Manual for users

This manual's purpose is to give for the network engineers the exact idea how to run this prototype. We once again warn of the low power of this tool, that it is mostly good for proving of some concepts and managing minor ranges of LAN. We assume the user is familiar with HTML processing otherwise we want to recommend him The Beginner's Guide to HTML

1. Installation

First copy the file installsnmp.pl into your cgi-bin directory and edit the file install.html to match it in the form. Then fill the install.html form. Load install.html into your HTML browser.
You are required to fill in all the fields there are in the form.

SNMPWALK - your snmpwalk with the full path like '/usr/OV/bin/snmpwalk'
HOST - your host with the full path like '/bin/host'
SNMPRARG - your argument for snmpwalking into the routers - consult your routers MIB for more information.
SNMPHARG - your argument for snmpwalking into the hubs - consult your hubs MIB for more information.
ROUTERS - the file containing your routers list.
HUBS - the file containing your hubs list.
SNMPWDIR - the directory where your snmp.ini is going to reside.
OTHERDIR - the directory where your data files will reside.
KEY - the main key for the data tables ('mac' recommended).
FILES - the files for the collection before arrangement according to KEY

After you have filled the forms you should press INSTALL and the required configuration file will be created.

Then copy all the .pl files files, the hubs list and the routers list into the place specified by OTHERDIR, and the files squery-cgi.pl, and queries.pl also into your cgi-bin directory. Now, the installation is complete.

Data collection

You are recommended to run them first from the command line and verify they work correctly. Estimate the time they take.
Then to schedule them to some time of heavy load in LAN in order to get as much nodes as possible into the database. The order of running should be from your selected main key down the data chain tree and the time span between the collecting scripts should be some five minutes plus your estimated time.

After that schedule the arranging script 'cfg.pl'

The collection engine is scheduled completely now.

Queries through Command Line

The files for queries are: squery.pl for the single query and mquery.pl for a ranged one.

squery.pl:

# Usage: squery.pl h|i|m|n|u himnu|a item
#        h stands for hub, i for ip, m for mac, n for name, u for user,
#        a for all possible data.
# It is possible that some data would be unaccessible. Then nothing would
# be printed for it.

mquery.pl:

# Usage: mquery.pl h|i|m|n|u himnu|a item[,item2]
#        h stands for hub, i for ip, m for mac, n for name, u for user,
#        a for all possible data.
# It is possible that some data would be unaccessible. Then nothing would
# be printed for it.

Queries through Graphic Interface

Load the file squery.html into your HTML browser. Fill in all the parts of the form correctly:

Enter the item of the query - a name, an IP number etc.
Select whether the query is simple or ranged.
Select what type of data your argument belongs to.
Select what types of data about it you want to know.
Send your query by pressing QUERY.

romm@empire.tau.ac.il

Last modified: Thu Jun 5 07:26:09 1997