Combine 2 SNMP Service Checks into a graph & calculation

[eloyd]

I think he was trying to emphasis what he added, not what was actually there. But I could be wrong.

@perric,

Can you please post your entire script, without any modification as an attachment, rather than a code../code block?

[perric]

Bandit,

My mistake on the <b>, </b>. It was left off from editing in the forum editor.

I got the script to work after swapping all $1, $2 variables for static values. How do I make use of the following variables (arguments) from the Nagios GUI?

1. IP Address/Host - do I use $HOSTNAME in the script to allow this service check to get the IP address of the host?
2. Community string
3. Link Size - will $1 map to $ARG1$ automatically?
4. WARNING thresholds - $2 will map to $ARG2$, etc.?

David

[eloyd]

$ARGx$ is an internal Nagios variable. It does not map automatically.

However, in Unix shell scripting, $1 is the first parameter, $2 the second, and so on.

So if you called the script from Nagios as: /usr/local/nagios/libexec/plugin "$ARG1$" "$ARG2$" "$ARG3$" and so on, then in your script, ARG1 would be $1 and ARG2 would be $2, etc.

Edit - corrected "ARG2 would be $3" to "ARG2 would be $2" like a normal person would have written.

[perric]

Eloyd,

Thanks.

Can the hostname be mapped directly using the $HOSTNAME variable in the shell script? Or is there a way to pass the IP address automatically as a variable, without having to specify it as an argument ($ARGx$)?

If I change the IP address on the host object, I would like this to carry through to the service objects without having to edit the service objects. I will eventually have hundreds and possibly thousands of these services.

David

[BanditBBS]

Eloyd,

Thanks.

Can the hostname be mapped directly using the $HOSTNAME variable in the shell script? Or is there a way to pass the IP address automatically as a variable, without having to specify it as an argument ($ARGx$)?

If I change the IP address on the host object, I would like this to carry through to the service objects without having to edit the service objects. I will eventually have hundreds and possibly thousands of these services.

David

Its actually $HOSTNAME$ and you reference in the script/plugin by passing it as a commandline variable. You have to get this working from commandline before even thinking about how to set it up in NagiosXI ccm. You had to hard code $1 and $2??? That makes no sense to me, I had it working just great with those, the only one I had worries about was the IP. Just for the time being, try my original version again, but make sure you put 20000000000 in for the first variable and put the IP inside double quotes and try that. If it doesn't work, then let me know and I'll find a switch to actually test it with here.

[perric]

Bandit,

Your script was fine. The static variables were only for my own troubleshooting, because it was not working for me. The cause was a problem with the variables I was using.

The script is working great with the variables ($1, $2, etc.) added back in. See code below. Hope you do not mind, but I modified some of the calculations. And I also modifed the variables as follows:

$1 - Host IP Address
$2 - Community String
$3 - Link Bandwidth in Mbps
$4 - Cisco Device Interface Index

WARNING and CRITICAL threshold variables are still static, but that's easy to resolve.

Code: Select all

#!/bin/bash
inbound_temp="$(/usr/local/nagios/libexec/check_snmp $1 -C $2 -o 1.3.6.1.2.1.2.2.1.10.$4)"
#inbound_temp="SNMP OK - 3788642629 | iso.3.6.1.2.1.2.2.1.10.15=3788642629c"
outbound_temp="$(/usr/local/nagios/libexec/check_snmp $1 -C $2 -o 1.3.6.1.2.1.2.2.1.16.$4)"
#outbound_temp="SNMP OK - 4243923118 | iso.3.6.1.2.1.2.2.1.16.15=4243923118c"
inbound=`echo $inbound_temp|grep -oP '(?<=\-)(.*?)(?=\|)'`
outbound=`echo $outbound_temp |grep -oP '(?<=\-)(.*?)(?=\|)'`
### Perform Logic here
inbound_per=$(( $inbound / ($3 * 10000000)))
outbound_per=$(( $outbound / ($3 * 10000000)))
total=$(( ($inbound_per + $outbound_per)/2 ))

echo "Inbound Percent = $inbound_per"
echo "Outbound Percent = $outbound_per"
echo "Total Percent = $total"

perfdata="| 'total'=$total%;80;90;0;100 'inbound'=$inbound_per%;;;0;100 'outbound'=$outbound_per%;;;0;100"

if [ "$total" -ge "90" ];then
        echo "CRITICAL - Total traffic above threshold - $total $perfdata"
        exit 2
elif [ "$total" -ge "80" ];then
        echo "WARNING - Total traffic above threshold - $total $perfdata"
        exit 1
else
        echo "OK - Traffic at good percentage $perfdata"
        exit 0
fi

[BanditBBS]

Do what ya want, was just helping and wrote that super quick. Glad it's working

[perric]

The script is working fine, but the values are cumulative from the SNMP check, which throws off the "percentage utilization values". As a fix, I can take two readings and then use the difference as the bandwidth usage.

To do this, I have added a sleep command (60 seconds) to the script and then I take a second reading. See below. Does this pose any impact to Nagios in terms of performance, etc.? Is there a better way to do this?

Code: Select all

#!/bin/bash
#### Collect first reading ####
inbound_temp="$(/usr/local/nagios/libexec/check_snmp $1 -P 2c -C $2 -o 1.3.6.1.2.1.31.1.1.1.6.$4)"
outbound_temp="$(/usr/local/nagios/libexec/check_snmp $1 -P 2c -C $2 -o 1.3.6.1.2.1.31.1.1.1.10.$4)"
inbound=`echo $inbound_temp|grep -oP '(?<=\-)(.*?)(?=\|)'`
outbound=`echo $outbound_temp |grep -oP '(?<=\-)(.*?)(?=\|)'`

sleep 60

#### Collect 2nd reading ####
inbound_temp2="$(/usr/local/nagios/libexec/check_snmp $1 -P 2c -C $2 -o 1.3.6.1.2.1.31.1.1.1.6.$4)"
outbound_temp2="$(/usr/local/nagios/libexec/check_snmp $1 -P 2c -C $2 -o 1.3.6.1.2.1.31.1.1.1.10.$4)"
inbound2=`echo $inbound_temp2|grep -oP '(?<=\-)(.*?)(?=\|)'`
outbound2=`echo $outbound_temp2 |grep -oP '(?<=\-)(.*?)(?=\|)'`


### Perform Logic here
#echo "$inbound"
#echo "$outbound"
inbound_per=$(( (($inbound2 - $inbound) * 8) / ($3 * 1000000)))
outbound_per=$(( (($outbound2 - $outbound) * 8) / ($3 * 1000000)))
total=$(( ($inbound_per + $outbound_per)/2 ))

#echo "Inbound Percent = $inbound_per"
#echo "Outbound Percent = $outbound_per"
#echo "Total Percent = $total"

perfdata="| 'inbound'=$inbound_per%;;;0;100 'outbound'=$outbound_per%;;;0;100"

if [ "$total" -ge "90" ];then
        echo "CRITICAL - Total traffic above threshold - $total $perfdata"
        exit 2
elif [ "$total" -ge "80" ];then
        echo "WARNING - Total traffic above threshold - $total $perfdata"
        exit 1
else
        echo "OK - Traffic at good percentage $perfdata"
        exit 0
fi

Note - I also changed the SNMP to version 2c, so that I could take readings from 64 bit counters, which are required for high speed Cisco interfaces. Values collected are multiplied by 8 to convert octets (bytes) to bits.

David

[perric]

The "sleep 60" command is giving "(Service check timed out after 60.00 seconds)" errors in Nagios, so I need to come up with another approach.

David

[BanditBBS]

The "sleep 60" command is giving "(Service check timed out after 60.00 seconds)" errors in Nagios, so I need to come up with another approach.

David

You can use it, you just have to tell Nagios it is OK for service checks to take that long. Its a variable in the nagios.cfg file.