Bug in ramon?

Carter Bullard carter at qosient.com
Fri Sep 5 11:00:50 EDT 2003

Hey Jesper,
   This is a feature so that we don't miss anything.  With
any aggregation operation, [ragator, ramon, rabins] if the
addition of the next record would roll over a counter,
we spit it out the accumulator immediately and then continue
aggregating records.  So the counts are accurate, but they
end up scattered in multiple records.

   One thing to do is to write your own perl script to eat
the output of ramon() letting perl do the 64-bit variable
thing.  Ragraph does this type of operation, but its buried
deep in the script.

   One thing that may help is to pipe the output of ramon()
to rasort() sorting on the 'saddr field', so that all the
records for a given address are together.  While that messes
up 'top talker' like reports, it at least will show you
those entries that have multiple records.

   64-bit integrated support will be a major release number,
so it probably won't happen this year.  More likely
contributed perl scripts will happen before that!!!


> -----Original Message-----
> From: owner-argus-info at lists.andrew.cmu.edu
> [mailto:owner-argus-info at lists.andrew.cmu.edu] On Behalf Of
> Jesper S. Jensen
> Sent: Friday, September 05, 2003 5:52 AM
> To: argus-info at lists.andrew.cmu.edu
> Subject: Re: Bug in ramon?
> Russell Fulton wrote:
> >>First of all, I hope this is the right place to ask, if not
> please let
> >>me know where. :-)
> >
> > Yup this is the right place :)
> Hehe good, I hoped I was right, but wasn't 100% sure.
> >>32bit barrier
>  >
> > This is a feature.
> Good one. :-)
> Carter may have plans to extent the counters to 64 bits.
> Ok sounds good, any one know when we might expect this to happen?
> --
>    Jesper S. Jensen
> Basisnet og Sikkerhed
> Uni-C - Århus, Danmark
>     +45 8937-6666

More information about the argus mailing list