I wouldn't say I'm likely to get this to progress quickly; yes, I wrote the NFS handling code, but soon came to recognize there were bugs in the resulting logic that I couldn't see a way to address with the current upstart support.
That said, I don't see the reason for this runlevel check. Is this meant to split the portmap start condition (start when portmap is restarted) away from the startup condition (start when mountall mounts an NFS mount)? If so, what happens when there are no NFS shares mounted at boot time, and the user calls mount manually post-boot in which case no 'mounting' event is emitted?
If that's not the intent of your check, then I definitely don't understand what you're trying to achieve here.
I wouldn't say I'm likely to get this to progress quickly; yes, I wrote the NFS handling code, but soon came to recognize there were bugs in the resulting logic that I couldn't see a way to address with the current upstart support.
That said, I don't see the reason for this runlevel check. Is this meant to split the portmap start condition (start when portmap is restarted) away from the startup condition (start when mountall mounts an NFS mount)? If so, what happens when there are no NFS shares mounted at boot time, and the user calls mount manually post-boot in which case no 'mounting' event is emitted?
If that's not the intent of your check, then I definitely don't understand what you're trying to achieve here.