VFS Quota project

Francois Tigeot ftigeot at wolfpond.org
Thu Aug 18 08:20:14 PDT 2011


On Thu, Aug 18, 2011 at 10:57:37AM -0400, Justin Sherrill wrote:
> Are the PFSs in /pfs/ unusable for this quota work?

With the current state of thing, yes.

> The null mounts
> are used to place them in the normal file hierarchy, but on my
> machine, /var is the null mount to /pfs/@@-1:00001, or /pfs/var,
> depending on how you look at it. so I assume quotas could go against
> /pfs/whatever.

Quota initialization is correctly done against the right mount point:

  vfs_quota: enabling accounting for /boot (ufs)
  vfs_quota: enabling accounting for /tmp (tmpfs)
  vfs_quota: enabling accounting for /home (nullfs, /pfs/@@-1:00001)

but the nullfs mount code uses a hack to replace it's mount point information
with the one of the underlying non-nullfs filesystem

In the above example, /pfs/@@-1:00001 is a symlink on the root volome. So,
once the real write operations are done, they are all counted against /

> I may be totally misunderstanding this.

I'm afraid not :-(

So far, I've not found a way to obtain the _real_ mount point information
when a write operation is done on a nullfs volume.

-- 
Francois Tigeot





More information about the Kernel mailing list