Author | Vernon Everett |
Compatibility | Xymon 4.2 |
Requirements | ZFS |
Download | None |
Last Update | 2020-04-20 |
An update to the BB ZFS monitor to graph capacity and keep tabs on snapshots as well
Client side
Server side
GRAPHS="...,xymond,zfs"
TEST2RRD="...,xymond,zfs=ncv"
SPLITNCV_zfs="*:GAUGE"
#!/bin/ksh
# Revision History:
# 1. Mike Rowell <Mike.Rowell@Rightmove.co.uk>, original
# 2. Uwe Kirbach <U.Kirbach@EnBW.com>
# 3. T.J. Yang: add in some comments.
# 4. Vernon Everett <everett.vernon@gmail.com : Added check for old snapshots
# : Added graphing data
# 5. FreeSoftwareServers <admin@freesoftwareservers.com>:
# Update to new "zpool list -h" format
# Add Available/Free to table
# Full "zpool status" always displayed
# Set DEGRADED=RED
# DISKYELL=90 & DISKRED=95
DISKYELL=90
DISKRED=95
SNAPRED=90 # Days old
SNAPYELL=60 # Days old
SNAPCOL=true # Set to true if snapshot age should effect test colour
CHECKSNAPS=true # Set to true to do snapshot checking
TEST="zfs"
DISPCOLOR="green"
FIRST_LINE="zfs - okay"
FIRST_LINE_HEALTH="okay"
FIRST_LINE_CAP="okay"
DATA=""
#What: beautify the page display by html code.
STRING="<table border=1 cellpadding=10><tr><th></th><th>Zpool Name</th><th>Status</th><th>Allocated</th><th>Free</th><th>Capacity</th></tr>"
#What: a loop to parse output of "zpool list -H" output.
# bash-3.00# zpool list
# NAME SIZE USED AVAIL CAP HEALTH ALTROOT
# mypool 33.8G 84.5K 33.7G 0% ONLINE -
#####UPDATE####
#
#:# zpool list
#NAME SIZE ALLOC FREE EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT
#raid-z 18.1T 13.6T 4.54T - 0% 74% 1.00x ONLINE -
# bash-3.00# zpool list -H
# mypool 33.8G 84.5K 33.7G 0% ONLINE -
# bash-3.00#
zpool list -H | while read name size alloc free expandsz frag cap dedup health altroot
do
LINE_COLOR="green"
if [ "${health}" == "ONLINE" ]; then
HEALTH_COLOR="green"
elif [ "${health}" == "DEGRADED" ]; then
HEALTH_COLOR="red"
elif [ "${health}" == "FAULTED" ]; then
HEALTH_COLOR="red"
fi
CAP_COLOR="green"
cap=`echo ${cap} | cut -d% -f1`
if [ "${cap}" -ge $DISKYELL ]; then
CAP_COLOR="yellow"
elif [ "${cap}" -gt $DISKRED ]; then
CAP_COLOR="red"
fi
DATA=$(echo "$name : $cap\n$DATA")
# Determine the line colours
[ "$HEALTH_COLOR" == "yellow" -o "$CAP_COLOR" == "yellow" ] && LINE_COLOR="yellow"
[ "$HEALTH_COLOR" == "red" -o "$CAP_COLOR" == "red" ] && LINE_COLOR="red"
# Determine the messages
[ "$HEALTH_COLOR" == "red" -a "$FIRST_LINE_HEALTH" != "faulted" ] && FIRST_LINE_HEALTH="DEGRADED"
[ "$HEALTH_COLOR" == "red" -a "$FIRST_LINE_HEALTH" != "DEGRADED" ] && FIRST_LINE_HEALTH="FAULTED"
[ "$CAP_COLOR" == "yellow" -a "$FIRST_LINE_CAP" != "full" ] && FIRST_LINE_CAP="nearly full"
[ "$CAP_COLOR" = "yellow" ] && FIRST_LINE_CAP="full"
#Determine the final colour status
[ "$LINE_COLOR" == "yellow" -a "$DISPCOLOR" != "red" ] && DISPCOLOR="yellow"
[ "$LINE_COLOR" == "red" ] && DISPCOLOR="red"
STRING="$STRING <tr><td>&${LINE_COLOR}</td><td>${name}</td><td>${health}</td><td>${alloc}</td><td>${free}</td><td>${cap} %</td></tr>"
done
DATA=$(echo "$DATA \n\n")
STRING="$STRING </table><br><br>"
STRING="$STRING`zpool status`"
FIRST_LINE="zfs - Health Report: $FIRST_LINE_HEALTH - Capacity Report: $FIRST_LINE_CAP"
# Snapshot check
if [ "$CHECKSNAPS" = "true" ]
then
NOW=$(perl -e 'print time(), "\n" ')
SNAPLIST=$(zfs list | grep @ |awk '{ print $1 }' \
| while read a
do
echo "$(zfs get -H -o name,value -p creation $a) $( zfs get -H -o value used $a)"
done)
if [ -z "$SNAPLIST" ]
then
SNAPTABLE="<br>&green No snapshots found"
else
SNAPTABLE="<table border=1 cellpadding=10><tr><th></th><th>Snapshot</th><th>Age</th><th>Size</th></tr>"
echo "$SNAPLIST" | while read SNAPSHOT CREATION SIZE
do
LINE_COLOR=green
((SNAPREDS=SNAPRED*43200))
((SNAPYELLS=SNAPYELL*43200))
((AGES=NOW-CREATION)) # AGES=Age in seconds
[ $AGES -gt $SNAPYELLS ] && LINE_COLOR=yellow
[ $AGES -gt $SNAPREDS ] && LINE_COLOR=red
[ $AGES -gt 120 ] && ((AGE=AGES/60))&& AGE="$AGE Minutes"
[ $AGES -gt 7200 ] && ((AGE=AGES/3600)) && AGE="$AGE Hours"
[ $AGES -gt 172800 ] && ((AGE=AGES/86400)) && AGE="$AGE Days"
SNAPTABLE="$SNAPTABLE <tr><td>&${LINE_COLOR}</td><td>${SNAPSHOT}</td><td>${AGE}</td><td>${SIZE}</td></tr>"
if [ "$SNAPCOL" = "true" ] #Only if true will it effect test colour
then
if [ "$DISPCOLOR" != "red" ] # If it's already red, it's not getting any worse
then
[ "$LINE_COLOR" != "green" ] && DISPCOLOR=$LINE_COLOR
fi
fi
done
SNAPTABLE="$SNAPTABLE </table><br><br>"
fi
STRING="$STRING <br><br><B>SNAPSHOT STATUS</B> $SNAPTABLE"
fi
# What: Sent out the final bb message to hobbit server.
$BB $BBDISP "status $MACHINE.$TEST $DISPCOLOR `date` $FIRST_LINE $STRING"
$BB $BBDISP "data $MACHINE.$TEST $DISPCOLOR $(echo; echo "$DATA" ;echo;echo)"
Known Bugs and Issues
None known at this stage.
I am eternally grateful to those that have assisted in creating this script. All I have done is add a little extra functionality to monitor snapshots and a little bit of graphing.
From the script, it appears credit is owed to