Re: [IPFIX] Hash-based filtering config parameters

["Zseby, Tanja" <Tanja.Zseby at fokus.fraunhofer.de>]

Hi all,

in general the parameters are configurable (section 7.2 of RFC5475) in order to support different (proprietary) hash functions. But for the usage of the recommended BOB function these parameters have to be filled with the according values and are fixed (see section 6.2.4.1. of RFC5475 which is in line with RFC 5476).

Kind regards
Tanja


> -----Ursprüngliche Nachricht-----
> Von: Benoit Claise [mailto:bclaise at cisco.com]
> Gesendet: Mittwoch, 21. Oktober 2009 10:54
> An: Gerhard Muenz
> Cc: ipfix at ietf.org; Zseby, Tanja; DUFFIELD, NICHOLAS G (NICK); me;
> Aamer Akhter (aakhter)
> Betreff: Re: Hash-based filtering config parameters
> 
> Gerhard,
> 
> >
> > Hi Tanja, Benoit, all,
> >
> > I have two questions regarding the configuration of hash-based
> filtering.
> >
> > 1) Are IP header fields/bits fixed or configurable?
> >
> > RFC 5475, Section 7.2 says:
> > http://tools.ietf.org/html/rfc5475#section-7.2
> >
> >    Case hashing:
> >       - Hash Domain (input bits from packet)
> >            - <Header type = IPv4>
> >            - <Input bit specification, header part>
> >            - <Header type =  IPv6>
> >            - <Input bit specification, header part>
> >            - <payload byte number N>
> >            - <Input bit specification, payload part>
> >
> > => header bits seems to be configurable
> >
> >
> > RFC 5476, Section 6.5.2.6 says:
> > http://tools.ietf.org/html/rfc5476#section-6.5.2.6
> >
> >    In Hash-based Selection, a Hash Function is run on IPv4 traffic.
> The
> >    following fields MUST be used as input to that Hash Function:
> >
> >       - IP identification field
> >
> >       - Flags field
> >
> >       - Fragment offset
> >
> >       - Source IP address
> >
> >       - Destination IP address
> >
> >       - A number of bytes from the IP payload.  The number of bytes
> and
> >         starting offset MUST be configurable if the Hash Function
> >         supports it.
> >
> > => header fields seems to be fixed.
> I would be tempted to believe the protocol. So RFC5476.
> 
> I'll let Tanja or Nick answer the question below.
> 
> Regards, Benoit.
> >
> >
> >
> > 2) Is the output range of the hash function a configuration parameter
> > or just an invariant property of the hash function itself?
> > If yes, should it be configured as min and max or as length and
> bitmask?
> >
> > RFC 5475, Section 7.2 says:
> > http://tools.ietf.org/html/rfc5475#section-7.2
> >
> >       - Hash Function
> >            - Hash Function name
> >            - Length of input key (eliminate 0x bytes)
> >            - Output value (length M and bitmask)
> >            - Hash Selection Range, as a list of non-overlapping
> >              intervals [start value, end value] where value is in
> >              [0,2^M-1]
> >            - Additional parameters are dependent on specific Hash
> >              Function (e.g., hash input bits (seed))
> >
> > => output value length and bitmask seem to correspond to the output
> > range defined by hashOutputRangeMin and hashOutputRangeMax
> >
> > Thanks,
> > Gerhard
> >