Greetings,
Hmm... well, I read Alex's and some other people's advice and it mostly made sense. Then we tried some stuff. What I don't get is this: Our numbers are way off the router's counters for the respected interface we are watching. We did a "clear counters [interface]" on our Cisco 7500 router where this particular point to point customer was terminating. We noted the time. It was 1:35 pm PDT. We then watched it for a 24 hour period and got the numbers for bytes moved off of the router interface for that 24 hour period for IN and OUT. This was done with a simple "sh interface foo" Then we fetched data from the .RRD and the numbers for inbound and outbound were off by over 1 Megabyte for just a 24 hour period. Here's the data: Bytes: INBOUND from Router's Counter: 134472313 INBOUND from RRD: 135519035.148 Excel Sheet Method [explained at bottom of this email]: 135855200.42 We were Off: 1046722.148 (1.05MB) ---------------------------------- OUTBOUND from Router's Counter: 217643750 OUTBOUND from RRD: 219438500.976 Excel Sheet Method: 220948884.67 We were Off: 1794750.976 (1.79MB) ------------------------------------ Note using the Excel method: rrdtool fetch... Paste each column in, for each row, multiply by 300, add up the numbers. This is more similar to the method used in Steve Shipway's routers.cgi. The excel sheet method seemed more to like what Alex was suggesting and we were still off. Very confusing. The command I used to grab the Bytes from the RRD was: serengeti# cat working-cmd-day-stdout-OUT ./rrdtool graph /dev/null DEF:outoctets=myrouter.rrd:ds1:AVERAGE CDEF:out=outoctets,8,* PRINT:out:AVERAGE:"%lf" (I had read that the default time frame was 24 hours so this worked fine) I just took this number and divided by 8 to give us bytes. Any ideas as to what this all means would be *greatly* appreciated. Thanks! John -- 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
