Article ID: 115344, created on Dec 24, 2012, last review on Jan 11, 2015

  • Applies to:
  • Operations Automation 5.5
  • Operations Automation 5.4

Symptoms

A customer cannot manage FTP users in their Linux Shared Hosting subscription in POA CCP (create, remove, change password and so on) - corresponding POA tasks fail with error message 'Disk quota exceeded (errno 122)' like in the example below:

Task ID    100000
Queue name    ftp_vhost_3
Task name    Change user 'f100001.u1' for virtual host (vhost_id=1) on ProFTPD service with id 10
Task description    Change user 'f100001.u1' for virtual host (vhost_id=1) on ProFTPD service with id 10
Queue status    Failed    
Start not earlier than    Dec-10-2010 10:10
Method name    taskChangeUser on OBJREF:ProFTPD:0:getService:10
Last execution output    Destination host_id #10 : lchown /usr/local/pem/vhosts/100001: Disk quota exceeded (errno 122).

Cause

The customer exceeded disk quota in their webspace. It may be checked using the following command on the Linux Shared Hosting server where customer's webspace is created(in case of NG hosting repquota should be executed on NFS server in case it returns empty results on the webserver):

# repquota -a | grep WEBSPACE_ID

where WEBSPACE_ID is the ID of the customer's webspace in question, it may be found in POA Customer or Provider CP.

During any operation on webspace (including FTP users management) POA is trying to set correct ownership on customer's webspace: set the Linux system user created for customer's webspace as the owner of webspace. Since the customer exceeded disk quota for their webspace the operation to change webspace ownership fails with the error message 'Disk quota exceeded (errno 122)'.

Resolution

In general, exceeded disk quota should not affect FTP user management in POA CCP, customer just must not be able to upload more data in their webspace than disk quota allows.

So, in order to work the problem around one may temporarily increase disk quota for the problem system user using the Linux edquota utility. Should be executed on NFS server in case of NG, as well as repquota. The following solution should be applied on a physical server or a VM. If the hosting server in question is a Parallels Virtuozzo container, see the solution at the end of the article.

First, find the current disk quota and usage for the system Linux user using the command repquota mentioned above:

# repquota -a | grep WEBSPACE_ID

where WEBSPACE_ID is the ID of the customer's webspace in question. Then, set the quota slightly above the current disk usage:

# edquota WEBSPACE_USER

where WEBSPACE_USER is the system Lunux user created for the customer's webspace. The name of the system user may be found in the /etc/passwd file on the proper Linux Shared Hosting server:

# grep WEBSPACE_ID /etc/passwd | awk -F: '{print $1}'

where WEBSPACE_ID is the ID of the customer's webspace in question, it may be found in POA Customer or Provider CP.

If the hosting server in question is a Parallels Virtuozzo container, the following solution should be used:

  1. Increase the disk limit for the container:

    vzctl set CTID --save --diskspace LIMIT
    
  2. Re-initialize the disk quota:

    vzquota drop CTID
    vzctl quotainit CTID
    
  3. Start the container:

    vzctl start CTID
    

Search Words

Disk quota exceeded

NG file manager doesnt work

caea8340e2d186a540518d08602aa065 5356b422f65bdad1c3e9edca5d74a1ae e12cea1d47a3125d335d68e6d4e15e07 5b048d9bddf8048a00aba7e0bdadef37 2554725ed606193dd9bbce21365bed4e ac82ce33439a9c1feec4ff4f2f638899

Email subscription for changes to this article
Save as PDF