Network Interface Utilization on Windows Servers

[joe.ward]

What's the best way to monitor bytes in/out on the NIC interface on a Windows server? It seems it should be simple like CPU or memory utilization but I sure haven't been able to find and configure anything so far

When using the Windows Server configuration wizard, it has a place for Performance Counters. Is there something I can put there to capture NIC utilization? (Is there a list of what can be put in Performance Counters?)

Thank you

[slansing]

Do you have counters for your NIC? The way to find this out would be to take a look at the windows "perfmon" application.

[joe.ward]

The Performance Counters are enabled.

I've tried "check_xi_service_nsclient -H $HOSTADDRESS$ -I "\\Network Interface\\Bytes Sent/sec" ...

I also tried \\Network Interface\\INTEL[R] PRO_1000 MT Network Adapter\\Bytes Sent/sec"

What have I missed?

[slansing]

Run the following and let us know what the output is, fill in the spots which require your own info:

Code: Select all

/usr/local/nagios/libexec/check_nt -H windows.server.ip -p 12489 -v COUNTER -l "\\Network Interface(Intel[R] PRO_1000 MT Network Connection)\Bytes Sent/sec" -w 1000 -c 3000

[joe.ward]

I added "-s password" and it works! Thank you for helping figure out the syntax!

[joe.ward]

OK - I guess I jumped the gun... It works on the Nagios server at the command line. How does it translate to the service in Nagios XI?

I have defined a command "check_nic" as "$USER1$/check_nt -H $HOSTADDRESS$ -s $ARG1$ -p 12489 -v COUNTER -l $ARG2$"

Then the check command in the service screen is:
check_nic
$USER1$/check_nt -H $HOSTADDRESS$ -s $ARG1$ -p 12489 -v COUNTER -l $ARG2$
$ARG1$ = password
$ARG2$ = "\\Network Interface(Intel[R] PRO_1000 MT Network Connection)\Bytes Sent/sec"
$ARG3$ = -w 1000000 -c 3000000

Again, this works on the Nagios server from the command line:

/usr/local/nagios/libexec/check_nt -H ip.addr -s password -p 12489 -v COUNTER -l "\\Network Interface(Intel[R] PRO_1000 MT Network Connection)\Bytes Sent/sec" -w 1000000 -c 3000000

[sreinhardt]

Are you seeing this fail via the test check button, or in actually letting core run this? I ask because the test check button runs things via apache and can heavily alter and escape commands like this with somewhat special characters, such that it may throw false flags when a command should work fine via core, as in this case.

[joe.ward]

Are you seeing this fail via the test check button? YES

If I log in on the LINUX host that is running Nagios, and run the command at the command line, then it returns a non-zero value (correctly).

When I configure the service in Nagios XI Core Configuration , Services, and use the Test button, it fails. I thought it might be the way the test button escapes things so I saved the configuration, let the service check run, then looked at the status. Zero!

How do I translate from the command line in LINUX to the GUI Service configuration window in Nagios XI?

Thank you

[sreinhardt]

I would believe the way you presently have it should work perfectly. Let me do some testing with a similar interface and get back to you!

[joe.ward]

I would believe the way you presently have it should work perfectly. Let me do some testing with a similar interface and get back to you!

Any luck in figuring out what I've done wrong?