The cornucopia of my trip 2 weeks ago also included a Sun SPARCstation 4 - after a general cleaning of the outside, I checked the inside - CPU, memory, floppy: present. HD: missing.
Fortunately, I had one of those proprietary SUN HD holders left and I also had a suitable SCA hard disk in my stash.
Since no CD-ROM was installed in the station, I connected my external Pioneer CD/DVD-ROM - probe-scsi - and everything is displayed properly. Test booted briefly from CD - works.
Next up was the usual suspect, the NVRAM, an 08 type in station 4. I also found this in the pool (and put the removed one in the refurbishment queue). The NVRAM worked perfectly, reprogrammed and the installation could begin.
I chose Solaris 2.6 for the machine - it should run smoothly with its 110 MHz and 160 MB RAM (full configuration).
The installation went smoothly in one go, including the correct detection of the (IBM) hard disk and partitioning.
Reboot - and, surprise ... various unclear error messages Short read - Elf read error - watchdog reset.
Very annoying and somehow inconclusive. The guesses (including those of my club colleagues) ranged from SCSI to HD to NVRAM.
Now began a small odyssey …
- reinstalled a 4 GB WD hard disk again - without success
- repaired the removed NVRAM LIFO-style - without success.
- Tried SCSI settings (parity) - without success.
- Tried CD-ROM settings 2048/512 byte block - without success.
I was about to put the machine on the “Extensive Care needed” shelf when I realized that I had the SCSI cable with the external CD-ROM connected to the machine the whole time.
At that point the IBM disk was back in - so cable off, reboot and ... it worked fine!
Start Solaris 2.6, final configuration, reboot again. All great.
The question remains as to what exactly caused it ... Voodoo :), of course.
The external cable is of very good quality and not too long - so all that remains is the external housing ... although I still don't really understand why the installation went off without a hitch.
VOODOO!