On 2013-03-24 09:21, Martin Husemann wrote:
On Sun, Mar 24, 2013 at 03:22:43AM +0100, Johnny Billquist wrote:Correct. I worked around the "problem" right now by just dd:ing the first block of xxboot onto the disk, and then restore the disklabel. But that is maybe less than trivial for someone who is new to the whole system.It looks to me like installboot puts all of xxboot starting at sector 1, and only fills in magic (keeping the actual disklabel) in sector 0.
Right. However, some of that magic is values that can be different things, and installboot don't agree with block 0 in xxboot.
Can you zero the disk (image) again and provide hexdumps of the first two sectors for the working and the not working state?
The non-working version can be examined in any bootable cd that is available. The bootable version is basically xxboot.
I didn't have old enough disklabel sources around, will check there.
Old versions of disklabel copied pretty much all of block 0 of xxboot as well, just going around the area used by the disk label itself.
If you really need me to send you block 0 of a working and non-working boot block, I'll do that.
However, the "non"-working version do work on MicroVAXen. As I mentioned at the start, the bootblock as produced by installboot just do not work on VAX-11 machines, which have a loadable VMB (or in the case of the 11/750 have a very primitive boot prom).
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt%softjar.se@localhost || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol