nfs shares not unmounted properly

Asked by Maarten Jacobs

I've been having an issue on one of my machines (a Dell Latitude 830) where the shutdown process isn't working properly and essentially hangs up on restart or shutdown.

I have been trying for a couple days now to try and figure this thing out. I thought perhaps it was related to the following bug:

https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/869635

But not 100% sure.

In essence, when I shutdown the machine I get a number of messages indicating that nfs.mount fails to unmount my nfs shares as it cannot resolve the hostname of the machine they're mounted to. I'd be happy to post the exact error messages if somebody can explain how to capture them.

Anyways it appears to me that the network (or some element of it) is torn down before the system gets to cleanly unmount the nfs shares... Which probably causes the hang a little later in the process.

What I'd like to get some help with is how to figure out in what order the scripts in /etc/init and /etc/rc6.d are called. I'm pretty sure that somehow it's not getting done in the right order, but without being able to figure out what the order actually is that's hard to ascertain.

Anybody out there who can give me some pointers on how to figure that out?

(Incidentally if anybody knows how to fix my problem I'm happy to listen too - but in the first instance I'm looking for some help on debugging this).

Thanks,

Question information

Language:
English Edit question
Status:
Expired
For:
Ubuntu util-linux Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Maarten Jacobs (maarten256) said :
#1

Nobody?

Revision history for this message
Maarten Jacobs (maarten256) said :
#2

Still nobody??

Revision history for this message
Launchpad Janitor (janitor) said :
#3

This question was expired because it remained in the 'Open' state without activity for the last 15 days.

Revision history for this message
delance (olivier-delance) said :
#4

The question is too technical for me. I see you already subscribed the bug.

Revision history for this message
Maarten Jacobs (maarten256) said :
#5

For future reference, the underlying issue is described here, with a workaround:

https://bugs.launchpad.net/ubuntu/+source/netbase/+bug/903825

Revision history for this message
Andrew McMillan (karora) said :
#6

The scripts are called in the order of K00 -> K99 with the 'stop' parameter followed by S00 -> S99 with the 'start' parameter. Hence (as described in #903825) S20sendsigs is being called before S31umountnfs.sh. The fact that the script for *unmounting* NFS drives is called with the 'start' parameter is definitely a curiousity, and I expect there's some long-standing reason for that which might need re-examining, however it is also followed by S40umountfs and S60umountroot, so I guess that's the way it's done...