As previously posted, reseating the boot ROMs in their sockets cured the
corrupted boot loader issue. Got lucky there :)
In my system there are two RL02 drives. I will call them "Top" and "Bottom"
here, after their physical locations.
I put the OS/8 pack that had been acting strangely in the bottom drive
(faulting the disk and halting the 8/A when OS/8 was interrupted with
Ctrl-C) in the top drive, and switched unit plugs.
It works perfectly there! Ctrl-C returns to the "." prompt as expected.
The pack I formatted also still reads four 4000+ blocks partitions when
switched to the bottom drive.
As before, the AJRLHA diagnostic (Seek/Fctn) still faults and prints errors,
but only at the start of testing the drive. Thereafter that drive shows no
errors. Then the program will switch to the other drive and the same problem
occurs. Specifically (edited and reformatted for clarity, with results
interpreted from the RL02 user manual):
(Ready light goes out, Fault light on)
STATE NOT 5 AFTER SEEK WITH 0 DIFFERENCE (this error and register dump
prints twice in succession. State 5 = Lock-on, keeping on track).
WD1 0317 (lower head, heads not out, spin-down)
WD2 0204 (write data error, write gate error). This one worries me. WDE is
"write gate on but no transitions on write data line". The drive should
never be trying to write when there is no data being sent, and this
diagnostic program does not do any writes to disk! So a fault, head
retraction and shutdown is the proper response to this fatal error. I'll
have to put a logic analyzer on the appropriate bits and see why it thinks
it's supposed to be writing... or where the command came from.
ERROR FLAG SET
ER 2000 (Operation incomplete within 200 ms). Probably because the drive
shut down when the WDE error occurred.
CB 0003 (Seek)
(Fault light goes out)
ER NOT AS EXPECTED BUT ERROR FLAG NOT SET (prints this block twice)
WD1 0235 (heads out, lock-on, keeping on track)
WD2 0002 (Volume check bit, no errors)
ER 0003 (Drive Error and Drive Ready)
CB 1004 (8-bit mode, Read Header)
BAD STATUS RECEIVED FROM DRIVE (also prints this block twice)
WD1 ACTUAL 0235 (heads out, lock-on, keeping on track)
WD2 ACTUAL 0002 (volume check bit)
WD1 EXPECTED 0234 (heads out, seek - track counting)
WD2 EXPECTED 0000 (no errors)
Some of these errors look like a "cascade"... because the first false-write
error occurred, the others are flagged before the drive can become ready
again.
Then it goes on to run for about 10 mins, flickering the Ready light
constantly so it's very dim, obviously accessing as fast as it can.
No further errors.
Next the program switches to drive 1 and the results are identical, except
the "STATE NOT 5" message has WD1 0217 (upper head, heads not out,
spin-down).
So only the Seek diagnostic is giving errors... the AJRLIA diagnostic
(Read/Write) and 'KA (Performance Exerciser) do not show errors. Writes the
entire pack and reads it with no errors. I suspect it has automatic error
retry but without the source code, that is speculation.
Anyhow it may be time to break out the scope and RL02 service manual and
check some settings and waveforms...
But I still think this is actually a problem with the RL8A controller and
not *both* RL02's, since both drives return the same errors, and never had
problems like this before...
Any help making sense of this mishmash of errors would be greatly
appreciated :)
-Charles
