MRTG can handle single roll-overs just fine... its when the counter rolls 2 or more times within a polling interval that there is a problem. I don't remember the specific numbers, but if you get over 180 Mbps or maybe it was over 200 Mbps, your counters will reset twice within 300 seconds using a 32-bit counter. (I'm to lazy and tired to do the math to find Try to use 64-bit counters if they are available on high speed links, or use RRDTool and poll more frequently on devices that don't have 64-bit counters. I'm currently doing a 1-minute interval on some gigabit links and haven't had counter wrap problem since, even doing a sustained 350 Mbps. However, I haven't measured the added load of that frequent polling. YMMV.
If you have 64-bit counters, check using SNMP2 in the MRTG Reference documentation. (:::::2) Eric Brander ACS Texas CHIP Account Sr. Communications Engineer - Information Systems Department 512.336.3331 Eric dot Brander at acs-inc dot com -----Original Message----- From: Jim McAtee [mailto:[EMAIL PROTECTED] Sent: Monday, January 13, 2003 2:42 PM To: [EMAIL PROTECTED] Subject: [mrtg] Counter Rollovers How does mrtg handle counter rollovers - or does it? I notice that on a number of very busy interfaces (switch ports) that I periodically get "dropouts", where the graph will go to zero. I'm wondering if these are evidence of the counters rolling over and starting again from zero. Jim -- Unsubscribe mailto:[EMAIL PROTECTED] Archive http://www.ee.ethz.ch/~slist/mrtg FAQ http://faq.mrtg.org Homepage http://www.mrtg.org WebAdmin http://www.ee.ethz.ch/~slist/lsg2.cgi -- Unsubscribe mailto:[EMAIL PROTECTED] Archive http://www.ee.ethz.ch/~slist/mrtg FAQ http://faq.mrtg.org Homepage http://www.mrtg.org WebAdmin http://www.ee.ethz.ch/~slist/lsg2.cgi
