[Nagiosplug-help] stale NFS handles detection

[Matthias Eble]

Rick Garland schrieb:
> Hi all
> 
> Working with Nagios 2.9 on CentOS5 and plugins 1.4.9 on RHEL4 systems.
> 
> Have had problems with stale NFS. I see lots google posts and other 
> scripts but they don’t seem to fit the bill.
> 
> Example, on the RH systems it will not say “stale” in the df output.
> 
>  
> 
> Looking for feedback. What do you use to detect stale NFS.

Check_disk has been detecting stales for long time, but dependent on the 
argument selection, though. However, this has been fixed in 1.4.10.
You'll find some entries in the NEWS file in the plugins tarball.

Here's what NEWS says:

Check_disk now calls stat() on all filesystems to check.
   (Old: only the ones selected using -p)
   A meaningful error message (eg "Stale NFS Handle") is printed if stat 
fails.

New check_disk option -L: Only check local filesystems, but call stat() 
on remote ones, too.
   Thus accessibility of remote filesystems can be checked without any 
threshold comparison.

So -L doesn't check free space on remote filesystems.

hth
Matthias