Issue Plesk Backup Aborts at ~128 GB Uploading to FTP (Curl Error 55)

[audiobras]

Server operating system version

AlmaLinux 9.5 (Teal Serval) and Centos 7 in ftp server

Plesk version and microupdate number

18.0.68 #2

Hello Plesk Community,

I’m encountering a problem when sending full-server backups from my Plesk panel to a remote FTP server. The transfer always stops at approximately 128.79 GB, with a steady upload speed (~109–110 MB/s), and Plesk logs the following error:

Unable to create the remote backup:
Curl error: Unable to resume an interrupted upload: (55) Failed sending data to the peer

In the /var/log/xferlog, each attempt shows the file marked as “i” (incomplete), and only a few kilobytes are ever retried before Plesk aborts and deletes the partial archive. Example excerpt:

Fri Mar 28 02:34:40 2025 1167 xxxxxxxx 128789846920 /xxxxx/backup_2503272315.tar b _ i r rootftp ftp 0 * i
Fri Mar 28 02:34:44 2025 1 xxxxxxxx 14480 /xxxxxx/backup_2503272315.tar b _ i r rootftp ftp 0 * i

...

In my panel.ini, I’ve set:
streamCacheSize = 1024
ftpMaxResumeAttempts = 10
ftpMaxResumeFailures = 15
ftpFileTransferTimeout = 36000
ftpForbidReuseConnection = 1
backupsValidationPeriod = 0

What I’ve tried so far:

1. Adjusted streamCacheSize in panel.ini (currently 1024) to increase buffer, without success.
2. Verified network stability—no packet loss or obvious timeouts on my side.
3. Confirmed the partial file is always purged by Plesk immediately after the abort.

Questions

1. Could any of these settings be causing the backup to abort at ~128 GB?
2. What quick configuration tweaks or tests can I try—either in Plesk or on the FTP server—to allow the backup to complete?

Thanks in advance for any pointers!

Delton

 

[mixos]

Hi Delton,

Did you find a solution for this large backup? I'm experiencing the same issue with a 100GB backup file.
Incremental backups work fine, but when it comes to full backups, I get

Curl Error 55​

and unable to upload to FTP

 

[audiobras]

Hi,

Thanks for your message — it’s helpful to know I’m not alone in this!

I haven’t found a solution yet. I haven’t tested incremental backups so far — only full-server backups. I recently tried switching the destination to Dropbox instead of FTP, but even there, the full backup takes several hours and sometimes fails as well.

I’m still unsure whether the issue is on the Plesk server side or with the remote FTP server. I’m planning to test with a different FTP destination soon to try to isolate the cause.

Out of curiosity, could you share some details about your FTP server? (Type, OS, or provider, if applicable?) It would be interesting to see if our environments are similar.

If I find any workaround, I’ll be sure to update here — and I’d appreciate the same from you!

Best regards,
Delton

 

[mixos]

I use a Hetzner dedicated server with Plesk, and I configured it to take a backup at a "Storage Box" (1TB) via FTP(S).

The issue was that there was not enough space left on the Storage Box, although it had only one older backup (100GB); I supposed that there was enough space left.

The truth was that some box snapshots occupied almost all the box storage. (I never enabled snapshots).

These snapshots are clones of the box, and I am not aware of how they were created or progressively grew.

Also, I didn't find a way to reset the box to the original state or delete the snapshots.

So decided to go with a new storage box, and everything worked like a charm. Check below how the usage indicator looked.

 

[audiobras]

Thanks for the info.

It’s possible the issue on my side is also related to the FTP destination. I’ll check for hidden disk usage, quotas, or anything that might be limiting available space, even though it seems to have enough.

I also plan to test with a different FTP server to help isolate the cause.

Delton