Nagios Support Forum

Hi there,
We've been using the Network Router/Switch Config Wizard quite a bit for when we roll out a new WAN router at the customer site.
However, we have sites where we do not control the WAN side of things and do not want to open ports on the firewall to allow the XI server to run that scan on the customers internal switches, etc.

We run small Linux client on customer premise running NRDS for passive checks. I have a few questions:

1) Can I run that wizard or something similar on the Linux client that I have on premise to detect the switch ports/port names, etc like the Wizard does? I don't mind if I have to manually copy that info to make the NRDS commands, but anything is better than a manual snmpwalk and trying to pull the info that way.

2) Once I get the OIDs and such for the ports I want to monitor, how does the graphing/perf data work? The NRDS agent on premise would be "gathering" that info from the customer switches and reporting back to XI in our datacenter -- would the perf data be transferred back to XI To be graphed properly?

1) Not currently, but you could run the cfgmaker command on the remote server and transfer that output back to the XI server. You would still need to create the host/services manually though which kind of defeats the purpose.

2) Please see here for an architecture overview:

https://support.nagios.com/kb/article.php?id=62

ssax wrote:1) Not currently, but you could run the cfgmaker command on the remote server and transfer that output back to the XI server. You would still need to create the host/services manually though which kind of defeats the purpose.

2) Please see here for an architecture overview:

https://support.nagios.com/kb/article.php?id=62

Hmmm cfg maker might work. I really am just interested in the output of what the network switch wizard creates.
Specifically about just automatically finding the port names and such. How would I get something like that without crawling thru an snmpwalk? I don't mind picking out the info and dropping into an NRDP config in the XI GUI.

There is a lot more going on in the background when the wizard it run.
To get the bandwidth information, the XI server runs MRTG every 5 minutes to gather the data from the device and stores that in a file that is local to the XI sever.
Then the check_mrtgtraf plugin looks at that file to get the bandwidth information to display it in XI.
You would have to replicate that whole process just for the bandwidth info.

For simplicity sake, I'd be ok with just having something pull the port name and numbers like the wizard does. I can figure out something else for the BW.

Alternatively, if I put an XI server at customer premise, can I somehow see those hosts/services on my main XI server? Just thinking out loud...

You can run a snmpwalk and grep for the string ifDescr to get the description and grep for ifIndex to get the number.

If you do install an XI server at the remote site, you could setup an outbound transfer from the remote system to the inbound at the Main XI server.
That will transfer the check status from the remote system to the main system.
Take a look at these links for more details on that.
https://assets.nagios.com/downloads/nag ... ith_XI.pdf
https://assets.nagios.com/downloads/nag ... ith_XI.pdf