[comp.unix.wizards] panic: iinit on booting 4.3bsd miniroot, VAX 11/750

mullen@nrl-css.arpa (Preston Mullen) (08/14/87)

Trying to bring up 4.3bsd on a VAX 11/750, I booted vmunix from the
miniroot on disk ra0; upon specifying ra0* as my root device, I got an
immediate panic: iinit.  Does this imply something like bad blocks in
the swap area of the RA81 disk I am using?

Thanks.

chris@mimsy.UUCP (08/15/87)

In article <8799@brl-adm.ARPA> mullen@nrl-css.arpa (Preston Mullen) writes:
>Trying to bring up 4.3bsd on a VAX 11/750, I booted vmunix from the
>miniroot on disk ra0; upon specifying ra0* as my root device, I got an
>immediate panic: iinit.  Does this imply something like bad blocks in
>the swap area of the RA81 disk I am using?

An iinit panic means that the kernel could not mount the root file
system.  The reason is never mentioned, but it is either an I/O
error (preceded by a message from the appropriate disk driver), a
bad super block, or a nonexistent root device (not the case for
the generic kernel).  The bad super block error could occur if the
standalone copy program and the kernel do not agree as to the
offsets of the `b' partition.
-- 
In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690)
Domain:	chris@mimsy.umd.edu	Path:	seismo!mimsy!chris

haynes@ucscc.UCSC.EDU.UUCP (08/16/87)

In article <8799@brl-adm.ARPA> mullen@nrl-css.arpa (Preston Mullen) writes:
>Trying to bring up 4.3bsd on a VAX 11/750, I booted vmunix from the
>miniroot on disk ra0; upon specifying ra0* as my root device, I got an
>immediate panic: iinit.  Does this imply something like bad blocks in
>the swap area of the RA81 disk I am using?
>
I had the same experience with ra-type disks.  Worked fine with hp
disks.  Someone on the net mentioned that this is the result of upgrading
from 4.2 to 4.3 and using the 4.2 cassette.  So I got a real(tm) 4.3
cassette from Berkeley and have only tried it once, but it seemed to
work as expected.


haynes@ucscc.ucsc.edu
haynes@ucscc.bitnet
..ucbvax!ucscc!haynes

steve@dartvax.UUCP (Steve Campbell) (08/16/87)

In article <7986@mimsy.UUCP> chris@mimsy.UUCP (Chris Torek) writes:
>In article <8799@brl-adm.ARPA> mullen@nrl-css.arpa (Preston Mullen) writes:
>>Trying to bring up 4.3bsd on a VAX 11/750, I booted vmunix from the
>>miniroot on disk ra0; upon specifying ra0* as my root device, I got an
>>immediate panic: iinit.  

>An iinit panic means that the kernel could not mount the root file
>system.  ...  The bad super block error could occur if the
>standalone copy program and the kernel do not agree as to the
>offsets of the `b' partition.

...and a good way for this to happen is by trying to use the 4.2bsd
standalone copy program from the 4.2 distribution cassette.  When our
4.3 package arrived, Berkeley had not enclosed a cassette, presumably
because we were already known to be running 4.2.  I tried using the 4.2
cassette and got the iinit panic.  I grumbled and resorted to manually
bootstrapping the miniroot.  See Appendix B of "Installing &
Operating."  That worked.
						Steve Campbell
						Dartmouth College

torben@uhmanoa.UUCP (08/17/87)

In article <6882@dartvax.UUCP> steve@dartvax.UUCP (Steve Campbell) writes:
>In article <8799@brl-adm.ARPA> mullen@nrl-css.arpa (Preston Mullen) writes:
>>Trying to bring up 4.3bsd on a VAX 11/750, I booted vmunix from the
>>miniroot on disk ra0; upon specifying ra0* as my root device, I got an
>>immediate panic: iinit.  
>
>...and a good way for this to happen is by trying to use the 4.2bsd
>standalone copy program from the 4.2 distribution cassette.  When our
>4.3 package arrived, Berkeley had not enclosed a cassette, presumably
>because we were already known to be running 4.2.  I tried using the 4.2
>cassette and got the iinit panic.  I grumbled and resorted to manually
>bootstrapping the miniroot.  See Appendix B of "Installing &
>Operating."  That worked.
>						Steve Campbell
>						Dartmouth College

I had that problem when trying to upgrade a 780 from 4.2BSD to 4.3BSD about a
year ago.

I kept running into the panic and I was stumped for a couple of hours. The
problem occurs because there was a change in the layout of the RA81 disks
between 4.2BSD and 4.3BSD. This problem only affected RA81's. Bootstrapping
manually is one possible solution. However, if you happen to have at least
two RA81's (and many who have RA81's do since they got them as package deals
from DEC), there's another way.

Say you have drives ra0 and ra1. Use the 4.2BSD standalone copy program to copy
the miniroot to ra1(0,0) and boot that up. When asked for the root partition,
just answer ra1 (NO asterisk). This will give you a running miniroot on the
second RA81. Now use the miniroot to load the real root onto ra0(0,0) and
reboot. You can now proceed with a normal installation. As soon as you can get
to it, make yourself a new floppy (or cassettte as the case may be) and you're
ok.

There are some warnings to people with RA81's in the documentation. But there's
no procedure for overcoming the problem.

					Torben N. Nielsen
					University of Hawaii