This is just too weird.

I'm trying to bring up a MicroVAX-II.  Unfortunately it has only one
disk, and it appears to be broken, so it has to be diskless.

Most things seem to work: "b xqa0" tickles mopd, which downloads the
booter.  The booter says something about "press any key to interrupt
autoboot 5", but generally doesn't count down from 5 - it started to
once, but only once.  When I press a key and get the > prompt, I tell
it to "boot -s" and it comes up and (since I used the install kernel)
drops into sysinst, root on md0a.

Used to be I could then exit sysinst, mount the diskless area, chroot
to it, and do stuff.  But then I took the machine apart to see what was
really there.  When I put it back together again, I left out the TK50
drive and its controller, moving the DEQNA up the bus.

Now, it all boots fine.  And I can ifconfig qe0 and lo0 correctly.  But
when I try to ping the server, or mount the diskless area...nothing
happens.  tcpdump on the server sees nothing when I ping from the VAX.

But obviously I didn't just render the DEQNA nonfunctional, since it
can boot from it!  Repeatedly!

Any idea what gives?  This hardware is a bit odd already; initially the
DEQNA was in yet a third place, and when it was there, the ROM couldn't
talk to it even enough to boot.

dmesg output (well, "cat /kern/msgbuf" output - there's no dmesg in
md0a) is:

MicroVAX II
realmem = 5234688
avail mem = 2072576
Using 63 buffers containing 258048 bytes of memory.
mainbus0 (root)
cpu0 at mainbus0: KA630
uba0 at mainbus0: Q22
uda0 at uba0 csr 172150 vec 774 ipl 17
mscpbus0 at uda0: version 2 model 3
mscpbus0: DMA burst size set to 4
ra0 at mscpbus0 drive 0: RD51
ra0: attempt to get status failed:  unit offline (not mounted) (code 3, subcode 1)
qe0 at uba0 csr 174440 vec 770 ipl 17
qe0: deqna, hardware address 08:00:2b:05:10:ba
md0: internal 1056K image area
boot device: qe0
root on md0a dumps on md0b

ra0 is the other odd thing.  According to the stickers on the drive,
it's an RD53, not an RD51, and what does this error mean?  I've never
seen it give anything else.

I may try putting the hardware back as nearly as I can exactly the way
it was last "working"...but I can't help feeling that there's something
significant lurking here in the way it can boot from the DEQNA but not
use it once it's up.

					der Mouse

			       mouse@rodents.montreal.qc.ca
		     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B