It's probably some race condition, so it depends on what happens first in the initialization.
Your suggestion that this might be bug #470776, doesn't seem to be correct because when I run "mount -a -v", it claims that the NFS mount has already happened: > mount.nfs4: mount(2): Device or resource busy > mount.nfs4: /export/big is busy or already mounted So, mountall left it in a state with NFS mounted but without the bind mounts. I've worked around the bug by editing /etc/init/?mount?nfs? and adding a call to mount -a -v after the call to mountall. (Sorry for the lack of detail - that machine is booted into XP at the moment.) Why use mountall, anway? What does it do that "mount -a" doesn't? -- upstart/mountall doesn't handle bind mounts reliably (cannot reproduce) https://bugs.launchpad.net/bugs/459642 You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
