Friday, December 3, 2010

How To Monitor Network Switch and Ports Using Nagios

Nagios is hands-down the best monitoring tool to monitor host and network equipments. Using Nagios plugins you can monitor pretty much monitor anything.

I use Nagios intensively and it gives me peace of mind knowing that I will get an alert on my phone, when there is a problem. More than that, if warning levels are setup properly, Nagios will proactively alert you before a problem becomes critical.

Earlier I wrote about, how to setup Nagios to monitor Linux Host, Windows Host and VPN device.

In this article, I’ll explain how to configure Nagios to monitor network switch and it’s active ports.

1. Enable switch.cfg in nagios.cfg
Uncomment the switch.cfg line in /usr/local/nagios/etc/nagios.cfg as shown below.
<!-- @page { margin: 0.79in } P { margin-bottom: 0.08in } PRE.cjk { font-family: "DejaVu LGC Sans Mono", monospace } -->
[nagios-server]# grep switch.cfg /usr/local/nagios/etc/nagios.cfg
cfg_file=/usr/local/nagios/etc/objects/switch.cfg
2. Add new hostgroup for switches in switch.cfg
Add the following switches hostgroup to the /usr/local/nagios/etc/objects/switch.cfg file.
define hostgroup{
hostgroup_name  switches
alias           Network Switches
}
3. Add a new host for the switch to be monitered
In this example, I’ve defined a host to monitor the core switch in the /usr/local/nagios/etc/objects/switch.cfg file. Change the address directive to your switch ip-address accordingly.
define host{
use             generic-switch
host_name       core-switch
alias           Cisco Core Switch
address         192.168.1.50
hostgroups      switches
}
4. Add common services for all switches
Displaying the uptime of the switch and verifying whether switch is alive are common services for all switches. So, define these services under the switches hostgroup_name as shown below.
# Service definition to ping the switch using check_ping
define service{
use                     generic-service
hostgroup_name          switches
service_description     PING
check_command           check_ping!200.0,20%!600.0,60%
normal_check_interval   5
retry_check_interval    1
}

# Service definition to monitor switch uptime using check_snmp
define service{
use                     generic-service
hostgroup_name          switches
service_description     Uptime
check_command           check_snmp!-C public -o sysUpTime.0
}
5. Add service to monitor port bandwidth usage
check_local_mrtgtraf uses the Multil Router Traffic Grapher – MRTG. So, you need to install MRTG for this to work properly. The *.log file mentioned below should point to the MRTG log file on your system.
define service{
use                             generic-service
host_name                       core-switch
service_description     Port 1 Bandwidth Usage
check_command           check_local_mrtgtraf!/var/lib/mrtg/192.168.1.11_1.log!AVG!1000000,2000000!5000000,5000000!10
}
6. Add service to monitor an active switch port
Use check_snmp to monitor the specific port as shown below. The following two services monitors port#1 and port#5. To add additional ports, change the value ifOperStatus.n accordingly. i.e n defines the port#.
<!--
google_ad_client = "pub-8090601437064582";
/* TGS Inside Content Big */
google_ad_slot = "8006040097";
google_ad_width = 336;
google_ad_height = 280;
//-->google_protectAndRun("ads_core.google_render_ad", google_handleError, google_render_ad);# Monitor status of port number 1 on the Cisco core switch
define service{
use                  generic-service
host_name            core-switch
service_description  Port 1 Link Status
check_command        check_snmp!-C public -o ifOperStatus.1 -r 1 -m RFC1213-MIB
}

# Monitor status of port number 5 on the Cisco core switch
define service{
use                  generic-service
host_name            core-switch
service_description  Port 5 Link Status
check_command          check_snmp!-C public -o ifOperStatus.5 -r 1 -m RFC1213-MIB
}
7. Add services to monitor multiple switch ports together
Sometimes you may need to monitor the status of multiple ports combined together. i.e Nagios should send you an alert, even if one of the port is down. In this case, define the following service to monitor multiple ports.
# Monitor ports 1 - 6 on the Cisco core switch.
define service{
use                   generic-service
host_name             core-switch
service_description   Ports 1-6 Link Status
check_command         check_snmp!-C public -o ifOperStatus.1 -r 1 -m RFC1213-MIB, -o ifOperStatus.2 -r 1 -m RFC1213-MIB, -o ifOperStatus.3 -r 1 -m RFC1213-MIB, -o ifOperStatus.4 -r 1 -m RFC1213-MIB, -o ifOperStatus.5 -r 1 -m RFC1213-MIB, -o ifOperStatus.6 -r 1 -m RFC1213-MIB
}
8. Validate configuration and restart nagios
Verify the nagios configuration to make sure there are no warnings and errors.
# /usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg

Total Warnings: 0
Total Errors:   0
Things look okay - No serious problems were detected during the pre-flight check
Restart the nagios server to start monitoring the VPN device.
# /etc/rc.d/init.d/nagios stop
Stopping nagios: .done.

# /etc/rc.d/init.d/nagios start
Starting nagios: done.
Verify the status of the switch from the Nagios web UI: http://{nagios-server}/nagios as shown below:

Fig: Nagios GUI displaying status of a Network Switch

9. Troubleshooting
Issue1: Nagios GUI displays “check_mrtgtraf: Unable to open MRTG log file” error message for the Port bandwidth usage
Solution1: make sure the *.log file defined in the check_local_mrtgtraf service is pointing to the correct location.

Issue2: Nagios UI displays “Return code of 127 is out of bounds – plugin may be missing” error message for Port Link Status.

Solution2: Make sure both net-snmp and net-snmp-util packages are installed. In my case, I was missing the net-snmp-utils package and installing it resolved this issue as shown below.
[nagios-server]# rpm -qa | grep net-snmp
net-snmp-libs-5.1.2-11.el4_6.11.2
net-snmp-5.1.2-11.el4_6.11.2

[nagios-server]# rpm -ivh net-snmp-utils-5.1.2-11.EL4.10.i386.rpm
Preparing...       ########################################### [100%]
1:net-snmp-utils   ########################################### [100%]

[nagios-server]# rpm -qa | grep net-snmp
net-snmp-libs-5.1.2-11.el4_6.11.2
net-snmp-5.1.2-11.el4_6.11.2
net-snmp-utils-5.1.2-11.EL4.10

No comments:

Post a Comment