Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!ucbvax!MSU.EDU!08071TCP From: 08071TCP@MSU.EDU (Doug Nelson) Newsgroups: comp.protocols.tcp-ip.ibmpc Subject: Re: Wanted : ethernet protocol list... Message-ID: <9105301356.aa15353@louie.udel.edu> Date: 30 May 91 17:35:42 GMT References: Sender: daemon@ucbvax.BERKELEY.EDU Reply-To: Doug Nelson Distribution: world Organization: The Internet Lines: 19 >Could anyone give me the translation of these ethernet protocols or give a >pointer to a site I could FTP an extensive list of protocols along with their >translation ,as I might discover new protocols on our network and I wouldn't >like to send such a request every week or so ... The list is maintained in the "Assigned Numbers" RFC (RFC 1060, unless it has been republished recently), available from the standard places. But I think that most of the numbers you are seeing are bogus values. Values like 2020, aaaa, and ffff are not legit - aaaa is commonly seen in test packets, often with a source Ethernet address of aa-aa-aa-aa-aa-aa. But I suspect that most of the rest are simply from corrupted packets that happen to have a valid checksum - odds are that 1 in 65,536 corrupted packets will have a good checksum, or maybe more (this would require a detailed electrical analysis to know for sure). If you can count packets of a given type, make sure you're seeing them with a higher frequency than your checksum error rates would account for. Doug Nelson Michigan State University