Thanks for the fast answer. You're right, this isn't a problem of mount. Maybe it would be helpful to add a hint in the list of mounted device (output of mount), when the mount is not "active" anymore (because of a moved file). It's the special combination of tools I use which caused this trouble. The "mount"-configuration is part of the named-chroot package in Redhat/CentOS. Webmin as administration tool works like vi (create tmp-files and move them). If you add a new zone via Webmin, only /etc/named.conf will be updated, not /var/named/chroot/etc/named.conf. If you then restart the service, it works, because the chroot environment will be destroyed and recreated. But a "rndc reload" doesn't work anymore because it uses the unchanged /var/named/chroot/etc/named.conf. When you aware of the problem, it's no problem to work around. Since /etc/named is also a mount in chroot, the zone configuration can be separated in a file in this directory and then webmin works fine. Thanks for your help, Rudi! >On Thursday 10 December 2015, Yoss@xxxxxxx wrote: > >> Additional notes: >> The problem seem to be more complex, because if a 'echo "test" >> >> /etc/testa' is used, it works. But after a change with vi, echo >> doesn't works too. Practical background is the use of mount --bind in >> a bind (named) chroot environment. A change of /etc/named.conf will >> not result in an update of /var/named/chroot/etc/named.conf. > >Why don't you use a hardlink? >$ cp -al /etc/named.conf /var/named/chroot/etc/named.conf > >Then vim will automatically change it's behavior and saves changes >in-place. > >cu, >Rudi -- To unsubscribe from this list: send the line "unsubscribe util-linux" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
