• If you are still using CentOS 7.9, it's time to convert to Alma 8 with the free centos2alma tool by Plesk or Plesk Migrator. Please let us know your experiences or concerns in this thread:
    CentOS2Alma discussion

Resolved Backup broken since update to 18.0.50 (when using FTPS+"active mode")

lj-shadow

New Pleskian
Server operating system version
Ubuntu 22.04.1 LTS
Plesk version and microupdate number
18.0.50
Hello,
Since the update to 18.0.50, the backups to the FTP server no longer work.
Under "Tools & Settings -> Backup Manager" I only get a white page with endless loading.
So this setting isn't available/usable anymore.

The daily task for incremental backup (to FTP only) runs in an timeout but the hint to check it via cli by

" curl -v -P - --ssl -k -u USERNAME 'URL' "

works like a charm.
What can I do?
 
Same here. On all our servers.

I do apologize up front, but

EVERY ... SINGLE .... TIME ... Plesk changes something to with [FTP] backups, THEY ... F**K ... IT UP !!!!

We've been using Plesk since version 8 (!). Backups, ... EVERY SINGLE TIME!!
I really really have to hold back not to plaster this post with swearwords...
 
Could you please check into /var/log/plesk/PMM/backup-2023-MM-DD-HH-ii-... of the hanging backup? Do you see any error messages in the pmmcli.log or backup.log files?
 
Hi Peter,

the pmmli.log shows no errors at all.
the backup.log shows:

[2023-02-05 00:50:39.987|1011371] INFO: Find incremental backups in '' for prefix 'backup' and base version 2301250040
[2023-02-05 00:50:39.995|1011371] INFO: 0 backups found in cache
[2023-02-05 00:50:39.995|1011371] INFO: Backup info not found in cache
[2023-02-05 00:50:39.995|1011371] INFO: Ftp init url ftps://xyz.your-storagebox.de//somefolder/
[2023-02-05 00:55:40.179|1011371] INFO: TransportError Transport error: unable to list directory: Curl error: (28) Timeout was reached: Last FTP request: LIST Last FTP response: 150 Opening ASCII mode data connection for file list [common/plesk-utils/PMM/repository-transport/transport.cpp:TransportError]
virtual void plesk::tRepositoryFtp::ListDirEx(const string&, std::__cxx11::list<plesk::FileInfo>&)
[2023-02-05 00:55:40.180|1011371] INFO: Repository 'ftps://xyz.your-storagebox.de//somefolder': Delete packed backup backup_2302050045.tar
[2023-02-05 00:55:40.180|1011371] INFO: Ftp init url ftps://xyz.your-storagebox.de//somefolder/backup_2302050045.tar
[2023-02-05 00:55:40.565|1011371] INFO: Pause before next attempt...
[2023-02-05 00:55:41.589|1011371] INFO: Pause before next attempt...
[2023-02-05 00:55:42.624|1011371] INFO: RepositoryError Repository error: file backup_2302050045.tar not exists [common/plesk-utils/PMM/repository-transport/repository.cpp:RepositoryError]
virtual void plesk::tRepository::DeletePackedBackup(const string&)
[2023-02-05 00:55:42.624|1011371] INFO: pmm-ras finished. Exit code: 151

I don not believe that the transport error is the main reason. We have had that on rare occasions in the past, but was never anything to worry about.

When you go Settings&Tools, Tools&Resources, Backup Manager, it never opens the page that's supposed to be displayed. The just sits there waiting for a response. It like its waiting for input that never comes.
 
Forgot the mention, all servers are on Almalinux 8, latest update. In case that helps narrow it down.

Noticed the Extension "Scheduled Backups List" bug. Can now access that, however clicking on Backup Manager in the breadcrumbs has the page loading endlessly again.
 
@TomBoB Thank you for the details. I went ahead and created a ticket for developers to check into the issue. In case you have questions on this case to support staff, please refer to ID [PPS-13913]. I'll have an eye on it, too.
 
Support engineers are asking for more information:
1) Please enable "Debug mode" as described in https://support.plesk.com/hc/en-us/articles/213373669. Please make sure that "FORCE_DEBUG_LOG 1" is set as described in the "For Scheduled Backup Tasks" section.
2) Then let the scheduled backup run and also, please run backup manually.
3) Please provide all logs generated from this (/var/log/plesk/PMM/backup-**/). I suggest sending them in a private message to me, not to expose the details publicly here on the forum.
 
Hi,

can't access the GUI page [still loads endlessly] to run the manual and scheduled backups.

From CLI I get after a few seconds:
Code:
plesk bin pleskbackup --server -ftp-server xyz.your-storagebox.de -ftp-login 123loginuser -ftp-password 123pw

The backup failed with errors!
exit status 255

Will send logs shortly.
 
Figured it out.

Log into your servers, go straight to path


untick "Use FTPS"

Save.

Working again.

I feel I may have to apologize to Plesk. It may after all not have been caused by a Plesk update issue but by curious coincidence with Hetzner doing something their side.
Still would be nice to get backups to FTP using FTPS going again. Plus Plesk detecting the non-connection / timeout when opening the Backup Manager, instead of just sitting there until an eventual timeout occurs and a 503 bad gateway is displayed.

tested above on 3 servers so far and the "workaround" did the trick on all.
 
Thank you for the update. Good to have a workaround, but FTP without encryption cannot be a final solution. We'll keep checking this for sure. It must also work with FTPS.
 
@TomBoB Ever thought of using sshfs to mount the storagebox directly on your server? That way, you can skip the FTP part and create backups to something like: /mnt/storagebox

@maartenv Yes we did look into that a long while back, but it wan't feasible for us for multiple reasons. But will definitely look into that again! Thanks!
 
@lj-shadow, @TomBoB could one or both of you please open a ticket with Plesk support and allow SSH access to your server(s) for support? Developers would like to check the issue directly on the server, because we were yet not able to determine the cause. It does not seem to occur everywhere.
Please refer support to ID [PPS-13913].
 
@lj-shadow, @TomBoB could one or both of you please open a ticket with Plesk support and allow SSH access to your server(s) for support? Developers would like to check the issue directly on the server, because we were yet not able to determine the cause. It does not seem to occur everywhere.
Please refer support to ID [PPS-13913].

I am using a Hetzner Storage Box, same problem and quick fix.
Opened the ticket - Id #400924
Would be nice if you could submit tickets without a subscription, at least if you try to help on a bug.
 
You can get support from Plesk directly, even if your license is from Strato.

It's a subscription, but it's free for the first month:
 
Developers were able to reproduce the issue under the condition that "Passive Mode" is turned off. The product issue has been confirmed and will now be tracked with ID [PPPM-13895], [PPP-60041]. According to developers, the workaround when using FTPS is to also check "passive mode".
 
Last edited:
Back
Top