Article View: comp.dcom.lans.ethernet
Article #1004Re: le0: Transmit late collision - net problem?
From: Andrew Boyd
Date: Fri, 18 Dec 1992 19:49
Date: Fri, 18 Dec 1992 19:49
21 lines
1346 bytes
1346 bytes
In article <1992Dec16.181735@is.morgan.com> sergei@is.morgan.com (Sergei Poliakoff) writes: >le0: Transmit late collision - net problem? >Does anybody have any additional information, idea or pointers Too many repeaters/hubs/concentrators between any two transmitters. Excessive cable lengths (eg more than 185m of thin coax or 500m of thick coax in a segment). Out-of-window collisions mean that small packets can get lost, which makes the hardware guys feel bad. This incurs higher-level timeouts, and hurts your performance, but since software guys don't trust the hardware guys not to lose packets anyways, any protocol will retransmit a packet several times until an ACK comes back. Unless you've got a pathological case, where all your small packets always get "lost". Sorta like your bread always landing jelly side down. ---- #include <std.disclaimer>
Technical Details
Message-ID:
<6mbrzrk@quantum.on.ca>Path:
arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!cs.utexas.edu!torn!nott!uotcsi2!revcan!quantum!aboydReferences:
<1992Dec16.181735@is.morgan.com>Show Full Headers
Complete Article Headers
Newsgroups: comp.dcom.lans.ethernet Path: arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!cs.utexas.edu!torn!nott!uotcsi2!revcan!quantum!aboyd From: aboyd@quantum.on.ca (Andrew Boyd) Subject: Re: le0: Transmit late collision - net problem? Message-ID: <6mbrzrk@quantum.on.ca> Date: Fri, 18 Dec 92 19:49:54 GMT Organization: Quantum Software Systems References: <1992Dec16.181735@is.morgan.com> Keywords: SunOS, twisted-pair Ether, Synoptics Lines: 21