Re: NFS ACL strangeness

On Thu, Nov 18, 2004 at 12:03:44AM +0100, Andreas Gruenbacher wrote:

> This problem is difficult to solve: The client could try to retrieve the 
> default ACL of the parent directory and only apply the umask if there is 
> none, but this would result in a race (and require an additional round-trip 
> as well). A default ACL could be created or removed on the parent directory 
> after the client queried the default ACL, but before it could create the 
> file. The result could be too permissive permissions.

Does this effectively mean that the behaviour is a known bug?

Is there a workaround for the given Problem:

Restictive umask (0066) by default and a less restrictive one (0022) on
some specific directories?

Sven

-- 
"And I'm right.  I'm always right, but in this case I'm just a bit more
right than I usually am." (Linus Torvalds, Sunday Aug 27, 2000)

/me is giggls@ircnet, http://sven.gegg.us/ on the Web

_______________________________________________
acl-devel mailing list
[email protected]
http://acl.bestbits.at/mailman/listinfo/acl-devel