24 Oct
2012
24 Oct
'12
6:39 p.m.
I have been trying to get a microvax II fully functional over the last few weeks and I
seem to have a roadblock. The system used to boot and allow logins, but now after
connecting a VCB-02 cable to display, keyboard and mouse, it has become more and more
unreliable. It may be possible that some assistance I got from a colleague might have
helped to cause this situation, but I think its unlikely. When the system was booting
properly, it wouldn't bring up decwindows even when the display, kb, and mouse were
attached to the bulkhead cable (bc18z). I was asked to check the WINDOW_SYSTEM variable
within sysgen to determine if the current value was anything other than 1. In this case
it was 2. He asked me to change the value to 1, and then to write it to current and
active, then complete an autoconfigure all. I don't feel this is a factor, because
when I boot into conversational mode and set the WINDOW_SYSTEM variable to 0 the system
still faults when I perform a @sys$system:startup. In conversational mode, when I launch
the startup.com script, the kernel panics as it does when it boots in normal mode. I was
hoping to edit the startup.com script (or move it to a safe location and place a partial
startup script in its place), but I'm not sure what parts I need to have. If I dont
run the startup script, I can't run anal/crash because of protection issues.
At first when connected to a display the system would kernel panic during boot. It would
boot WITHOUT the display being connected, however, so I continued using it without the
display. Unfortunately, now it appears to panic every boot, and I can't get to a
prompt to run any diagnostics such as anal/crash. No user based changes were made to the
system between boots. The real brain teaser is that it doesn't always crash during the
same process load, I have been trying to get a microvax II fully functional over the last
few weeks and I seem to have a roadblock. The system used to boot and allow logins, but
now after connecting a VCB-02 cable to display, keyboard and mouse, it has become more and
more unreliable. The real brain teaser is that it doesn't always crash during the
same process load, sometimes its OPCOM, sometimes its STARTUP, sometimes its DECW$STARTUP
as it is below. Anyone have any insight on how to proceed? I'm not a very
experienced user. A crash example is below:
KA630-A.V1.3
Performing normal system tests.
7..6..5..4..3..
Tests completed.
>> boot
2..1..0..
VAX/VMS Version V5.2 Major version id = 1 Minor version id = 0
$! Copyright (c) 1989 Digital Equipment Corporation. All rights reserved.
The VAX/VMS system is now executing the system startup procedure.
**** FATAL BUG CHECK, VERSION = V5.2 MACHINECHK, Machine check while in kernel mode
CRASH CPU: 00 PRIMARY CPU: 00
ACTIVE/AVAILABLE CPU MASKS: 00000001/00000001
CURRENT PROCESS = DECW$STARTUP
REGISTER DUMP
R0 = 00000000
R1 = 7FF38058
R2 = 00000000
R3 = 00000268
R4 = 00000000
R5 = 00000000
R6 = 00004152
R7 = 00000001
R8 = 0000026C
R9 = 00003FDC
R10= 000042DC
R11= 00006CF0
AP = 7FF380FC
FP = 7FF380C0
SP = 803AF1E0
PC = 8028D977
PSL= 04DF0008
KERNEL/INTERRUPT/BOOT STACK
803AF1E8 0000000C
803AF1EC 00000082
803AF1F0 7FF38058
803AF1F4 02000002
803AF1F8 00001534
803AF1FC 03C00001
LOADED IMAGES
[SYSMSG]SYSMSG.EXE 800E0600 8010A400
[SYS$LDR]SYSLDR_DYN.EXE 80195200 80196C00
[SYS$LDR]DDIF$RMS_EXTENSION.EXE 80197000 80198200
[SYS$LDR]RECOVERY_UNIT_SERVICES.EXE 80198200 80198E00
[SYS$LDR]RMS.EXE 8010A400 80130200
SYS$NETWORK_SERVICES.EXE 80130C00 80130E00
SYS$TRANSACTION_SERVICES.EXE 80131400 80131600
CPULOA.EXE 80131C00 80133200
LMF$GROUP_TABLE.EXE 80133C00 80134600
SYSLICENSE.EXE 80134A00 80136000
SYSGETSYI.EXE 80136600 80137C00
SYSDEVICE.EXE 80138000 80139600
MESSAGE_ROUTINES.EXE 80139C00 8013C800
EXCEPTION.EXE 8014CC00 80155400
LOGICAL_NAMES.EXE 80155C00 80157600
SECURITY.EXE 80157C00 80159400
LOCKING.EXE 80159A00 8015C200
PAGE_MANAGEMENT.EXE 8015C800 80164800
WORKING_SET_MANAGEMENT.EXE 80165000 80169600
IMAGE_MANAGEMENT.EXE 8016A000 8016CA00
EVENT_FLAGS_AND_ASTS.EXE 8016D000 8016E400
IO_ROUTINES.EXE 8016EA00 80175C00
PROCESS_MANAGEMENT.EXE 80176200 8017E800
ERRORLOG.EXE 8018BC00 8018C600
PRIMITIVE_IO.EXE 8018CC00 8018DC00
SYSTEM_SYNCHRONIZATION_UNI.EXE 8018E000 8018F600
SYSTEM_PRIMITIVES.EXE 8018FC00 80193000
**** STARTING MEMORY DUMP....
ERROR WRITING 127 BLOCK(S), STARTING AT VBN 5471 - STATUS = 00000054
25 Oct
25 Oct
8:17 a.m.
At 06:39 PM 10/24/2012, Kevin Reynolds wrote:
.... A crash example is below:
**** FATAL BUG CHECK, VERSION = V5.2 MACHINECHK, Machine check
while in kernel mode
...
KERNEL/INTERRUPT/BOOT STACK
803AF1E8 0000000C
803AF1EC 00000082
803AF1F0 7FF38058
803AF1F4 02000002
803AF1F8 00001534
803AF1FC 03C00001
Here's what that machine check stack frame means:
803AF1E8 0000000C 12 bytes of parameters
803AF1EC 00000082 Machine check code = Write bus
error, virtual address
803AF1F0 7FF38058 Virtual address
803AF1F4 02000002 Internal CPU state.
803AF1F8 00001534 PC
803AF1FC 03C00001 PSL
It seems that you are getting hard memory write faults on the stack.
Maybe something is hanging up the Q-bus, or you have failing memory. If
you have diagnostics, running a memory test for a while may get you
somewhere. Otherwise, I'd look into trying to pinpoint the problem by
removing memory modules and see if the system becomes more stable
(assuming you have multiple memory cards, of course).
-Rick
3:29 p.m.
On 10/25/2012 08:17 AM, Rick Murphy wrote:
At 06:39 PM 10/24/2012, Kevin Reynolds wrote:
My bet is to suggest that the memory over the top connector is munged if not
pull out the video subsystem and move the cards up and try a boot then.
Allison
.... A crash example is below:
**** FATAL BUG CHECK, VERSION = V5.2 MACHINECHK, Machine check
while in kernel mode
...
KERNEL/INTERRUPT/BOOT STACK
803AF1E8 0000000C
803AF1EC 00000082
803AF1F0 7FF38058
803AF1F4 02000002
803AF1F8 00001534
803AF1FC 03C00001
Here's what that machine check stack frame means:
803AF1E8 0000000C 12 bytes of parameters
803AF1EC 00000082 Machine check code = Write bus
error, virtual address
803AF1F0 7FF38058 Virtual address
803AF1F4 02000002 Internal CPU state.
803AF1F8 00001534 PC
803AF1FC 03C00001 PSL
It seems that you are getting hard memory write faults on the stack.
Maybe something is hanging up the Q-bus, or you have failing memory.
If you have diagnostics, running a memory test for a while may get you
somewhere. Otherwise, I'd look into trying to pinpoint the problem by
removing memory modules and see if the system becomes more stable
(assuming you have multiple memory cards, of course).
-Rick
3:32 p.m.
It seems that you are getting hard memory write faults
on the stack.
Maybe something is hanging up the Q-bus, or you have failing memory. If
you have diagnostics, running a memory test for a while may get you
somewhere. Otherwise, I'd look into trying to pinpoint the problem by
removing memory modules and see if the system becomes more stable
(assuming you have multiple memory cards, of course).
-Rick
I do have multiple memory modules, there are two in the chassis. I have spare modules,
but they aren't the same, so I am not sure they will work in the system.
I did pull and replace the CPU, Memory and VCB02 modules, but the result is the same. The
system did work fine recently, so I assume that the CSR/Vectors are ok on the cards. I
don't have the manuals for everything, but maybe bitsavers does.
I don't have any spare grant continuity cards...will this cause me trouble if I just
pull a ramcard out and have nothing in its slot? I seem to remember that you couldn't
have an open slot between cards, unless you had a grant card in there...
Is there a standard VMS memory test I can run in conversational mode?
Kevin
26 Oct
26 Oct
2:46 a.m.
On Oct 25, 2012, at 3:32 PM, Kevin Reynolds <tpresence at hotmail.com> wrote:
The MicroVAX-II does not require that the memory boards be the same; you can
mix-and-match and it'll work fine.
It seems that
you are getting hard memory write faults on the stack.
Maybe something is hanging up the Q-bus, or you have failing memory. If
you have diagnostics, running a memory test for a while may get you
somewhere. Otherwise, I'd look into trying to pinpoint the problem by
removing memory modules and see if the system becomes more stable
(assuming you have multiple memory cards, of course).
I do have multiple memory modules, there are two in the chassis. I have spare modules,
but they aren't the same, so I am not sure they will work in the system. I did pull and replace the CPU, Memory and VCB02
modules, but the result is the same. The system did work fine recently, so I assume that
the CSR/Vectors are ok on the cards. I don't have the manuals for everything, but
maybe bitsavers does.
Most (all?) of the relevant docs are on bitsavers, but I'd agree with your thought
that it's likely not a configuration problem.
I don't have any spare grant continuity
cards...will this cause me trouble if I just pull a ramcard out and have nothing in its
slot? I seem to remember that you couldn't have an open slot between cards, unless
you had a grant card in there...
You'll need something in that empty slot, yes.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
4:21 a.m.
> I don't have any spare grant continuity
cards...will this cause me trouble $
I don't know which of you was responsible for the paragraph-length
lines, but whoever it was please try to avoid it...?
Compensating manually,
I don't
have any spare grant continuity cards...will this cause me
trouble if I just pull a ramcard out and have nothing in its slot?
You'll need
something in that empty slot, yes.
1:35 p.m.
On 10/26/2012 04:21 AM, Mouse wrote:
Well, in the A/B side of it, since (if it used to be a RAM card) it's a
slot with memory interconnect, rather than Q-bus, on the C/D side.
Not that it'll hurt anything to (try to) do grant continuity on the C/D
side; in order to accommodate quad-width cards (I think that's the
term; I mean cards as wide as the KA630 or most (all?) RAM), the C/D
interconnect has to be tolerant of grant continuity connections. But
you don't need it.
That's why I said "that empty slot" rather than "those empty
slots". ;)
Also, there's a 1MB PMI memory board for the KA630 that's dual-width!
I've seen one in person, once, decades ago. I don't think DEC sold
them for very long.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
I don't have any spare grant continuity
cards...will this cause me
trouble if I just pull a ramcard out and have nothing in its slot?
You'll need
something in that empty slot, yes.
1:55 p.m.
You'll need something in that empty slot, yes.
Well, in the A/B side of it [...] Also, there's a 1MB PMI memory board for the KA630
that's dual-width!
When I was writing that, it occurred to me that such a thing could
exist, though it also seemed unlikely to me that they were made. I'm
interested to hear such a thing was made.
I've seen one in person, once, decades ago. I
don't think DEC sold
them for very long.
It occurs to me that a dual-width 16M PMI board is something that would
border on trivial to make today and probably would have a small but
happy market. I know I'd definitely want to hear more if I heard of
such a thing. Is enough known about the PMI for that to be a
reasonable thing to contemplate?
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
2:14 p.m.
On 10/26/2012 01:55 PM, Mouse wrote:
It occurs to me that a dual-width 16M PMI board is
something that would
border on trivial to make today and probably would have a small but
happy market. I know I'd definitely want to hear more if I heard of
such a thing. Is enough known about the PMI for that to be a
reasonable thing to contemplate?
Well the schematics for the KA630 are out there, and the documentation
seems pretty complete. I think it wouldn't be too tough, but that's a
guess, as I've not looked at the PMI area in any detail.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
2:29 p.m.
On Oct 26, 2012, at 14:14, Dave McGuire <mcguire at neurotica.com> wrote:
On 10/26/2012 01:55 PM, Mouse wrote:
According to at least the KA650 manual, it uses the
CD interconnect for address and the ribbon cable for the
data. Presumably, the configuration would be very much
like the PDP-11 PMI with the data along a sideband.
I'd be interested in cooking something up at some point.
- Dave
It occurs to me that a dual-width 16M PMI board
is something that would
border on trivial to make today and probably would have a small but
happy market. I know I'd definitely want to hear more if I heard of
such a thing. Is enough known about the PMI for that to be a
reasonable thing to contemplate?
Well the schematics for the KA630 are out there, and the documentation
seems pretty complete. I think it wouldn't be too tough, but that's a
guess, as I've not looked at the PMI area in any detail.
2:43 p.m.
On 10/26/2012 02:29 PM, David Riley wrote:
According to at least the KA650 manual, it uses the
CD interconnect for address and the ribbon cable for the
data. Presumably, the configuration would be very much
like the PDP-11 PMI with the data along a sideband.
I'd be interested in cooking something up at some point.
I'd guess the KA630 would be similar.
The KA630 tops out at 16MB, while the KA650 can handle 64MB.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
It occurs to me that a dual-width 16M PMI board is
something that would
border on trivial to make today and probably would have a small but
happy market. I know I'd definitely want to hear more if I heard of
such a thing. Is enough known about the PMI for that to be a
reasonable thing to contemplate?
Well the schematics for the KA630 are out there, and the documentation
seems pretty complete. I think it wouldn't be too tough, but that's a
guess, as I've not looked at the PMI area in any detail. 
10:36 a.m.
I did pull and replace the CPU, Memory and VCB02
modules,
but the result is the same. The system did work fine recently, so I
assume that the CSR/Vectors are ok on the cards. I don't have
the manuals for everything, but maybe bitsavers does.
Reading this, I wonder, could the power supply be overloaded?
I do not know how much current the VCB02 modules draw ...
Just a thought. I know the QDA50 is a power hog!
- Henk, PA8PDP
2:07 p.m.
Reading this, I wonder, could the power supply be
overloaded?
I do not know how much current the VCB02 modules draw ...
Just a thought. I know the QDA50 is a power hog!
- Henk, PA8PDP
My other Microvax II has the same exact build and seems to work well (they were from a
clustered set)...
I will play with the memory boards, ribbon cables, and check the PSU monday.
Kevin
25 Oct
25 Oct
5:12 p.m.
On Wed, 24 Oct 2012, Kevin Reynolds wrote:
I have been trying to get a microvax II fully
functional over the last
few weeks and I seem to have a roadblock. The system used to boot and
allow logins, but now after connecting a VCB-02 cable to display,
keyboard and mouse, it has become more and more unreliable.
At first when connected to a display the system would
kernel panic
during boot. It would boot WITHOUT the display being connected,
however, so I continued using it without the display. Unfortunately,
now it appears to panic every boot, and I can't get to a prompt to run
any diagnostics such as anal/crash. No user based changes were made to
the system between boots. The real brain teaser is that it doesn't
always crash during the same process load, I have been trying to get a
microvax II fully functional over the last few weeks and I seem to have
a roadblock. The system used to boot and allow logins, but now after
connecting a VCB-02 cable to display, keyboard and mouse, it has become
more and more unreliable. The real brain teaser is that it doesn't
always crash during the same process load, sometimes its OPCOM,
sometimes its STARTUP, sometimes its DECW$STARTUP as it is below.
Anyone have any insight on how to proceed? I'm not a very experienced
user. A crash example is below:
If you haven't already, check the outputs of the power supply with an
oscilloscope. This is exactly what my BA123 MicroVAX II / MicroVAX III
(tried different board sets) did and it turned out to have very bad ripple
on the outputs. I'm fairly certain my BA123's power supply has bad
electrolytics, but unfortunately I don't have a list of parts because I've
not yet rebuilt it. At the time I didn't have room on the bench to tear it
down so I put the project aside and I've not yet gotten back to it.
4287
days inactive
4289
days old
13 comments
8 participants
participants (8)
-
ajp166＠verizon.net -
fraveydank＠gmail.com -
henk.gooijen＠hotmail.com -
mcguire＠neurotica.com -
mouse＠Rodents-Montreal.ORG -
rick＠rickmurphy.net -
tothwolf＠concentric.net -
tpresence＠hotmail.com