home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!ames!sgi!rhyolite!vjs
- From: vjs@rhyolite.wpd.sgi.com (Vernon Schryver)
- Newsgroups: comp.sys.sgi
- Subject: Re: late collision errors
- Message-ID: <nv8m6t4@rhyolite.wpd.sgi.com>
- Date: 30 Jul 92 23:45:17 GMT
- References: <1992Jul30.190642.29534@aio.jsc.nasa.gov>
- Organization: Silicon Graphics, Inc. Mountain View, CA
- Lines: 31
-
- In article <1992Jul30.190642.29534@aio.jsc.nasa.gov>, strong@sweetpea.jsc.nasa.gov (Renee Strong 283-4257) writes:
- > We keep getting TONS of late collision errors in our console windows. I under-
- > stand that this indicates that there is a major network problem. We know our
- > network runs very slowly. Is there some way I can gather statistics on the
- > late
- > collisions? I've tried using netstat -i, but the collisions are only
- > about 0.3%
- > of the total packets sent and recieved so according to those statistics
- > we don't
- > have a problem.
-
-
-
- Late collisions mean your network is broken.
- They will cause terrible performance (by SGI standards).
-
-
- Typical causes are:
- -execessively long "drop cables" connectint twisted pair
- transceivers to workstations
- -excessively long ethernet coax
- -partially connected computers--as in half-connected cables
- -broken computers.
-
- I know of no way to find such problems, except by turning things
- off until they go away. (after first walking around loooking
- for bad cables, etc.)
-
-
-
- Vernon Schryver, vjs@sgi.com
-