Why do you think stopping MRTG will stop the traffic counters on the devices from accumulating? The only way I see to do this is to develop a process to "remove" the backup spikes. One way is to "synthesize" the data. Although not smart, if you know that the backup spikes will last an hour, just keep on polling through the period when the backups run. Then, after that period has completely finished, take the previous hour's data (prior to when the backups started) and replace the slice while the backups were going on with the previous hours' data. That should give the PHB's the warm feeling that nothing is going on with the network during the period that the backups are running.
There are many other ways, and I'm not even saying this is a good idea, but it will serve the purpose. Paul >>> "Thomas Deliduka" <[EMAIL PROTECTED]> 01/25/06 12:59 PM >>> I have this annoying problem on my network. The backup runs every morning at 4am. When it does, traffic spikes to over 4Mbps on the network for my servers. The backup I'm using doesn't allow me to choose the SAN (faster/non-routable) network to backup for a few servers thus the spike. I can't avoid this, at this time, until the company updates the backup client. I have been running MRTG daemonized and I had the idea of stopping the daemon 5 minutes before the backup and activate it 1 hour later when it's done. However, even though the log evidence shows that the daemon is shutting down and starting up again at the appropriate times, I still have data for the whole hour! So, my question is, if I run it simply as a cron job every 5 minutes instead of between 4 and 5 am, will it still fill in the whole hour that I don't want tracked? -- Unsubscribe mailto:[EMAIL PROTECTED] Archive http://lists.ee.ethz.ch/mrtg FAQ http://faq.mrtg.org Homepage http://www.mrtg.org WebAdmin http://lists.ee.ethz.ch/lsg2.cgi -- Unsubscribe mailto:[EMAIL PROTECTED] Archive http://lists.ee.ethz.ch/mrtg FAQ http://faq.mrtg.org Homepage http://www.mrtg.org WebAdmin http://lists.ee.ethz.ch/lsg2.cgi
