Cheshire Cat Computing

So, here's an interesting one. I'm working on primarily Cisco MDS9500's here, and use MRTG to track port trending and hotspots. So obviously, when I went to routers2, I noticed a tenfold increase. All of a sudden, all of my ports are at or near cap. I broke out more diagnostic tools, and lo and behold, they're nowhere NEAR cap.

Best example is that I have an interface that averages 1.26MB/s outbound, which routers2.cgi reports as 11.62MB/s outbound! Obviously, this is horribly wrong.

I'm using "routers.cgi*Options[iis.totalPosts]: fixunit" in the MRTG config, and "usebigk = mixed" in the routers2.cfg. Any ideas what's going on here, and how to fix it?

Thanks!

The usual reason for this sortof discrepancy is bits/bytes confusion.

Routes2, by default, will display network traffic in bits/sec rather than bytes/sec. This is the 'bits' option in MRTG which normally defaults to 'off' but with Routers2 will default to 'on' for Targets identified as network interfaces. The labels will also be changed to reflect this, by default.

If you want 'bytes' anyway, you can either -

1. Set 'bytes=yes' in the routers2.conf to make everything default to 'bytes', whether network interface or not.
2. Set 'routers.cgi*Options[targetname]: bytes' in your cfg file (of course removing any 'bits' option already defined)

Note that the 'fixunit' option tells routers2 to NOT use any unit prefixes, such as 'k' or 'M' which can sometimes confuse things if you're dealing with very large numbers.

The 'mixed' option for usebigk tells it that 1MB = 1024000B. 'no' would set it to 1000000B and 'yes' would set it to 1048576B (1024x1024).

Steve

_________________
Steve Shipway
UNIX Systems, ITSS, University of Auckland, NZ
Woe unto them that rise up early in the morning... -- Isaiah 5:11