Re: Nasty little network problem

Toby <tkb9@xxxxxxxxxxxx> · Sun, 25 May 2003 08:40:02 -0400

Harry Putnam wrote:

Ed Greshko <Ed.Greshko@xxxxxxxxxxx> writes:
 BROADCAST=192.168.0.255

                    ^^^^
                    ^^^^
Is this a typo or how it is really configured??

If you look at the ifconfig output above you can conclude that this is
how it is configured....incorrectly configured of course.  :-)

It wasn't a typo and infact the error existed on 2 machines on that
hub.  However it doesn't appear to have been significant.  Both have
now been corrected and ran `service network restart' on both machines.
So to summarize quickly:  4 machines on 1 hub all in 192.168.1.0

network.  All with broadcast set correctly.  All closely examimed for

other errors like having network set to 192.168.0.0 or something.  
I also found one that had that error.
Now after a close examination I believe all have correct settings.
Further 3 of four were able to talk to each other even with the
errors.  M2 could not and still M2 cannot.
Now here is some more info that really confuses me.
Unplug M1 and M2 ethernet cables from hub1.  Plug them into hub2
which is not connected to anything else.  Now M1 and M2 can talk to
each other.  (With no further network changes).
Plug them back into hub1 and they can't even with switching around
port holes and cables.   One might think a switch/hub would act like
that but it isn't supposed to.   Further, 3 of the machines on hub1
can all talk to each other.
Again I emphasis that I have switched port holes and cables around
several times.  Tested by unplugging and reconnecting to hub2, seeing
the results described above.
I've collected a set of data from the problem machine.  Wrote it to
floppy and insert it here.
But to describe the setup only in part for posters ease in seeing the

setup:       
                    INTERNET

                        |

                 (MORE ROUTERS and cable modem)

                        |

    M5 (rh7.1) eth1 is 192.168.1.6 eth0 is 192.168.0.5 
                        |
                        |
    ======================  HUB1  =====================
    |            |             |                   |
192.168.1.7  192.168.1.5   192.168.10         192.168.1.11
   M1         M2            M3                   M4
                ==== HUB2 ==== (not in use here)
Pulling M1 and M2 then reconnect to hub2 (and nothing else) and the
problem on M2 disappears.  It can talk to M1.  Very puzzling.
====================================
cat /etc/sysconfig/network:
NETWORKING=yes
HOSTNAME=hgp.local.lan
GATEWAY=192.168.1.6

With things connected as they are supposed to be, have you watched the 
what the gw M5 (I assume it's a gw ) is doing? It may be taking the 
packets & throwing them out on 192.168.0 net, which is the wrong place 
of course.
Personally, I would just remove completely the NETWORK & BROADCAST lines 
out of the ifcfg scripts. You don't need them here - NETMASK 
255.255.255.0 perfectly describes the network on its own.

-toby

--
Psyche-list mailing list
Psyche-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/psyche-list