On Thu, 30 Jan 2003, Blaz Antonic wrote:

> As i said earlier - since serial console attached to it was
> unavaliable i used telnet sessions to log on. There was no mention of
> kernel panic in telnet window but tehn again i'm not sure if they'd
> pop up there at all.

The kernel isn't going to send output from a panic over a networked
telnet session.  If you think about it a bit, it should be pretty clear
as to why doing so would be problematic.

Without information about the specifics of the panic, it's difficult
to know what's going wrong.  It could be a bug in the kernel, or it
equally, it could be that you've something like a flaky SIMM.  Maybe
when you're running big compiles or multiple terminal sessions, your
memory usage gets to the point where those processes start using that
memory?  But these are only speculations, possible scenarios, which
may or may not reflect reality.

If you've another system, say a PC with some spare serial ports on the
back of it, I'd recommend getting or making a null-modem serial cable to
run between it and your VAXstation.  Then you can run terminal emulator
software on the PC to access the VAX console over the serial port.
(i.e minicom, cu, Telix, kermit, Hyperterminal, etc.)

> BTW: with all the gimmicks VAXen have DEC engineers couldn't leave one
> HALT level for automatic reboot ? It boots of DKA0 after powerup but
> there is no way to have it boot after kernel crash, is it ?

There is, and I believe some other people already mentioned it their
replies.  IIRC, one of the conditions for it working properly is that
all of the startup tests pass (or at least pass without any serious
complaints?)

-brian.