I am facing problems assigning the owner and group to an ext3 filesystem being mounted from an external storage. I created a mount point /vol and changed its owner and group to alpha and beta respectively (as required). The ext3 filesystem (LV from the external storage) was then mounted on /vol using the mount command. However, after mounting the filesystem the owner and group of /vol changed back to root.
How do I ensure that the owner and group for this (and other) external filesystem to be mounted dynamically are preserved as required ? The mount options uid and gid don't seem to work for ext3 filesystems. Are there any other options for this purpose?
Thanks,
Manish
Manish Kathuria writes:
I am facing problems assigning the owner and group to an ext3 filesystem being mounted from an external storage. I created a mount point /vol and changed its owner and group to alpha and beta respectively (as required). The ext3 filesystem (LV from the external storage) was then mounted on /vol using the mount command. However, after mounting the filesystem the owner and group of /vol changed back to root.
How do I ensure that the owner and group for this (and other) external filesystem to be mounted dynamically are preserved as required ? The mount options uid and gid don't seem to work for ext3 filesystems. Are there any other options for this purpose?
Try to run chown also after you have mounted the filesystem. That should solve it.
-- Nux! www.nux.ro
At Sun, 16 Jan 2011 14:20:40 +0530 CentOS mailing list centos@centos.org wrote:
I am facing problems assigning the owner and group to an ext3 filesystem being mounted from an external storage. I created a mount point /vol and changed its owner and group to alpha and beta respectively (as required). The ext3 filesystem (LV from the external storage) was then mounted on /vol using the mount command. However, after mounting the filesystem the owner and group of /vol changed back to root.
How do I ensure that the owner and group for this (and other) external filesystem to be mounted dynamically are preserved as required ? The mount options uid and gid don't seem to work for ext3 filesystems. Are there any other options for this purpose?
The ownership of the mount point is irelevant and has no effect on the ownership of the mounted file system.
You need to change the ownership of the *file system* root directory itself.
Mount the ext3 filesystem and *then* change the owner and group. Of course, whatever you set it to will be whatever it is on any other Linux system the volumn is mounted on.
Thanks,
Manish _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On Sun, Jan 16, 2011 at 8:25 AM, Robert Heller heller@deepsoft.com wrote:
At Sun, 16 Jan 2011 14:20:40 +0530 CentOS mailing list centos@centos.org wrote:
How do I ensure that the owner and group for this (and other) external filesystem to be mounted dynamically are preserved as required ? The mount options uid and gid don't seem to work for ext3 filesystems. Are there any other options for this purpose?
The ownership of the mount point is irelevant and has no effect on the ownership of the mounted file system.
You need to change the ownership of the *file system* root directory itself.
And you need to have the privileges to do so. If it's an NFS mount, and the NFS server has not included "no_root_squash" in the settings of /etc/exports, you're not going to be able to change this from an NFS client. The "root_squash" setting is enabled by default. You need collaboration with the owner of the upstream filesystem to export it appropriately to change this. The owner of the exported file system should set this before exporting it.
Otherwise, anyone with local root privileges could change the ownership of any casually exported NFS filesystem, and that would be...... chaotic.
Mount the ext3 filesystem and *then* change the owner and group. Of course, whatever you set it to will be whatever it is on any other Linux system the volumn is mounted on.
See above. The "root_squash" setting is the default and prevents this for NFS exports.
-
Manish Kathuria -
Nico Kadel-Garcia -
nux＠li.nux.ro -
Robert Heller