw-engine restart
service sw-cp-server restart

Timeout still happens and got these messages :
--
Warning: Could not display the list of backups in the FTP Storage: Unable to get ftp dumps list: FTP network error. Please check the FTP storage settings.
Warning: Unable to access the storage: Transport error: unable to list directory: Curl error: (28) Timeout was reached: Last FTP request: Last FTP response:
Make sure you have entered the correct storage settings.
--


But, after migrating the account to another Plesk server, this issue disappears.

Now, when I click "Backup Manager" its execute with option

--dump-storage=/mnt/rbackup/server12/

not

--dump-storage=ftps://[email protected]/.

But still, I don't know what exactly happens in the background.
Our backup setting is NFS mounted on /mnt/rbackup.
Maybe this is a very rare case, this happens on a single account only.