Mailinglist Archive: opensuse (4053 mails)

< Previous Next >
Re: [SLE] Two Hubs
Are your hubs both set to 10M or 100M? I know it's obvious but...


On Wednesday 08 August 2001 4:50 pm, Paul Abrahams wrote:
> David Grove wrote:
> > I have a problem. Thought you might have a clue or have run into this
> > before.
> >
> > I have two hubs, let's call them A and B. (Physically, A is a big mama,
> > and B is a dinky D-Link.) I have four rooms, A, B, C, and D. (Physically,
> > A is where I work, and B is where I've moved the machines to lower the
> > heat in this small room.) Hub A is in room A, hub B is in room B. I have
> > Cat5 going from rooms B, C, and D to room A to connect the house to a
> > central point.
> >
> > Some background. Two days ago, Hub B didn't exist, and all of the
> > machines were in room A save one. This made it almost too hot to work
> > during summer days in here. So I moved all machines but one into room B
> > last night.
> >
> > There are now 4 machines in Room B on hub B. There is one machine in room
> > A.
> >
> > My intention was to link Hub A to Hub B. I've tried the "uplink" port of
> > hub B as well as a standard port going to Hub A. Neither of them allowed
> > any machine connected to Hub A communicate with any machine connected to
> > Hub B. I have no idea what's wrong with this setup. The cable was
> > straight-through Cat5 and not a crossover. The only solution that I've
> > been able to get working is to directly connect the sole machine in Room
> > A to the hub in Room B using single wire connecting the two rooms. This
> > leaves rooms C and D dead, and no ability to use my laptop which is
> > usually connected to room C.
> >
> > I need to have a hub in both rooms A and B.
> >
> > Any ideas what might be wrong?
> I might be stating the obvious since you clearly know what you're about,
> but have you applied a cable tester to all your cables? It sounds as
> though you might have made some physical changes to them. The cable
> tester I have has a remote unit and a sender (in one configuration), and
> I already found one bad connection with it. Failing that, you might be
> able to test your cables by hooking them directly between machines and
> bypassing the hubs. Another thing I'd try would be interchanging the
> locations of the hubs, if that's possible.
> Paul

< Previous Next >