FWIW - I ran into a similar problem a while back that I decided had to
do with something having to do with the .so files.

I don't remember the details.  I was going from system X to system Y,
where the two systems were 1.3.3 and 1.5.2, not necessarily in that order.

I had system X installed and running, and I decided that it would be
quicker to simply inpack the tarballs right on top of the running
system and reboot system Y.

Oops.

The odd part was that what I ended up with was something similar to the
above.  Many commands didn't work.  I messed around some to try to fix
it, because I did figure out that it had something to do with the
.so files/libraries.  I finally gave up, partly because so few commands
worked.

I ended up re-installing from scratch (from MacOS).  Too bad I didn't
keep notes, because this was not -current, but two released systems.

-dgl-

>To start with, there will not be a snapshot done this weekend....
>
>I currently have my system running on 1.5.2, but it was a tortuous process
>getting there.
>
>For the curious, this is what I've done to try to recover:
>
>As you may recall, my system got into a state where statically-linked
>programs would run, but dynamically-linked ones didn't.  I'm not sure
>it's a kernel problem, as none of the 1.5X kernels that I have access
>to, nor the 1.5Y (neither the one built under 1.5X nor the one built
>clean under 1.5.2) are able to get past the problem.
>
>The first thing I did was to mount a directory off my Indy which contained
>the tarballs from my earlier snapshot.  (I'm kinda regretting my agressive
>pruning of old snapshots. ;-)  I was able to restore from the tarballs,
>but that didn't fix the problem.  So I decided to set up a second disk and
>work from that.