系统监视

转载自：http://www.linuxnote.org/system-monitoring.html

一.监视并重启httpd服务

#!/bin/bash
# Apache Process Monitor
# Restart Apache Web Server When It Goes Down
# RHEL / CentOS / Fedora Linux restart command
RESTART="service httpd restart"
# uncomment if you are using Debian / Ubuntu Linux
#RESTART="/etc/init.d/apache2 restart"
#path to pgrep command
PGREP="/usr/bin/pgrep"
# find httpd pid
$PGREP httpd
if [ $? -eq 0 ]
then
# restart apache
$RESTART
fi

　　

check process

#!/bin/bash
process="/usr/sbin/httpd"
start="service httpd restart"
ps ax | awk `{print $5}` | grep -q "^$process$" || {
# Apparently not running, so start the process
eval "$start"
exit $?
}
exit 0

　　

check apache

#!/bin/bash
URL="http://localhost/netsaint/index.html"
TIMEOUT=10 #(seconds)
MATCH=""
restart="service httpd restart"
wget -q -O – -T=$TIMEOUT –tries=1 "$URL" | grep -q "$MATCH" || {
# Something is wrong, so restart
eval "$restart"
exit $?
}
exit 0

　　

check port

#!/bin/bash
port="80″
restart="service httpd restart"
netstat -ln | awk `/^tcp/ {print $4}` | grep -q ":$port$" || {
# Apparently not listening, so run restart command
eval "$restart"
exit $?
}
exit 0

　　

log watch all

#!/bin/bash
HOSTS=`
host1|admin1@someplace.com
host2|admin2@someplace.com
host3|admin3@someplace.com
for entry in $HOSTS ; do
host=`echo "$entry" | sed `s/|.*$//``
email=`echo "$entry" | sed `s/^.*|//``
logwatch –mailto "$email" –hostname "$host"
done

　　

web

#!/bin/bash
SLEEP=30 #seconds
TIMEOUT=10 #seconds
URL=`http://localhost/index.html`
MATCH=``
while true ; do
wget -q -O – -T=$TIMEOUT –tries=1 "$URL" | grep -q "$MATCH" || {
/usr/local/sbin/report ERROR "Web server not responding"
/etc/rc.d/init.d/httpd restart && {
/usr/local/sbin/report INFO "Web server restarted"
}
}
sleep $SLEEP
done

　　

log watch example

#!/bin/bash
# Note, needs to be placed in /etc/log.d/scripts/services and named
# according to your configuration file (i.e. a config file of
# `example.conf` would require this file to be named `example`.
# This is as nice script that will show you the lines you will
# be processing and reporting on. It will first display the
# standard environment variables and then it takes STDIN and
# dump it right back out to STDOUT.
# These are the standard environment variables.
echo "Date Range : $LOGWATCH_DATE_RANGE"
echo "Detail Level: $LOGWATCH_DETAIL_LEVEL"
echo "Temp Dir: $LOGWATCH_TEMP_DIR"
echo "Debug Level: $LOGWATCH_DEBUG"
# Now take STDIN and dump it to STDOUT

　　

按如下方式设置调度作业 (cronjob):

*/5 * * * * /path/to/script.sh >/dev/null 2>&1

二下面的脚本用业监视系统进程 , 根据指定的服务端口是否打开，输出相应的信息

执行该脚本，看能否得到正确的结果，如果不能，修改脚本使之能正常运行．

#!/bin/bash
# Shell script to monitor running services such as web/http, ssh, mail etc.
# If service fails it will send an Email to ADMIN user
# service port
ports="22 80 25″
# service names as per above ports
service="SSH WEB MAIL"
# No of services to monitor as per (above ports+1)
SCOUNTER=4
#Email id to send alert
ADMINEMAIL="admin@myispname.com"
# counter
c=1
echo "Running services status:"
# use sudo if you want i.e. sudo /bin/netstat
/bin/netstat -tulpn | grep -vE `^Active|Proto` | while read LINE
do
sendMail=0
# get active port name and use : as delimiter
t=$(echo $LINE | awk `{ print $4}` | cut -d: -f2)
[ "$t" == "" ] && t=-1 || :
# get service name from $services and : as delimiter
sname=$(echo $service | cut -d` ` -f$c)
sstatus="$sname: No"
# now compare port
for i in $ports
do
if [ $i -eq $t ]; then
sstatus="$sname: Ok"
sendMail=1
fi
done
# display service status as OK or NO
echo "$sstatus"
#next service please
c=$( expr $c + 1 )
[ "$sendMail" == "0" ] && echo $sstatus | mail -s "service down $sstatus" $ADMINEMAIL || :
# break afer 3 services
[ $c -ge $SCOUNTER ] && break || :
done

　　

　　

三下面的脚本用ping来检查远程主机是否有响应，输出相应的信息

执行该脚本，看能否得到正确的结果，如果不能，修改脚本使之能正常运行．

#!/bin/bash
# Simple SHELL script for Linux and UNIX system monitoring with
# ping command
# add ip / hostname separated by while space
HOSTS="cyberciti.biz theos.in router"
# no ping request
COUNT=1
# email report when
SUBJECT="Ping failed"
EMAILID="me@mydomain.com"
for myHost in $HOSTS
do
count=$(ping -c $COUNT $myHost | grep `received` | awk -F`,` `{ print $2 }` | awk `{ print $1 }`)
if [ $count -eq 0 ]; then
# 100% failed
echo "Host : $myHost is down (ping failed) at $(date)" | mail -s "$SUBJECT" $EMAILID
fi
done

　　

check drives

#!/bin/bash
types="ext2|ext3|ufs|vfat"
cutoff="90″
for drive in `mount | awk "/type ($types)/ {print \\\$1}"` ; do
df "$drive" | awk -v "cutoff=$cutoff" `/^\// {
gsub(/%$/, "", $5);
if ($5 > cutoff)
print "Drive " $1 " (" $6 ") is " $5 "% Full"
}`
done | /usr/local/sbin/report –stdin WARNING `Drives almost full`

　　

check for failure

#!/bin/bash
# Check for any drive failures
egrep `I/O error: dev .+, sector` /var/log/messages | \
/usr/local/sbin/report –stdin CRITICAL `Drives failure`
# Check for any RAID failures
egrep `raid.*Disk failure on` /var/log/messages | \
/usr/local/sbin/report –stdin CRITICAL `RAID drive failure`

　　

services

#!/bin/bash
SLEEP=30 #seconds
while true ; do
cat /usr/local/etc/services.conf | while read line ; do
match=`echo "$line" | sed `s/|.*$//``
cmd=`echo "$line" | sed `s/^.*|//``
ps ax -o `%c %P` | awk -v "process=$match" \
`{if (($1 == process) && ($2 == 1)) exit 1}` && {
# Process not found!
/usr/local/sbin/report ERROR "Process $match not running"
[ -n "$cmd" ] && {
$cmd && {
/usr/local/sbin/report INFO \
"Process $match was restarted ($cmd)"
}
}
}
done
sleep $SLEEP
done