Has anyone successfully installed and used OpenBSD on the Celeron-based
Compaq iPAQ PC (they are quite old now but I'd like to reuse it as a
gateway/firewall).
At first, the install process (booting from CDROM) hung/panic-halted a few
times just after USB checks during the boot process but I persevered,
rebooted and tried again. On one occasion it then went on to successfully
run the complete install.
Now, after the installation, it dies during the boot just after the USB
checks again and I can't get past that.
Kunle
--
Don't talk unless you can improve the silence.
It remains as originally delivered except for the addition of a supported
LinkSys USB network adapter (identified as 'axe0'). It ran fine with no
issues (using Windows XP) until I decided to use it as an OpenBSD gateway
box. I elected to use all of 'wd0' for OpenBSD.
Kunle
........
uhub0 at usb0
uhub0: Intel UHCI root hub, rev 1.00/1.00, addr 1
uhub0: 2 ports with 2 removable, self powered
uhci1 at pci0 dev 31 function 4 "Intel 82801BA USB" rev 0x02: irq 10
uvm_fault(0xd05ccc00, 0x0, 0, 1) -> e
kernel: page fault trap, code=0
Stopped at usb_schedsoftintr+0xd: call *0x4(%edx)
ddb>
<http://www.openbsd.org/OpenBSD/faq/faq4.html#getdmesg>
k> ........
k> uhub0 at usb0
k> uhub0: Intel UHCI root hub, rev 1.00/1.00, addr 1
k> uhub0: 2 ports with 2 removable, self powered
k> uhci1 at pci0 dev 31 function 4 "Intel 82801BA USB" rev 0x02: irq 10
k> uvm_fault(0xd05ccc00, 0x0, 0, 1) -> e
k> kernel: page fault trap, code=0
k> Stopped at usb_schedsoftintr+0xd: call *0x4(%edx)
ddb>
FWIW, your problem appears similar to PR kernel/5001
<http://cvs.openbsd.org/cgi-bin/query-pr-wrapper?full=yes&textonly=yes&numbers=5001>
Thanks, I'd already read that. The redirection to a serial port looks like
the option for my environment but I haven't got the right cables. What
cables do you use?
> k> ........
> k> uhub0 at usb0
> k> uhub0: Intel UHCI root hub, rev 1.00/1.00, addr 1
> k> uhub0: 2 ports with 2 removable, self powered
> k> uhci1 at pci0 dev 31 function 4 "Intel 82801BA USB" rev 0x02: irq 10
> k> uvm_fault(0xd05ccc00, 0x0, 0, 1) -> e
> k> kernel: page fault trap, code=0
> k> Stopped at usb_schedsoftintr+0xd: call *0x4(%edx)
> ddb>
>
> FWIW, your problem appears similar to PR kernel/5001
>
<http://cvs.openbsd.org/cgi-bin/query-pr-wrapper?full=yes&textonly=yes&numbe
rs=5001>
I suppose it does except that lacking 'legacy' ports like PS/2 ports, the
usb keyboard attached to the box is useless by the time the 'ddb>' prompt
appears so I can't 'trace/ps/show registers'.
Thinking about it now, I can see that the computer and the OpenBSD
installation might still be active despite the [usb] keyboard being dead. It
is a new install with all the X-related file sets excluded but I did opt to
start sshd by default. This was to be the firat reboot after the
installation.
Would I be able to access and use box via the network even if the usb errors
have effectively killed the 'console'?. That might just be enough until the
usb/kernel-panic issue is resolved.
Kunle
You'll need a null-modem cable, i.e. one where the Tx and Rx lines (at
least) are swapped.
> Thinking about it now, I can see that the computer and the OpenBSD
> installation might still be active despite the [usb] keyboard being dead.
> It
> is a new install with all the X-related file sets excluded but I did opt
> to
> start sshd by default. This was to be the firat reboot after the
> installation.
>
> Would I be able to access and use box via the network even if the usb
> errors
> have effectively killed the 'console'?. That might just be enough until
> the
> usb/kernel-panic issue is resolved.
If the machine is indeed active, yes indeed. Try pinging it.