• 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

1.2 Gb Backup job to FTP repository causes /tmp to be 100% utilized

JasperV

JasperV

Basic Pleskian
Hello,

After the upgrade to 12.0.18 #7, after a 1.2Gb backup job to the FTP-repository I received the message:
ERROR: Zend_Db_Statement_Exception: SQLSTATE[HY000]: General error: 1030 Got error 28 from storage engine (Pdo.php:234)
Which means "Not enough diskspace". Running a df -m shows that the /tmp volume is 100% in use. As a result there is no backup file in the FTP-repository.

The backup causing this is 1.2Gb in size. /tmp is 500Mb in size.

Before the upgrade this backup tasks was performed daily without any such problem.

My FTP repository has enough disc space and is reachable. Smaller backups (up to 500Mb) to the FTP repository succeed without any error.

Please advice on how to solve the problem.
 
I have read both kb articles before submitting the issue. Somehow they do not seem to apply here. Again, all was working fine running Plesk Panel 11.5.
 
We are having this exact issue. Since upgrading to Plesk 12 the backup is taking the server offline as the /tmp folder is filling up.

Having read the above articles they both referecne the following folder /usr/local/psa/PMM/tmp however in our case it is the /tmp/ folder at the root that is filling up.

The backups have worked correctly until the upgrade to 12, we already resized /usr/local/psa/PMM/tmp to 40Gb to account for this many years ago. We have no space left to resize the tmp folder.

Why ii back up suddenly using /tmp/ instead of the original ]/usr/local/psa/PMM/tmp ?? and will this be fixed / put back to how it used o work ?

Thanks in advance, with no backups :-(

R
 
PRODUCT, VERSION, OPERATING SYSTEM, ARCHITECTURE
12.0.18 Update #8 CentOS 6.4 (Final) 64bit

PROBLEM DESCRIPTION

Server is knocked offline when /tmp/ folder is filled up with a file called repo_transport_tmp. This happens after the zip phase

STEPS TO REPRODUCE

Run an Persona lFTP repository backup

ACTUAL RESULT

Server goes offline until tmp folder is emptied, no backup is created

EXPECTED RESULT

Server backs up

ANY ADDITIONAL INFORMATION

The log file is over 1Gb in size and is crashing PC's we are tryng to open it in, so we cannot provide any log file data ! We wil continue trying to access it through other means !
 
Guys, we are working on this issue. Bugreport PPP-11008 is created, for your reference.
Now as possible workaround you may use following steps:

1. change DUMP_TMP_D in /etc/psa/psa.conf to /usr/local/psa/PMM/tmp
2. change /etc/sw-cp-server/conf.d/plesk.inc as described in http://kb.parallels.com/en/120242
 
Hi Igor,

Thanks for the update, we have only upgraded to 12 on a test server so will wait for the fix to be rolled out before apply 12 to our production servers. Could you advise when the fix will appear in a Micro Update ?

Is there any way we can watch progress on bug reports ?


Cheers
R
 
I'm wondering if Parallels Plesk 12.0 is a Beta/Alpha/Preview/etc version or is the Release version...
 
IgorG said:
Guys, we are working on this issue. Bugreport PPP-11008 is created, for your reference.
Now as possible workaround you may use following steps:

1. change DUMP_TMP_D in /etc/psa/psa.conf to /usr/local/psa/PMM/tmp
2. change /etc/sw-cp-server/conf.d/plesk.inc as described in http://kb.parallels.com/en/120242
Click to expand...

/etc/sw-cp-server/conf.d/plesk.inc is missing on the server I have. Can I skip the adjustments to this file?
 
You must log in or register to reply here.

Similar threads

P
Lost Storage Space After Failed Backup Centos6
Replies
2
Views
778
Peter_R
P
Back
Top