[Gluster-devel] ACL-related glitch in latest TLAs?

[Anand Avati]

Brent,
Most likely this seems to be caused by a server crash. Can you confirm if
glusterfsd died while this operation was under way ('Invalid argument' was
wrongly returned which is now again fixed to 'Transport endpoint not
connected'). 'preserving ACL' internally means a setxattr() call. A setxattr
related bug was fixed in the server earlier today which likely caused the
server death, and hence 'invalid argument' reply. Now both should be fixed.

thanks,
avati

2007/7/1, Brent A Nelson <brent at phys.ufl.edu>:
>
> When doing a cp -a from a directory in a glusterfs to a different
> directory of the same glusterfs, I get messages such as the following
> (presumably for every file copied):
>
> cp: `usr/include/linux/atmppp.h': Invalid argument
> cp: preserving ACL for `usr0/include/linux/atmppp.h': Invalid argument
>
> This did not occur when copying the original directory from a local ext3
> to the glusterfs.
>
> Is GlusterFS claiming to have ACL support when it doesn't? Or does it now
> have some support, and I should mount the underlying ext3 volumes with the
> acl option? Or...?
>
> Thanks,
>
> Brent
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>



-- 
Anand V. Avati