The reason for this is likely that when server B is rebooting, it
initiates the serial ports and sends out a break signal. As this is
connected to server A's console, it brings it down to the Ok> prompt.
(The OS is frozen, typing go at the ok prompt would get it running
normally again.)

/usr/bin/kbd can be used to temporarily change the abort command, and
/etc/default/kbd can be used for permanent changes. Good links for more
info appears to be

* http://www.columbia.edu/~rtt2101/iaoq/#I.D5 - How can I prevent my
system from halting when my terminal server is rebooted?

*
http://docs.sun.com/app/docs/doc/816...D_ABORT&a=view
- man page for /usr/bin/kbd (and /etc/default/kbd)


/ Per


Lorin.Kobashigawa@gmail.com skrev:

> Hi, I have a setup with two SunFire v100s, where they both are
> connected to each other by serial cables. One of them is currently
> having a problem, so I'm trying to diagnose it remotely using tip from
> the other server over SSH.
>
> Server A is fine, Server B is not booting properly.
>
> If i shell into server A, I can use 'tip hardwire' to get into server B
> without any problems, I can drop it to the ok prompt with ~~# (I'm over
> ssh, so i need the extra ~). The problem is that when I try to boot
> server B from the ok prompt, or power it up from the lom prompt,
> something in the boot process is causing server A, which I'm connecting
> from to drop to the ok prompt, and I have to call and have it power
> cycled, since now both boxes are off the internet.
>
> Anyone have any idea if there are specific tip parameters I should be
> setting? I've tried disabling tandem mode, (not sure what that does,
> but it sounded likely).
>
> -Lkb