[Explained] Unending Restore Problems: Not all specified mountpoints are using quota

Not Solved, but Explained:
My VPS host (Rose Hosting) uses virtualization technology from Linux-VServer. Apparently that virtualization layer does not permit quotas, which explains the restore problems described below.

I am having a heck of a time doing a Virtual Server Restore to Virtualmin 3.73 GPL. The virtual servers were created on an unrestricted “Vanilla” CentOS 5.3 home server running Virtualmin 3.73 GPL.

Due to some problems with my residential cable connection, I have purchased a VPS from Rose Hosting. It came installed with CentOS 5.3 Webmin, and various and sundry other things. I downloaded the install.sh for Virtualmin, and (eventually) ran it such that I now have Virtualmin 3.73 GPL on it as well.

[Off Topic: Rose Hosting sets the /tmp mount point to be noexec by default, for “security” (?). They will change it if you ask, or what I did is I simply created a /temp directory, and modified the install.sh script to use /temp instead of /tmp…hope this didn’t break anything during the install? It appeared to run and install everything correctly.]

Once Virtualmin was installed, I SCP/transfered my tar.gz Virtualmin backups from my home server to my new VPS, and attempted to restore them.

First off, I tried to install, and got an error about no home directory (I think). The result was that some AWSTATS information was created, even though the virtual server was not created.

[For Future Reference: If this happens, you need to go back to Webmin - Servers - AWStats Reporting and delete any reference to the failed virtual server restore.]

I tried to restore again, and received an error about quotas: “setquota: Not all specified mountpoints are using quota.” I assume this once again has something to do with Rose Hostings default settings for VPS. The following is the output of “mount”:

/dev/hdv1 on / type ufs (defaults)
none on /proc type proc (defaults)
none on /dev/pts type devpts (gid=5,mode=620)

And this is my /etc/fstab:

LABEL=/ / ext3 grpquota,defaults,usrquota 0 1
none /dev/pts devpts gid=5,mode=620 0 0
none /dev/shm tmpfs defaults 0 0
none /proc proc defaults 0 0
none /sys sysfs defaults 0 0
/dev/sda1 swap swap defaults 0 0
/dev/cdrom /mnt/cdrom udf,iso9660 noauto,owner,kudzu,ro 0 0
/dev/fd0 /mnt/floppy auto noauto,owner,kudzu 0 0

Which appears to this untrained linux beginner’s eye to say that quotas are enabled on /. However, when I go to Webmin - System - Disk Quotas and try to “Enable Quotas” it continues to state, “User and Group Quotas Inactive” no matter how many times I press the “Enable Quotas” button.

Anyway, I decided to try the import without quotas (I deselected the “Administration user’s password, quota and Cron jobs” feature…I really just wanted to deselect quotas, but there doesn’t appear to be a way to do that.)

When I tried to restore again, apparently it created the user and group for the failed restore, so now I need to go back, delete (userdel and groupdel) those, delete the AWStats, and remember to deselect the “Quotas” stuff before attempting to restore…whereupon I receive the:

Restore failed : setquota: Mountpoint (or device) / not found. setquota: Not all specified mountpoints are using quota.

error again, so I presume that leaving out the “Administration user’s password, quota and Cron Jobs” doesn’t do what I want it to.

So…can anyone help me any or all of the below with:

  1. Why quotas appear to be enabled in fstab but aren’t according to Virtualmin/Webmin?
  2. If I need to have Rose Hosting change something in regards to quotas, are there any security implications to that?
  3. A better way to restore “everything” from one Virtualmin installation to another?
  4. A way to not have to go back and delete AWstats stuff, Users/Groups, etc. every time a restore fails?
  5. Anything else I’m doing wrong?!!

Thanks in advance for any assistance you can render.

  • Acorp