> Hi,
> 
> > Beronet's manual is stating that the LEDs turn red when the driver
is
> > loaded but the card is not physically connected to the NTBA. When
> > connected properly they should turn green. Here it seems different:
The
> > LEDs indicate green no matter whether connected or not. Please
> > understand I was about to make my first ISDN-call with a computer,
so I
> > wouldn't spend too much interest in that matter. But if anyone
> > (Daniele?) could explain the correct status indications to me, I'd
> > appreciate it.
> 
> The led color is the opposite you are used to:
> 
> Off for disabled port
> Green for enabled port with deactivated layer 1
> Red for enabled port with active layer 1
> Fast flashing green/red for transitioning layer 1
> Flashing off for traffic on the D channel
> 
> I've already received some request to swap red and green colors and
I'll
> probably do it.

Since I assumed it's the way you explained above, at least /me could
stick with your definitions. Though people might get confused with the
meaning of a red indication. I couldn't help thinking of an error...

> > Back to the prob. As soon as I start to dial a number on my
SIP-phone
> > that'd be routed via vISDN the FC4-system freezes immediately.
> 
> Ok, before going into deep debugging make sure you are using the
latest
> modules (today's snapshot) and, more important, that you don't have
> multiple
> copies of them.

Then I'll get the latest snapshot. I simply thought debugging with the
existing one would at least give more compareable results in terms of a
possible bug track-down.

> Look into /lib/modules/`uname -r`/extra, you should see a directory
for
> each
> module, if you see other modules or strange directories containing
vISDN
> modules please delete them and rerun depmod -a.

Ok, everything seems fine here. At least from the file-structure

> I've changed the way modules are compiled and installed and as a
result
> you
> may have multiple copies of the modules in the modules dir.

Seems that changes happened before my snapshot from Feb 10th. Nothing
multiple in there.

> Then, if you still experience freezes, the most straightforward
debugging
> technique is to connect your machine to another box via a null-modem
> cable.
> 
> Run minicom on both machines to test the connection.
> 
> Boot the * machine appending "console=ttyS0,9600" to the kernel
> parameters.
> 
> On the other machine run minicom at 9600,8,N,1 and you should see some
> boot
> messages. Change 9600 to 115200 if you prefer.
> 
> Finally make the * box hang, if you're not completely out of luck, the
> OOPS
> leading to the freeze should appear on the serial console. 

OK, all I need is a second linux machine then ;-)

>Post it here
> and
> I'll try to "decode" it.

I'll do my best!

> Anyway, IRQ 58 is not wrong, some APIC is able to handle more than the
> usual
> number of IRQ lines, 

I just found out that booting the system with noapic has no effect on
the behaviour at all. Only the assigned IRQ changes. The problems
persist.

"NUM CHANS: 0" is an old debugging message I've
> forgot
> in, you may ignore it.

And ignored it is then.

THX
TOM



> Bye,
> _______________________________________________
> Visdn-hackers mailing list
> [email protected]
> https://mailman.uli.it/mailman/listinfo/visdn-hackers

_______________________________________________
Visdn-hackers mailing list
[email protected]
https://mailman.uli.it/mailman/listinfo/visdn-hackers

Reply via email to