[ILUG] thoughts on vendors entries

[Timothy Murphy]

On Thu 14 Jul 2005 12:51, Bryan O'Donoghue wrote:
> > Advice sought: when I turned on my second-best laptop today it said
> > =========================================
> > Jul 14 11:17:41 martha kernel: hub 1-0:1.0: over-current change on port 1
> > Jul 14 11:18:12 martha last message repeated 121 times
> > Jul 14 11:19:13 martha last message repeated 244 times
> > =========================================
> > and so on.
> >
> > What sort of error message is that?
> > There should be a competition for the most useless error message,
> > or the most obscure.
>
> Disclaimer : ianaue : i am not a usb expert.

Thank you Brian for dipping into your well of knowledge.
Your comments have indeed thrown a bright light on martha (my laptop).

> This is either a kernel or a hardware bug.

Sorry to be mean, but this reminds me of the time years ago
when I rang a US helpline for advice on a motherboard,
and a guy with a very slow Texas accent -
these were the days when Eircom was really Eircom,
and calls to the States cost 1000 pounds a minute -
said, "What you have there <long pause>
is either a hardware bug <longer pause> or a software bug."

> Your USB port 1, reports an over current "change" state, then the port
> is told to clear this "feature" (I'm not sure if that is analogous to
> switching it 'off forever' or just clearing it 'right now'). ie if
> "clear feature"="switch off forever" in this domain, is not immediately
> obvious to me.

OK thanks - I wasn't really clear what "over current" meant
but I guess that doesn't matter too much.

> In any case, either the clear command fails due to a software bug or
> hardware bug, or the clear command succeeds, but for whatever reason,
> the condition gets reasserted.
>
> To find out for sure, remove whatever is in port 1.

Actually there is nothing in my USB port -
the hub is internal,
and there should be an internal Bluetooth connection to the hub,
but this has never worked since I bought the laptop over eBay
from a very smooth guy in Birmingham UK.
(I was going over the UK anyway,
and he treated me to a slap-up Indian meal when I collected the laptop,
which I must admit made me suspicious.
However, that was 3 years ago, and it has worked fine since.
Who needs Bluetooth anyway when we have WiFi?)

> If the error persists, upgrade to the latest vanilla kernel.
> If the error still persists... perhaps try out a 2.4...

I did actually try running a couple of kernels
(I'm not old enough to have a 2.4 kernel),
and the message was the same.
I looked back over my /var/log/messages
and the error messages seem to have started today.
Actually the only time they are very noticeable
is when booting, as the error comes up every second.

> If at that point the error still persists (sans misbehaving device in
> port 1 for all previous cases) ... then it _does_ probably mean that for
> whatever reason USB port 1, on your laptop, reports over-current change
> for no good reason... and is probably broken, somehow... due to, black
> magic... or a field effect/martians maybe.

Actually, I don't think any USB devices work on this computer anyway,
_even under Windows-2000_,
so I guess I might as well just remove the USB modules from the kernel.
[Maybe I ought to open the box, to see if there is a real bug in there,
I mean one with 6 legs.]

Who needs USB anyway, when we have PCMCIA?
The old ways are the best ways.
The whole world has gone to the dogs since Python came in.


-- 
Timothy Murphy  
e-mail (<80k only): tim /at/ birdsnest.maths.tcd.ie
tel: +353-86-2336090, +353-1-2842366
s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland