• We value your experience with Plesk during 2024
    Plesk strives to perform even better in 2025. To help us improve further, please answer a few questions about your experience with Plesk Obsidian 2024.
    Please take this short survey:

    https://pt-research.typeform.com/to/AmZvSXkx
  • The Horde webmail has been deprecated. Its complete removal is scheduled for April 2025. For details and recommended actions, see the Feature and Deprecation Plan.
  • We’re working on enhancing the Monitoring feature in Plesk, and we could really use your expertise! If you’re open to sharing your experiences with server and website monitoring or providing feedback, we’d love to have a one-hour online meeting with you.

[PPP-26478] After update to onyx backup does not work

Hi TomBoB,

as you already know, @Sergey Prohorov watches this thread and will reply here, maybe requesting additional informations from you... so stay tuned on Monday morning ( russian time :p ).
 
Hi, all. I've checked the content of MU#9. It contains all changes from KB 130278. Our test environment does not catch the issue on this updates. So I need your help. If you still have the issue could you please set "FORCE_DEBUG_LOG" to "1" at "/usr/local/psa/admin/share/pmmcli/pmmcli-rc" and send me PM with log from the latest directory '/var/log/plesk/PMM/backup-...'. If you already re-applied custom fix from the KB you could download the original pmm-ras from the following links:
for Ubuntu 14.04 http://autoinstall.plesk.com/PSA_17.0.17/microupdates/MU9/dist-deb-Ubuntu-14.04-x86_64/pmm-ras
for CentOS 7 http://autoinstall.plesk.com/PSA_17.0.17/microupdates/MU9/dist-rpm-CentOS-7-x86_64/pmm-ras
 
I did the Plesk to Onyx update as well and not only are back ups not working but ftp is not working. I did run Plesk Repair FTP which found nothing wrong..

I swapped the pmm-ras file and fixed permissions but not working still.

I set the debug to "1" but I do not have the log in that directory to send.

Thanks for the assistance!
 
Last edited:
Can confirm this bug.
OS‪Ubuntu 16.04.1 LTS‬ProductPlesk Onyx
Version 17.0.17 Update #12, last updated on Dec 30, 2016 04:37 PM

Also the backup page (/admin/backup) takes ages to load.

root_Dezember31_1483139201_CWBackup_Manager_-_Plesk_Onyx_17.0.17_-_Google_Chrom.jpg


I am seriously regretting the decision to Upgrade to Onyx. As a stable release.. shouldn't happen. Downgrade is unfortunately impossible.

Edit: Yes, I have done: plesk installer --select-release-current --reinstall-patch --upgrade-installed-components just in case someone wants to ask.
 
I'm experiencing the same problem: We upgraded to Onyx because backup with 12.5 didn't work anymore. In Onyx backup didn't work either. Found out to increase timeouts on ftp backup server as backups are between 70-90GB and enable rest command. Did this on proftpd.conf, restarted and deleted all old backups.

The changes to proftpd.conf were:
TimeoutNoTransfer 14400
TimeoutStalled 14400
TimeoutIdle 14400
AllowStoreRestart on

Since then backup worked for 5 days for 3 of the 4 servers. The fourth server still backs up - his backup is about 1.8GB only. Since four days backup doesn't work anymore for 3 of the 4 servers. Found out that we ran out of harddisk space on the ftp server. But the errors in logs have been the same as before. Maybe clearer messages in the logs would be helpful ;-)
 
On Plesk Onyx Version 17.0.17 Update #12 this happens (again). I have applied the manual pmm-ras fix as it helped in the past and will report back.
This thread should NOT be tagged as "resolved".
 
i´m in
we just installed a new server with Plesk Onyx Version 17.0.17 Update #12 on ‪Debian 8.7‬ and get the same error when using pleskbackup with output-file on ftp

Code:
/usr/local/psa/bin/pleskbackup domains-name domain.com --output-file=ftp://user:pass@ip-adress/path/to/backup/

Code:
error: Runtime error: ack Broken pipe: write( 14, 'backup_user-data_1701190954.tgz
' ) at /opt/psa/PMM/agents/shared/IPC/Run/IO.pm line 558.

Runtime error: The backup failed with errors!

Yes, I have done: plesk installer --select-release-current --reinstall-patch --upgrade-installed-components
Any suggestions?
 
Same problem here.
I made this https://support.plesk.com/hc/en-us/...d-backup-files-to-FTP-storage-get-broken-pipe some days ago and worked once, only once, but now doesn't work even if I repeat the steps.
I have "Versión 17.0.17 Actualización #12".

The error is always the same with 3 variables:
- Unable to upload archive clients/clientfolder/domains/domain.com/backup_user-data_1701180313.tgz to FTP storage: ack Broken pipe: write( 13, 'clients/clientforlder/domains/domain.com/backup_user-data_170118xxxx.tgz
' ) at /opt/psa/PMM/agents/shared/IPC/Run/IO.pm line 558.
- Database dbname, type mysql is skipped from backup due to error: ack Broken pipe: write( 13, 'clients/clientfolder/domains/domain.com/backup_user-data_170118xxxx.tgz
' ) at /opt/psa/PMM/agents/shared/IPC/Run/IO.pm line 558.
- Hosting on domain domain.com is skipped from backup due to error: ack Broken pipe: write( 13, 'clients/clientfolder/domains/domain.com/backup_user-data_170118xxxx.tgz
' ) at /opt/psa/PMM/agents/shared/IPC/Run/IO.pm line 558.

Thanks for your support.
 
Hey thanks for your fast suggestion.
but we are not working with cronjobs. the command
Code:
/usr/local/psa/bin/pleskbackup domains-name domain.com --output-file=ftp://user:pass@ip-adress/path/to/backup/
was fired manually and finished with the given error


edit: we are using plesk 12.5 on 8 other debian servers and dont have any problems with backing up clients to ftp folder. this one 17.0 server should backup on the same ftp space, so this cant be the problem here
 
Last edited:
Hi,

Exact same error message. Was in contact with Plesk. Let me quote from the PM.
Hi, TomBob. Thank you for help. I've found 2 issues in backup.log:
1. It seems that the ProFTPd at FTP server has TimeoutSession=50m it means all part of backup must be uploaded during this time.
and
2. When backup process catches the issue above it try to reconnect to the FTP server and resume upload but the server configuration does not allow to use APPE and REST FTP commands so we cannout resume upload and stop backup:

leads to
If you have root access to the FTP server you could allow APPE/REST commands (preferred way) or increase TimeoutIdle value (it fixes only full backup creation but incremental backup to FTP and selective restore still will not work).

On the Plesk side, I will submit 2 bugs:
1. Improve FTP storage diagnostic to check that FTP server allow to use APPE/REST command and warn user if it does not support them.
2. Add a new parameter to panel.ini which will force backup process to open new FTP connection per backup volume. It will allow users to workaround small value of 'TimeoutIdle' by decreasing split size value.

When I asked if something changed in the backup process between 12.5 and Onyx (as 12.5 backs up perfectly) he answered:
When you run server backup to FTP Plesk 12.5 creates a full backup (web, mail, DB archives for all subscription) in local storage and then it send them to the FTP storage. Plesk Onyx consumes less disk space because does not create full backup in local storage - it sends each mail, web or DB archive to the FTP storage as soon as it created. It leads to following changes in interaction with FTP server:
1. Plesk 12.5 uses only RETR command to get a backup index from FTP but Plesk Onyx uses pair REST / RETR commands to get it.
2. Plesk 12.5 opens a connection to the FTP server when a full backup is ready but Plesk Onyx opens it as soon as first part of content ready. So Onyx has to keep the connection alive longer than 12.5 to avoid upload interruption by timeout we try to reconnect and resume download by APPE command.

Think that sheds some light on the issue.

Cheers,
Tom
 
Edit: Please ignore this post
at first it doesnt work, but that was because of a mistake of mine

ftp backups works fine with update #13
thanks for your help
 
Last edited:
Back
Top