[hobbit] Missing partition in disk monitoring on Solaris 10
Dominique Frise
dominique.frise at unil.ch
Tue May 26 16:24:05 CEST 2009
Benoit Schmid wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hello,
>
> On a server running solaris 10, df gives:
>
> # df -k
> Filesystem size used avail capacity Mounted on
> / 0K 6.7G 38G 15% /
> /common 123G 1.6M 123G 1% /common
> /dev 45G 6.7G 38G 15% /dev
> /var 8.9G 1.8G 7.2G 20% /var
> proc 0K 0K 0K 0% /proc
> ctfs 0K 0K 0K 0% /system/contract
> mnttab 0K 0K 0K 0% /etc/mnttab
> objfs 0K 0K 0K 0% /system/object
> swap 17G 328K 17G 1% /etc/svc/volatile
> fd 0K 0K 0K 0% /dev/fd
> swap 17G 32K 17G 1% /tmp
> swap 17G 32K 17G 1% /var/run
>
What your are showing us is not the output of df -k but rather df -h.
> Unfortunately hobbit just shows:
> Filesystem kbytes used avail capacity Mounted on
> swap 17718392 328 17718064 1%
> /etc/svc/volatile
> swap 17718088 32 17718056 1% /tmp
> / 0 7029995 39920877 15% /
>
> Would you know why not all partitions are monitored?
>
From hobbitclient-sunos.sh you can see that Xymon uses following
commands to report disk usage
...
echo "[df]"
ROOTFSTYPE=`/bin/df -n / | awk '{print $3}'`
/bin/df -F $ROOTFSTYPE -k
# Then see what fs types are in use, and weed out those we dont want.
FSTYPES=`/bin/df -n -l|cut -d: -f2 | awk '{print $1}'|egrep -v
"^${ROOTFSTYPE}|^proc|^fd|^mntfs|^ctfs|^devfs|^objfs|^nfs|^lofs|^sharefs"|sort|uniq`
set $FSTYPES
while test "$1" != ""; do
/bin/df -F $1 -k | egrep -v " /var/run" | tail +2
shift
done
...
This means that /var/run, /dev/fd, /system/object, /etc/mnttab,
/system/contract, /proc and /dev mounts are excluded from the output.
/var and /common seem to be lofs mounts and will be excluded as well.
Dominique
More information about the Xymon
mailing list