Hello...
I installed ntop on a Debian etch system over the weekend, and so far
everything seems to be working normally, after a few caveats. I do
have two questions I've run into, though...
First off, I'm using sflow exports from my firewall (another Debian
box running pmacctd) as ntop's data source. I could only get this to
work accurately if I turned off pmacctd's sampling interval; as
mentioned earlier on this list, setting a sampling interval caused
ntop to only look at every xth sflow packet, which is obviously broken:
http://listgateway.unipi.it/pipermail/ntop-misc/2005-May/000296.html
There are posts where people have mentioned hard-coding the sampling
rate instead of reading it from the exporter, but I believe that will
mess up the multipliers as well...
Is there a fix for this in CVS to date? If so, will that fix be
rolled into a point release soon, or am I better off grabbing the
latest CVS release?
Next question is, I have the rrd plugin enabled, and after running
for a few days, my /var/lib/rrd directory now is 3.4Gbyts of disk
space. Other than playing with the historical dump values, is there a
way to limit the amount of disk space the rrd files can occupy? If I
create a separate partition for rrd data, will rrd play nice when
that partition becomes full?
Thanks in advance, and my apologies if these questions are available
elsewhere...
-Chris
_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop
