Advanced Network Interface Check – check_netint / check_snmp_netint

When trying in command line, I get output,
but using same command in Nagios 4.4.9 I just see
"no output on stdout".
Should'nt it write to some temporary files in e.g. /usr/local/nagios_temp ?

This plugin is excellent and works fine on all our Centos 5 and Centos 6 servers. But it doesn't work on Centos 7 system.

Difference in output:

Centos 6:
# /sbin/check_netint -r -n eth1 -fYB -k -w4000,15000 -c10000,25000
eth1:UP (no usable data - 5 rows) (1 UP): OK

Centos 7:
# /sbin/check_netint -r -n enp5s0 -fYB -k -w4000,15000 -c10000,25000
ERROR : Unknown interface enp5s0

Can you help me?. I guess many people face the same on Centos versions higher than version 6.

Hi,

I want use this script to check spanning tree status (fwd, blk, etc ...) on Cisco switch.
I use rstp and there is different state for the same port on different vlan.

How can I check spanning tree state on different vlan ?

thx for your help

Hi.
I have a lot of servers with more than one interface(eth0,eth1,...) and some of them are down.
It's possible to monitor just the UP ones and to ignore the rest without using regexp ?
If I use "-n eth* " I cant figure how to ignore the interfaces that are down.
Thank you very much for the good work.

Below is the command output:

[root@gjvatwb1 libexec]# ./check_snmp_netint.pl -H 172.16.1.105 -C nocgnr -k -M -w5,5 -c8,8 -n "QLogic"

QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #128:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #135:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #135-Trend Micro NDIS 6.0 Filter driver-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #131-QoS Packet Scheduler-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #130-QoS Packet Scheduler-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #128-WFP LightWeight Filter-0000:UP (no usable data - 5 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #135-QoS Packet Scheduler-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #130:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #131-Trend Micro NDIS 6.0 Filter driver-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #131-WFP LightWeight Filter-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #128-Trend Micro NDIS 6.0 Filter driver-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #128-QoS Packet Scheduler-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #130-WFP LightWeight Filter-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #135-WFP LightWeight Filter-0000:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #131:UP (no usable data - 4 rows) , QLogic BCM57840 10 Gigabit Ethernet (NDIS VBD Client) #130-Trend Micro NDIS 6.0 Filter driver-0000:UP (no usable data - 4 rows) :(16 UP): OK |


This isn't the final result. Can I have final command, which should show total bandwidth utilization of single network interface?

was up and running in 10 minutes checking bandwidth on all my important trunk ports. good job!

Does a good job. Pretty dizzying array of options to try and figure out. Have one combination that often results in illegal division by zero error. Any idea what that problem might be?

/ois/usr/nagios/libexec/check_netint.pl -n "eth" -S -k -e -f -w 80,80,2,2,2,2 -c 90,90,4,4,4,4 -q -d 10 -u
Use of uninitialized value $speed_metric in division (/) at /ois/usr/nagios/libexec/check_netint.pl line 2490.
Illegal division by zero at /ois/usr/nagios/libexec/check_netint.pl line 2490.

I had my first bandwidth check up and running 15 minutes after installing.

Initial usage:
Cisco 2921 router (my eline interface) bandwidth usage monitor... 10Mbps interface...

commands.cfg...
define command {
command_name check_eline_bw
command_line /usr/local/sbin/nagios-plugins/check_snmp_netint.pl -H $HOSTADDRESS$ -C -k -M -w5,5 -c8,8 -n $ARG1$
}

And then the service entry....
define service{
use generic-service
host_name ops_router
service_description Eline Bandwidth
check_command check_eline_bw!"GigabitEthernet0/1"
}

Hi, this is great plugin, but perfomrance data is not working. I moved to check_snmp_netint because in check_snmp_int is not working reqex matching name of interface, but on check_snmp_netint is not working perfomrance data generation so I am stuck. I try with command:

check_snmp_netint.pl -H xxx.xxx.xxx.xxx -C public -n "Marvell Yukon 88E8056 PCI-E Gigabit Ethernet Controller$" -f

But I get only this output, where is missing performance data.

Marvell Yukon 88E8056 PCI-E Gigabit Ethernet Controller:UP:(1 UP): OK |

I tried also -z option, but it is same. Can you help me what is needed to change?

Can you please help me solve a problem as sent in e-mail? I want to know exactly the command I use to monitor the traffic on Cisco interfaces, kind if 90Mbps is reached on interface an WARNING is generated etc.

But now the error is GigabitEthernet1/31:UP (no usable data - 201 rows) (1 UP): OK

Below my services.cfg

define service{
host_name br-spo-cbo-acc71
service_description Checagem de Trafego
check_command check_netint!GigabitEthernet1/31
max_check_attempts 3
normal_check_interval 2
retry_check_interval 2
check_period 24x7
notification_interval 28800
notification_period 24x7
notification_options w,c,r
contact_groups administrador
}

Below hosts.cfg

define host {
use switch ; Name of host template to use
host_name br-spo-cbo-acc71
alias hostfiber.com.br-10.99.0.71
hostgroups Infraestrutura
address 10.99.0.71
check_command check-host-alive
max_check_attempts 3
notification_interval 28800
notification_period 24x7
notification_options d,u,r
contact_groups administrador

}

Below my checkcommands:

# check_netint command definition
define command{
command_name check_netint
command_line /usr/lib64/nagios/plugins/check_netint.pl -H $HOSTADDRESS$ -C H0STL0C4T10N -2 -n $ARG1$ -f -Y -k -B -u -d -w 5,5 -c 90,90
}

Really nice plugin for monitoring the status and bandwidth of switch ports for instance.

One addition i would like is this:
You can check the administrative status of an interface with -a, but i would like an option for the check to allways return OK for a port, even though it is down, if the interface is administratively down

For me unfortunately, I have an issue with an Enterasys S6 switch in that the description table is less than useful. Here is the results I get:

IF-MIB::ifDescr.52001 = STRING: Enterasys Networks, Inc. 1000BASE-SX Mini GBIC w/LC connector
IF-MIB::ifDescr.52002 = STRING: Enterasys Networks, Inc. 1000BASE-SX Mini GBIC w/LC connector
IF-MIB::ifDescr.52003 = STRING: Enterasys Networks, Inc. 1000BASE-SX Mini GBIC w/LC connector
IF-MIB::ifDescr.52004 = STRING: Enterasys Networks, Inc. 1000BASE-LX Mini GBIC w/LC connector

Or using the -v switch I get:

OID : 1.3.6.1.2.1.2.2.1.2.52004, Clean Desc : Enterasys Networks, Inc. 1000BASE-LX Mini GBIC w/LC connector, Raw Desc: Enterasys Networks, Inc. 1000BASE-LX Mini GBIC w/LC connector

As you can see…it is just the type of connector. The index name gives me this:

IF-MIB::ifIndex.52001 = INTEGER: 52001
IF-MIB::ifIndex.52002 = INTEGER: 52002
IF-MIB::ifIndex.52003 = INTEGER: 52003
IF-MIB::ifIndex.52004 = INTEGER: 52004

And I know that I need the last entry 52004 (blade module 5, port 4)…so is there a way that I can reference this directly? I've read the instructions in the top of the plugin, I've tried quite a few things, but I can't figure it out.

Would you be able to suggest anything I can look at to resolve this?

I've tried with versions 2.34, 2.36 and 2.4.

Can You add checking is really traffic on interface on-the-fly? Check in|out_octets, sleep timeout, check2. If (check2-check1) we have no traffic. "Speed" have slow variation.

hi,
I would like to use the full scope of functions with our HP switches.
maybe I do not understand all the parameters but when I do the following is the output something little.

. / check_snmp_netint.pl-H xxx.xxx.xxx.xxx-C public-n
VLANXXXX: UP: (1 UP): OK |

especially because there is no VLANXXXX.

the second is we use pnp4nagios. can we use the performance data for this?

thank you for your help!

Hi, thanks for the contribution :)
Does this plugin work with DLink switches?

it's a great tool, but i can't make it work with performance graph, is this possible at all?

I changed a few oids to make it work with juniper FW ssg550 version 6.2
my $oper_table = '1.3.6.1.4.1.3224.9.1.1.5.';
my $in_octet_table = '1.3.6.1.4.1.3224.9.3.1.3.';
my $out_octet_table = '1.3.6.1.4.1.3224.9.3.1.5.';

also had to change the status to be 0 for Down
my %status=('UP'=>1,'DOWN'=>0,'TESTING'=>3,'UNKNOWN'=>4,'DORMANT'=>5,'NotPresent'=>6,'lowerLayerDown'=>7);
my %status_print=(1=>'UP',0=>'DOWN',3=>'TESTING',4=>'UNKNOWN',5=>'DORMANT',6=>'NotPresent',7=>'lowerLayerDown');

And finally the check command is for exmaple (i use another OID for the description table thats why the -n switch ):
./check_netint.pl -H xxx.xxx.xxx.xxx -C public -2 -N 1.3.6.1.4.1.3224.9.1.1.2 -n ethernet0/2.20 -k -B -M -Y -w10,10 -c 20,20

Great work willix! thanks!

hi, im trying to use this plugin with rhel6.4 and it give me the next error:
./check_netint.pl -H 127.0.0.1 -C privado -2 -n eth0 -z
Use of uninitialized value within @oid_perf_inoct in hash element at ./check_netint.pl line 2023.
eth0:UP (1 UP): OK

Any ideas?
Thanks a lot
Javier

Hi,
Brilliant!
Could you add the following feature to your plugin:

Interface selection by using its index instead of its name?

Regards

This plugin has proven extremely valuable in detecting errors on redundant links. We're using it to monitor 600 inter-switch trunks on Cisco, Force10 and Dell switches.

Works with the Cisco SG-300. You need to specify the oid of the description table since the SG line doesn't use the standard Cisco ones.
-N 1.3.6.1.2.1.31.1.1.1.1
-r is necessary if checking port 1 so you don't get port 10 as well.

Hi,
I use this plugin a lot and I would like to make some suggestions.

1. When -f is used with --stp and --intspeed, I get stp and intspeed in the performance output, is there a way to avoid that ? (I tried with and without -S)

2. Checking the duplex mode could be great, for example by adding --intspeed=100Mb/full or 100Mb/half as options.

Thanks anyway for that great software.

William, you can add a template for pnp4nagios to the repository. template can be downloaded from my page - http://worm.org.ua/projects/nagios/

This is a great SNMP plugin for switches and routers. Many features and great support.

I've used check_snmp_int for a long time until I came across this 'fork'. Well it seems to be working good so far (tested on Cisco switches), but I'd like to have the possibility to output perfdata of the interface usage _without_ having to declare warning and critical thresholds (-k). If this feature could be added that would be great.

well it does show if the interface is up or down but it didn't work with bandwidth at least not on windows machines. in the command line it does give output with bandwidth and sometimes says no usable data - 37 rows. in nagios it just gives an unknown error for the command check line:
/usr/bin/perl $USER1$/check_snmp_netint.pl -H 10.1.1.94 -C public -n Realtek -O 1.3.6.1.2.1.2.2.1.2
output: unknown 2 warning levels for bandwidth checks