[mrtg] problem with missed SNMP polls

[Chris Conn]

Hello,

I am quite anxious to find the reason why huge peaks are generated on graphs. 
I have determined that should a poll fail for a target, the following poll will 
generate a huge peak of traffic that was not actually present.  I configured a 
test link which is basically idle, and created a target for it.  The first 3 
polls were normal, I then changed the SNMP read password to something else to 
cause the request to fail.  I then changed it back, and the next poll showed a 
usage of several megabits!

Here is the top of the .log file

1066744812 50623335 35717956
1066744812 175167 123591 175167 123591
1066744523 973 638 973 638
1066744500 973 638 973 638
1066744200 960 629 973 638
1066743900 0 0 0 0

The link never went down and the unit was not rebooted, only the SNMP read 
community was changed to cause the poll to fail.

The following poll modified the .log file as such:

1066745102 50860221 35899136
1066745102 816 624 816 624
1066744812 175167 123591 175167 123591
1066744800 161812 114164 175167 123591
1066744500 973 638 973 638
1066744200 960 629 973 638
1066743900 0 0 0 0
1066743600 0 0 0 0
1066743300 0 0 0 0

Still showing a significant amount of traffic, even though the link is idle. 
The third poll after:

1066745402 51115419 36089590
1066745402 850 634 850 634
1066745102 816 624 816 624
1066745100 7790 5542 175167 123591
1066744800 161812 114164 175167 123591
1066744500 973 638 973 638
1066744200 960 629 973 638
1066743900 0 0 0 0
1066743600 0 0 0 0
1066743300 0 0 0 0
1066743

back to normal.

Any ideas what can cause this?  I am using Linux RedHat9, MRTG 2.9.17 and 
net-snmp 5.0.6.

Thanks,

Chris

--
Unsubscribe mailto:mrtg-request at list.ee.ethz.ch?subject=unsubscribe
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