On Thu, 2010-04-08 at 23:14 +0200, Karel Zak wrote: > On Thu, Apr 08, 2010 at 02:09:37PM -0400, Adam Jackson wrote: > > Largely functionally equivalent to chroot(1), but creates a new > > filesystem namespace so the chroot'd process can't escape. It's > > possible to do this with unshare/mount/chroot in a shell script, but > > it's a bit tricky to get right. > > Interesting idea. > > What about to add this functionality to the chroot(1)? For example > add a new option "--ns" to move the process to a separate namespace > with a real new root directory. That'd be fine too. - ajax
Attachment:
signature.asc
Description: This is a digitally signed message part
