On Tuesday 10 June 2003 14:30, you wrote: Hi Haris, > > The way my front end works is, it asks you IP address ,community string, > service to monitor, and index no (if needed) than it automatically:
Well, my frontend works in this way. 1) Discover the object that we will monitor -you need to provide IP address, community, port- Object can be a router, switch, server, etc, that the frontend recognize. 2) Discover what OID and index the frontend will poll.... for example the frontend will examine automatically how many interfaces you have and if his status is up start polling. For interfaces i have the following variables: Bits, Bits In, Bits Out, Bytes, Bytes In, Bytes Out, Errors, Erros In, Errors OUt, Bandwith Utilization%, Discarted frames, Frames In, will be some more :-)) 3) A 5 minute cron is configured to start the poll. This will save the data a database and next to an .rrd. The name of the rrd is determined by a table index. ( I can be more precise if you want how internally the frontend work) > 3. Creates graphs every 5 min (not a very good idea :) With my interface you can create the graph when you need, and specify the date from and date to. Check this example: http://aleg.freeshell.org/example.gif > > Note, currently available service_to_monitor={Network Utilization, CPU > load, Interface Error, Latency, Load Balancing} > The variables will depend on the agent, a server with net-snmp will have some variables that a switch will not. The good news is that I can expand the numbers of variables, just I need some knowledge of the mib) ( I am CC the list for information to another interested users ) --- Alejandro -- Unsubscribe mailto:[EMAIL PROTECTED] Help mailto:[EMAIL PROTECTED] Archive http://www.ee.ethz.ch/~slist/rrd-users WebAdmin http://www.ee.ethz.ch/~slist/lsg2.cgi
