Monitor Swith port utilization trend !!!

[aluk]

Does anyone know the procedure in using MRTG to monitor specific switch port utilization trend since MRTG can keep long trend instead of online monitoring.!! I know that some MIBs must be know in configuration, right?

[steve.barlow]

I am using it on NT, but you can install it on unix as well.

Here is the link:

http://mrtg.hdl.com/mrtg.html

Steps:

1. install perl on your monitoring station.

2. Install mrtg on your station

3. configure the router/switch/pix to allow snmp from that station

4. configure a script file (called mrtg.cfg) that contains the MIBs you want to grab.

eg.

### Interface 2 >> Descr: 'FastEthernet0/1' | Name: 'Fa0/1' | Ip: '' | Eth: '00-07-ec-05-8f-01' ###

Target[10.200.200.200_2]: 2:test@10.200.200.200:

SetEnv[10.200.200.200_2]: MRTG_INT_IP="" MRTG_INT_DESCR="FastEthernet0/1"

MaxBytes[10.200.200.200_2]: 12500000

Title[10.200.200.200_2]: Traffic Analysis for 2 -- Switch2

PageTop[10.200.200.200_2]:

Traffic Analysis for 2 -- Switch2

System:	Switch2 in
Maintainer:
Description:	FastEthernet0/1
ifType:	ethernetCsmacd (6)
ifName:	Fa0/1
Max Speed:	100 Mbits/s

WithPeak[10.200.200.200_2]: ym

Options[10.200.200.200_2]: bits

5. run the script from DOS using "perl mrtg mrtg.cfg" command.

For each event (eg CPU, mem, errors etc) you want to monitor you need to know the MIB (mrtg has some built in MIBs).

Very easy to set up, can do it in 30 minutes.

Hope that helps.

Steve

[aluk]

Dear Steve,

Thank you for your kindly advice, but something that make me confuse is that

1) where do i input the MIBs value??

2) is it necessary to use Cisco propertiary MIBs??

and do u have more detail or sample configuration for this, Thanks!

[steve.barlow]

1) MRTG by default will grab traffic stats (in/out) from your interfaces if you use the config I posted first. If you want different stats you enter MIBs as shown here:

# Router CPU load %

Target[cpu.1]:1.3.6.1.4.1.9.2.1.58.0&1.3.6.1.4.1.9.2.1.58.0:test@10.200.200.200

RouterUptime[cpu.1]: test@10.200.200.200

MaxBytes[cpu.1]: 100

Title[cpu.1]: CPU LOAD

PageTop[cpu.1]:

CPU Load %

Unscaled[cpu.1]: ymwd

ShortLegend[cpu.1]: %

XSize[cpu.1]: 380

YSize[cpu.1]: 100

YLegend[cpu.1]: CPU Utilization

Legend1[cpu.1]: CPU Utilization in % (Load)

Legend2[cpu.1]: CPU Utilization in % (Load)

Legend3[cpu.1]:

Legend4[cpu.1]:

LegendI[cpu.1]:

LegendO[cpu.1]:  Usage

Options[cpu.1]: gauge

WithPeak[cpu.1]: ym

This will grab CPU stats from a router. Notive second line (target...), that's where you enter the community string, IP and MIB.

2) You can use any MIB that is recognized by the device. So my cisco router recognizes cisco specific MIBs and industry standard MIBs.

If there is a MIB, MRTG can grab it. But remember the device must understand the MIB to send the trap.

If you would like I can email you an example of a mrtg.cfg file (it's too big to post here) that shows memory, cpu, traffic and input errors from a cisco pix, router and switch.

Steve

[aluk]

Dear Steve,

Actually, i want to monitor the utilization trends of some switch ports on a Catalyst 3548XL. My network configuration briefly describe as follow:

1) Two Catalyst 5500 running HSRP with 20 vlans loadsharing between 2 catalyst

2) Trunking to Distributed layer Catalyst 3548 and 3524 ( Server farm location)

3) Access layer with numbers of cat2924XL (user's workstation)

I would be grateful if u can send me an configuration file about a switch

Thank you for your help!!

George

[aluk]

Steve,

sorry, missing email address in previous message!! You can mail me at skkan.geo@yahoo.com

Thks a lot!

George