Resolved Complete Backup FTP with letsencypt error

[Dukemaster]

Hello at all,
I made a complete backup with backup manager on my FTP-storage:
(Plesk Onyx 7.0.17, Ubuntu 14.0.4 LTS over 1and1)

I got the following notifcation about a successfully created backup, but with a lil unimportant error

Das Backup 30. Nov. 2016 10:48:36 wurde erstellt und kann wiederhergestellt werden, jedoch traten geringfügige Probleme auf.

Details zur Backup-Aufgabe 30. Nov. 2016 10:48:36
Warning: server "server"
Unable to upload backup files to FTP storage: ack Broken pipe: write( 13, 'backup_ext_letsencrypt_1611301048.tgz ' ) at /opt/psa/PMM/agents/shared/IPC/Run/IO.pm line 558. STDOUT:
Warning: server "server"
Unable to upload archive backup_ext_letsencrypt_1611301048.tgz to FTP storage: ack Broken pipe: write( 13, 'backup_ext_letsencrypt_1611301048.tgz ' ) at /opt/psa/PMM/agents/shared/IPC/Run/IO.pm line 558.

Does anyone knows how to fix it or what to do now?
Greets

 

[UFHH01]

Hi Dukemaster,

this issue ( #PPPM-5234 ) should be fixed in Plesk Onyx, with the patches from "MU#8", according to => https://kb.plesk.com/130278

Pls. consider to use the ssh - command ( logged in as user "root" ):

plesk installer --select-release-current --reinstall-patch --upgrade-installed-components

Additional patch - files ( depending to your operating system ) have been added to the above KB - article, for manual fixes.

 

[Dukemaster]

Thanks a lot, @UFHH01 your command worked perfectly, Plesk was updated. The last days I got errors by updating over Updates/Upgrades panel or section in Plesk Onyx (by website).
Now I have MU#8.
Never ever was the support of Plesk so great like in present.

Later I started a complete backup over FTP on another server/host which failed with a new error:
Log:

ERROR:      ()      Runtime error: pmm-ras command completed with non-zero exit code at /opt/psa/PMM/agents/shared/Storage/ArchiveStorage.pm line 120.

I was a little afraid in doing the workaround, after the successful update. Don't know if this is the reason why now all complete backups don't work.
Never mind...
happy, that's what I do... thanks and honors to Plesk
Greets

 

[UFHH01]

Hi Dukemaster,

I was a little afraid in doing the workaround, after the successful update.

No, there is no need to do a manual workaround, when you updated successfully to MU#8.

ERROR: () Runtime error: pmm-ras command completed with non-zero exit code at /opt/psa/PMM/agents/shared/Storage/ArchiveStorage.pm line 120.

Pls. consider to add MORE of your depending log for further investigations.


In addition, pls. be informed, that there is a configuration file called "/usr/local/psa/admin/share/pmmcli/pmmcli-rc", where you are able to set several options, as for example:

Change:

...
# Add debug information to all utilities called by pmmcli
FORCE_DEBUG_LOG 0
...

to:

...
# Add debug information to all utilities called by pmmcli
FORCE_DEBUG_LOG 1
...

So that your log - level will be increased for your next backups. Pls. repeat one or two backup - processes and consider to send the complete logs to @Sergey Prohorov ( within a private conversation/message ), asking him, if he finds something suspicious or unusual here, which could be repaired.
Afterwards, pls. revert your changes for the option "FORCE_DEBUG_LOG" back again to "0" at "/usr/local/psa/admin/share/pmmcli/pmmcli-rc", pls.​

 

[Dukemaster]

I did what you told me, @UFHH01. Then two backups I made. One question in the beginning, would it be necessary to start one service or the whole server again, after the change from "0" to "1"?
In the log is only the same error like all the time, except there is another location of more detailed log files which I don't know?
task-4.log task-5.log and task-6.log:

ERROR: () Runtime error: pmm-ras command completed with non-zero exit code at /opt/psa/PMM/agents/shared/Storage/ArchiveStorage.pm line 120.

Don't know to send only this sentence to @Sergey Prohorov?
Greets

 

[UFHH01]

Hi Dukemaster,

now that you added the "@" in front of the forum - username, he will be informed about the "mentioned-me" - function at this forum and might have look at the thread. If you would like to communicate with a forum - username over a private conversation, pls. go to his/her INFORMATION - page and click on the link "Start a conversation".

One question in the beginning, would it be necessary to start one service or the whole server again, after the change from "0" to "1"?

I suggested the DEBUG - log - level, in order to get a more verbose output in your log - file, for investigation reasons. Developers/Coders ( or/and Plesk - Team - Member ) ( - or persons willing to help you with the investigations ) are able to see possible misconfigurations with such a log! If you insist of keeping such a log to yourself, instead of following the suggestion to send the file to a mentioned person, I really can't help you here.
You would certainly create all sorts of possible backup - options, when using the DEBUG - level, so that a developer/coder ( or/and Plesk Team - Member ) is able to compare the processes made on your server.

For better reproduction I made a few screens of my Plesk backup configuration in Plesk panel. Hope it helps to find out what's wrong and happening.

Well, such screens are nice and somehow informative, but log - files are far better here.

task-4.log task-5.log and task-6.log:

You only included one line here from your logs, because you think, that only warnings and errors point to possible root causes, which is absolutely not the case. Pls. consider to let the developer/coder ( or/and Plesk - Team - Member ) ( - or persons willing to help you with the investigations ) decide, which part of the demanded log - file will help to investigate the issue. This is FAR easier, than trying to explain in x-posts why this and that information is usefull and may essential to investigate the issue/error/problem.

 

[Dukemaster]

Thanks for the information about tagging user. I know this by my own com.
Sorry, you misunderstood me. There was only this one sentence in each log file. And I asked you if it's necessary to reboot the server after the change to debug level.
better to have had this info.
That was the reason why I got only this one error-sentence in three backups. Nothing more of information - Igave you everything out of each log file.
Later I changed the value back to "0".
Now I decided to make your suggested change again and rebooted the server and it really gives a little more information.
That's the new backup error log which I've sent to Sergey.
Lots of greets

Error:
Runtime error: ack Broken pipe: write( 12, 'backup_ext_letsencrypt_1612021936.tgz ' ) at /opt/psa/PMM/agents/shared/IPC/Run/IO.pm line 558.
Warning: server "server"
Server settings are skipped from backup due to error: ack Broken pipe: write( 12, 'backup_ext_letsencrypt_1612021936.tgz ' ) at /opt/psa/PMM/agents/shared/IPC/Run/IO.pm line 558.
Warning: server "server"
Unable to upload archive backup_ext_letsencrypt_1612021936.tgz to FTP storage: ack Broken pipe: write( 12, 'backup_ext_letsencrypt_1612021936.tgz ' ) at /opt/psa/PMM/agents/shared/IPC/Run/IO.pm line 558.

 

[Dukemaster]

*EDIT* @UFHH01 and @Sergey Prohorov
Don't really know why, but now a "Complete FTP Backup" was successfully created".
- First I deleted all old backups except the last three from server and FTP storages over Plesk Onyx.
- Increased the value of disk space from 6000MB to 10.000MB (default is 20MB) in backup settings (Plesk Onyx Panel)
- Then I did in case of normal updating processes:

# apt-get clean
# apt-get update
# apt-get upgrade

Got the information (not really important, it's more for older systems):

The link /vmlinuz.old is a damaged link
Removing symbolic link vmlinuz.old
you may need to re-run your boot loader[grub]
The link /initrd.img.old is a damaged link
Removing symbolic link initrd.img.old
you may need to re-run your boot loader[grub]

- Left the message without any action.
- Just

# reboot

- Started "Complete FTP-Backup" to other FTP-server
Backup successfully created
- Changed the

"FORCE_DEBUG_LOG" back again to "0" at "/usr/local/psa/admin/share/pmmcli/pmmcli-rc"

and rebooted server for the last time.

Thank for the great help from all of You! Happy weekend!

 

[VITEX]

Dear all,

I have been following this post since I am also struggling with backups since upgrading to Plesk Onyx. pmm-ras Hotfix has been applied to a CentOS server as well as updating to patch #9.
However, there are still issues on scheduled backups. I initially thought this is related to the Let´s Encrypt certificate used in the customers webspace - but no. To isolate the issue I created a new customer, installed Wordpress and Let´s Encrypt and did choose the same FTP Backup repository - guess what.... this backup works.

The customer webspace failing backups is running pimcore CMS and the backup tarball of the failed backup does only contain the database portion. The tarball itself is corrupted and about 6MB in size.

I am wondering if this is a database permission related issue in combination with pimcore.

How can I isolate the issue ? Which Database User is used to dump the customer databases using pmm-ras ?

Best regards
VITEX