Article View: comp.dcom.lans.misc
Article #281Re: Does transmitting station stop looking for collision after collision window?
From: Matt Healy
Date: Wed, 16 Dec 1992 16:28
Date: Wed, 16 Dec 1992 16:28
41 lines
2496 bytes
2496 bytes
In article <Bz0vA3.D9G@news.iastate.edu>, john@iastate.edu (John Hascall) wrote: > > [*please* stick to < 80 columns...] > > jim@mivax.mc.duke.edu (James D Dryfoos - Systems) writes: > }I am trying to find out if there is a point during transmission > }of an ethernet packet that the > }transmitting card stops listening to the wire to see if... > > Yes, the ethernet chip(set) will notice the collision. A collision > after 512 bits is a `late collision'. A late collision indicates your A late collision should not occur because any Ethernet card is supposed to *listen* before it starts talking. In their advertisements during the early days of Ethernet, XeroX used to compare this to *polite* people at a cocktail party: you don't interrupt somebody else; you wait for a pause. Thus a collision should only take place when two cards attempt to begin transmitting at about the same time. The so-called "collision window" is the time period during which a collision can occur on a properly-functioning network because collision detection cannot be instantaneous. This is the main reason for having a minimum packet size, a maximum segment length, and the 4-repeater restriction: to ensure a sufficient collision window for detection to work. A late collision means that some device is transmitting when it should *not* be doing so. Properly-functioning devices can detect and report a late collision as an error, but they cannot solve the problem because the device which is the cause has already demonstrated its inability to follow the rules. (of course, in the cocktail party analogy at this point the boor who interrupted would get shouted down...) Matt Healy "I pretend to be a network administrator; the lab net pretends to work" matt@wardsgi.med.yale.edu
Technical Details
Message-ID:
<matt-161292111741@wardmac9.med.yale.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!cis.ohio-state.edu!pacific.mps.ohio-state.edu!linac!uwm.edu!cs.utexas.edu!qt.cs.utexas.edu!yale.edu!news.yale.edu!wardmac9.med.yale.edu!userReferences:
<9DEC199218262989@mivax.mc.duke.edu> <Bz0vA3.D9G@news.iastate.edu>Show Full Headers
Complete Article Headers
Newsgroups: comp.dcom.lans.misc Path: arc1-ca.pugleaf.net!sparky!uunet!cis.ohio-state.edu!pacific.mps.ohio-state.edu!linac!uwm.edu!cs.utexas.edu!qt.cs.utexas.edu!yale.edu!news.yale.edu!wardmac9.med.yale.edu!user From: matt@wardsgi.med.yale.edu (Matt Healy) Subject: Re: Does transmitting station stop looking for collision after collision window? Message-ID: <matt-161292111741@wardmac9.med.yale.edu> Followup-To: comp.dcom.lans.misc Sender: news@news.yale.edu (USENET News System) Nntp-Posting-Host: wardmac9.med.yale.edu Organization: Yale U. - Genetics References: <9DEC199218262989@mivax.mc.duke.edu> <Bz0vA3.D9G@news.iastate.edu> Date: Wed, 16 Dec 1992 16:28:13 GMT Lines: 41