NRPE til icinga.aoit.dk

Vejledningen er til Redhat EL baseret distributioner, men konceptet og konfigurationen er den samme.

Installation

For at lette installationen anbefales det at tilføje repoforge til dit rpm reposetory, du kan altid fjerne det bagefter.
Følg vejledningen her http://repoforge.org/use/

Herefter kan du installere nrpe, nagios-plugins og dets afhængigheder på en gang med yum

#yum install nrpe nagios-plugins

Tag en kopi af nrpe konfigurations filen

cp -p /etc/nagios/nrpe.cfg /etc/nagios/nrpe.cfg.orig

Tilføj herefter IP adressen på icinga.aoit.dk, i konfigurationens filen.

allowed_hosts=127.0.0.1,84.22.97.202

Installer nrpe som en service.

#chkconfig --add nrpe

Tilføj service i runlevels

#chkconfig nrpe on --level 3

Start nrpe.

#service nrpe start

Firewall

Der skal åbnes for TCP trafik på port 5666 til din maskine fra 84.22.97.202

Tilpassning

Til forskel fra overvågning af internet services, er der ved lokal overvågning to forskellige måder at opsætte check kommandoer til Icinga.
Den ene, hvor du tillader at AOit sender kommando argumenter til NRPE på din server og du kan derfor bruge AOit's Icinga konfigurator til opsætning af kommandoer ligesom ved opsætning af internet services. Det giver den bedste brugervenlighed men også mindre fleksibilitet til at lave sine helt egne kommandoer.
Den anden måde, hvor der ikke tillades argumenter kræver du selv sætter dine alarmdefinationer op i din lokale konfigurationsfil. Ud over du så kan sammensætte dine helt egne check plugins giver det også en lidt større sikkerhed da du præcist ved hvad dine plugins udføre.

Når du har besluttet dig skal du i konfigurations filen angive hvilen metode du vil bruge.

dont_blame_nrpe=

Sæt til "0" for at tillade argumenter og bruge AOit's Icinga kongigurator.
Sæt til "1" for at selv opsætte check kommandoer og have mere kontrol med hvad der udføres på din server.

Til sidst skal de kommandoer du vil benytte indsættes i konfigurations filen. Nederst i filen kan de se en stribe eksempler på kommando definationer, de skal slettes og erstattes af nye.

Hvis du har valgt at bruge argumenter skal du bruge kommando nedenstående definationer.
På 64bits intel systemer ligger de plugins der køres i /usr/lib64/nagios/plugins, bruge den rigtige stig i forhold til din installation.

command[check_load_nrpe]=/usr/lib64/nagios/plugins/check_load -w $ARG1$ -c $ARG2$
command[check_disk_nrpe]=/usr/lib64/nagios/plugins/check_disk -w $ARG1$ -c $ARG2$ -p $ARG3$
command[check_swap_nrpe]=/usr/lib64/nagios/plugins/check_swap -w $ARG1$ -c $ARG2$
command[check_proc_states_nrpe]=/usr/lib64/nagios/plugins/check_procs -w $ARG1$ -c $ARG2$ -s $ARG3$
command[check_proc_nrpe]=/usr/lib64/nagios/plugins/check_procs -w $ARG1$ -c $ARG2$ -C $ARG3$
command[check_procs_nrpe]=/usr/lib64/nagios/plugins/check_procs -w $ARG1$ -c $ARG2$
command[check_sensors_nrpe]=/usr/lib64/nagios/plugins/check_sensors

Hvis du har valgt uden argumenter skal du bruge disse kommandoer. Vær opmærksom på argumenterne er eksempler og bør tilrettes.

command[check_load_nrpe]=/usr/lib64/nagios/plugins/check_load -w 15,10,5 -c 30,25,20
command[check_disk0_noarg]=/usr/lib64/nagios/plugins/check_disk -w 20% -c 10% -p /dev/hda1
command[check_disk1]=/usr/lib64/nagios/plugins/check_disk -w 20% -c 10% -p /dev/hdb1
command[check_disk2]=/usr/lib64/nagios/plugins/check_disk -w 20% -c 10% -p /mnt/backup
command[check_disk3]=/usr/lib64/nagios/plugins/check_disk -w 20% -c 10% -p /home
command[check_disk4]=/usr/lib64/nagios/plugins/check_disk -w 20% -c 10% -p /mnt/das
command[check_swap]=/usr/lib64/nagios/plugins/check_swap -w 75% -c 50%
command[check_zombies]=/usr/lib64/nagios/plugins/check_procs -w 5 -c 10 -s Z
command[check_proc0]=/usr/lib64/nagios/plugins/check_procs -w 0:0 -c 0:0 -C vnstat
command[check_proc1]=/usr/lib64/nagios/plugins/check_procs -w 150 -c 200
command[check_proc2]=/usr/lib64/nagios/plugins/check_procs -w 150 -c 200
command[check_proc3]=/usr/lib64/nagios/plugins/check_procs -w 150 -c 200
command[check_proc4]=/usr/lib64/nagios/plugins/check_procs -w 150 -c 200
command[check_sensors]=/usr/lib64/nagios/plugins/check_sensors

BOINC for a brighter future

Berkeley Open Infrastructure for Network Computing
Mest kendt er SETI@home the seach for extraterestial at home der for mere end ti år siden
http://boinc.berkeley.edu/download_all.php
Older Linux x86 and x64
wget http://boinc.berkeley.edu/dl/boinc_compat_6.12.22_i686-pc-linux-gnu.sh
./boinc_compat_6.12.22_i686-pc-linux-gnu.sh
mv BOINC /home/boinc
useradd boinc
chown -R boinc:boinc /home/boinc
rpm -qa | grep libstdc++.so.6
if [ !$? ]
then
rpm -ivh libstdc++.so.6
fi
ln -s libstdc++.so.6 libstdc++.so.3
su -c "/home/boinc/boinc > /home/boinc/boinc.log 2>&1 &" boinc
echo "Let it run some bencmarking and ctrl+c when it reaches "Resuming computation"
tail -f /home/boinc/boinc.log
./boinccmd --create_account http://boinc.bakerlab.org/rosetta EMAIL PASSWORD USER
./boinccmd --project-attach http://boinc.bakerlab.org/rosetta KEY
vi /etc/init.d/boinc

#!/bin/sh # # BOINC - start and stop the BOINC client daemon on Unix # # Unix start/stop script to run the BOINC client as a daemon at # system startup, as the 'boinc' user (not root!). # # This version works on Red Hat Linux, Fedora Core, Debian, Ubuntu, SUSE, # Mandrake, and Slackware Linux, and should work on generic Linux systems # provided that they have 'pidof' and a POSIX /bin/sh (most all do). # Metadata for chkconfig and the SUSE equivalent INIT info are included below. # # Usage: boinc { start | stop | status | restart } # ### # chkconfig: 345 98 03 # description: This script starts the local BOINC client as a daemon # For more information about BOINC (the Berkeley Open Infrastructure # for Network Computing) see http://boinc.berkeley.edu # processname: boinc # config: /etc/sysconfig/boinc # ### BEGIN INIT INFO # Provides: boinc # Required-Start: $network # Required-Stop: $network # Default-Start: 3 4 5 # Default-Stop: 0 1 2 6 # Description: This script starts the local BOINC client as a daemon # For more information about BOINC (the Berkeley Open Infrastructure # for Network Computing) see http://boinc.berkeley.edu ### END INIT INFO # # Eric Myers <myers@vassar.edu> - 27 July 2004 # Department of Physics and Astronomy, Vassar College, Poughkeepsie NY # Eric Myers <myers@spy-hill.net> # Spy Hill Research, Poughkeepsie, New York # @(#) $Id: boinc,v 1.14 2008/07/14 12:52:34 myers Exp $ ######################################################################## # Defaults, which can be overridden by putting new NAME=value lines # in /etc/sysconfig/boinc (for Red Hat/Fedora Linux and variants) # or /etc/default/boinc (for Debian/Ubuntu and variants) # Name of user to run as: # BOINCUSER=boinc # Working directory. Could be /home/boinc, /var/lib/boinc, etc.. # The reason /var/lib/boinc is preferable is that this works best for a # cluster of computers whereas /home/anything might be shared between machines # BOINCDIR=/home/boinc # Name of the client executable. This is the file "boinc" if you # unpacked the download file boinc_M.mm.rr_i686-pc-linux-gnu.sh, # but I like to rename it and put it in a public place. # (Hint: move boincmgr to /usr/local/bin too so anyone can easily use it). # BOINCEXE=/home/boinc/boinc # Log and error files (you should rotate these occasionally) # LOGFILE=boinc.log ERRORLOG=boinc.log # BOINC options for the command line when running the client. # Be aware that --allow_remote_gui_rpc opens up your machine to the world! # (Or at least to your local subnet) # #BOINCOPTS="--allow_remote_gui_rpc" BOINCOPTS="" # Deprecated. Was used to construct the executable name in days of old. # (Still here just as a last ditch to find an old executable, but # probably should be removed.) # #BUILD_ARCH=i686-pc-linux-gnu # Mandrake 10.1 really wants a subsys lock file ... #if [ -d /var/lock/subsys ]; then # LOCKDIR=/var/lock/subsys #elif [ -d /var/lock ]; then # LOCKDIR=/var/lock #fi # su on Linux seems to need this to be set to work properly in a script export TERM dumb ## # Init script function library. This stuff is Red Hat specific, # but if the functions are not found we create our own simple replacements. # (The idea for replacing the functions comes from OpenAFS. Thanks guys!) if [ -f /etc/rc.d/init.d/functions ] ; then . /etc/rc.d/init.d/functions ##TODO: Try to use LSB here for systems that have it? else export PATH=/sbin:/bin:/usr/sbin:/usr/bin echo_success () { echo -n " [ OK ] " ; } echo_failure () { echo -n " [FAILED] " ; } echo_warning () { echo -n " [WARNING] " ; } killproc() { PID=`pidof -s -x -o $$ -o $PPID -o %PPID $1` [ $PID ] && kill $PID ; } fi ## Look for any local configuration settings which override all above if [ -f /etc/sysconfig/boinc ]; then . /etc/sysconfig/boinc . /etc/sysconfig/boinc elif [ -f /etc/default/boinc ]; then . /etc/default/boinc fi ## Verify the working directory exists: if [ ! -d $BOINCDIR ]; then echo -n "Cannot find BOINC directory $BOINCDIR " echo_failure echo exit 7 fi # Some additional places to look for the client executable # (We do this after init.d/functions and sysconfig/boinc, which sets PATH) export PATH=$BOINCDIR:/usr/local/bin:$PATH ## Locate the executable, either boinc_client, boinc, ## or boinc_M.mm_.... with highest version number ## We only do this if BOINCEXE set above isn't found or is not executable. if [ ! -x $BOINCEXE ]; then BOINCEXE=`/usr/bin/which boinc_client 2>/dev/null` if [ ! -x "$BOINCEXE" ]; then BOINCEXE=`/usr/bin/which boinc 2>/dev/null` if [ ! -x "$BOINCEXE" ]; then BOINCEXE=`/bin/ls -1 $BOINCDIR/boinc_*_$BUILD_ARCH 2>/dev/null | tail -1 ` fi fi fi if [ ! -x "$BOINCEXE" ]; then echo -n "Cannot find an executable for the BOINC client." echo_failure echo exit 2 fi ## Functions: $1 is one of start|stop|status|restart case "$1" in start) cd $BOINCDIR if [ -f lockfile ] ; then echo -n "Another instance of BOINC is running (lockfile exists)." echo_failure echo exit 4 fi if [ ! -d projects ] ; then echo -n "The BOINC client requires initialization." echo_warning echo fi echo -n "Starting BOINC client as a daemon: " su $BOINCUSER -c "$BOINCEXE $BOINCOPTS" >>$LOGFILE 2>>$ERRORLOG & sleep 4 PID=`pidof -s -x -o $$ -o $PPID -o %PPID $BOINCEXE` if [ $PID ]; then [ -d $LOCKDIR ] && touch $LOCKDIR/boinc echo_success else echo_failure fi echo ;; stop) cd $BOINCDIR if [ ! -f lockfile -a ! -f $LOCKDIR/boinc ] ; then echo -n "BOINC is not running (no lockfiles found)." echo_success else echo -n "Stopping BOINC client daemon: " killproc $BOINCEXE && echo_success || echo_failure # clean up in any case rm -f $BOINCDIR/lockfile rm -f $LOCKDIR/boinc fi echo ;; restart) $0 stop $0 start ;; status) PID=`pidof -x -o $$ -o $PPID -o %PPID boinc_client` if [ "x$PID" = "x" ]; then PID=`pidof -x -o $$ -o $PPID -o %PPID $BOINCEXE` fi if [ "$PID" != "" ]; then echo "BOINC client is running (pid $PID)." else if [ -f $BOINCDIR/lockfile -o -f $LOCKDIR/boinc ]; then echo "BOINC is stopped but lockfile(s) exist." else echo "BOINC client is stopped." fi fi ;; *) echo "Usage: boinc {start|stop|restart|status}" exit 1 esac

chmod +x /etc/init.d/boinc
chkconfig --add boinc