SRM Console hang up

[Elroy_Starcke]

I did a firmwareupdate on a PWS 433a from 6.something to

apu 7.1 *

SRM 7.1-3 *

POST 5.90

AlphaBIOS 5.70 *

All goes fine. I did powecycle my PWS and the "NT-Firmware" did boot fine. Then I've switched to SRM-Console and powercycle. SRM says:"ea.e9.e8.e7.e6.insufficient dynamic memory for a request of 512 bytes" and a blinking cursor in next line. I can't type in anything! How can I switch back to NT?

[Steven Schweda · Accepted solution]

I don't know. Yank the battery?

I just powered up my PWS 500a[u] to see what
firmware was on it, and I got so far as
"AlphaBIOS Version 5.70", but then it wants
me to "Press F2 to enter SETUP", and I don't
know how to do that from a serial console.

The (possibly relevant) point is that I
believe that I left it pointing to the SRM
console (which was about all I ever used),
but now, years later, it seems to have
reverted to AlphaBIOS on its own (I assume,
because the battery's died in the interval).

[Bob Blunt]

What console type do you need, Elroy, what O/S are you trying to use? Does the machine need to boot AlphaBIOS or SRM?

I'm pretty sure that once you have the machine set to use one console type or the other that it has to finish self-test to get you to a point you can change back to the other.

As far as getting back from AlphaBIOS to SRM, there should be instructions on using a DEC VT terminal in 8-bit mode to work through the menus to reset, but I'd try to make sure that the battery was good first. Since most of the times I've done this type operation I've been remote from the lab my only option was using the 8-bit VT solution and it's limitations. I'm not sure how or if the procedures work with a graphics head and attached mouse and keyboard.

bob

[Steve Reece_3]

Did you read the release notes and make sure that you covered all of the items in there and all the pre-requisites before doing the upgrade?

Yanking the battery might do the trick on getting you back to Alpha-BIOS, provided that the system doesn't remember the setting that you did in NVRAM that's not battery backed or anything silly like that. There may be a jumper setting too I guess...

[Elroy_Starcke]

At first, thank you guys for your quick replying!!! The battery was a very good hint. I removed it for a minute and then the Bios starts again. But when I switch over to SRM (it equals if it's UNIX or VMS) the old error appears. My originally plan is to run OpenVMS on this machine. I had installed for the meantime debian over the UnixSRM. It runs, a bit slow, but it runs! After that I've made the firmwareupdate, and I did'nt read the release notes, because with the right informations there is no trouble and no trouble is no fun ;-))! And afterwards the insufficient memory error appears. Could it be that my memory is to small for the new firmware?

My machine has:

-256 MB Ram
-No external cache-module
-Disk 0 18GB (scsi)
-Disk 1 4GB (scsi)

[Steven Schweda]

> Could it be that my memory is to small for
> the new firmware?

> -256 MB Ram

And the AlphaBIOS sees 256MB?

I know nothing, but I'd expect 256MB to be
enough for the console firmware. (More would
be better for normal use.) If you had more
memory to add, then I'd add it, but I
wouldn't go buy more until I knew more.

Which graphics card is in there?

You may be able to find a document which says
what it's doing at or just after ".e6.", and
that might provide a clue.

I'll see if I can get a new Li cell for mine
(which is stuck in AlphaBIOS because of its
dead one), and I'll report what I learn (if
anything).

[Elroy_Starcke]

The graphics card is a MATROX POWER GRAPHICS ACCELERATOR MGA Series, VGA/VBE BIOS Version V 1.1!

I'd tryed to update the firmware a second time, but the machine detects that the file on the diskette is not newer as the installed file. Does thas mean that I'm not able to downgrade the firmware?

[Steven Schweda]

> The graphics card is [...]

Not one I've ever heard of. I don't know if
it would help, but you might try removing the
graphics card, and using a serial port as the
console. (You'd need a terminal/emulator and
an appropriate cable. Unplugging the console
keyboard should be enough to persuade it to
use the serial port. I also don't know which
serial port it likes better, so you may need
to try both.)

> [...] Does thas mean that I'm not able to
> downgrade the firmware?

I don't know that either. I've never tried
to go backward. Knowing nothing, I'd expect
it to ask if you really wish to go back,
rather than absolutely refuse to do it. I
probably also did it from a CD-ROM instead of
a floppy, but I wouldn't expect that to
matter.

[Steven Schweda]

Ok, with a new Li cell, I can now get the SRM
console setting to stick on my system, and I
see (at a serial console, using the lower
connector):

*** keyboard not plugged in ...

ff.fe.fd.fc.fb.fa.f9.f8.f7.f6.f5.CPU 0 speed is 2.00 ns (500MHz)

ef.ee.ed.ec.eb.f4.ea.e9.e8.e7.e6.Auto Negotiation restart failed
Auto Negotiation restart failed
Auto Negotiation restart failed

ewb0: link failed : Using 100BaseTX: full duplex
Auto Negotiation restart failed
Auto Negotiation restart failed
Auto Negotiation restart failed

ewa0: link failed : Using 100BaseTX: full duplex
Digital Personal WorkStation 500au
Console V7.2-1 Mar 6 2000 14:47:02

So, it appears that after ".e6.", it's doing
the network initialization. I seem to have
the onboard interface (EWA0) plus a DE500-BA
(?) (EWB0), and I had no cables connected,
hence the complaints.


>>>show pal
pal OpenVMS PALcode V1.20-16, Tru64 UNIX PALcode V1.22-18

>>>show version
version V7.2-1 Mar 6 2000 14:47:02

>>>show memory

Memory Size = 768Mb

Bank Size/Sets Base Addr
------ ---------- ---------
0 512Mb 00000000
1 128Mb 20000000
2 128Mb 28000000

BCache Size = 2Mb

Tested Memory = 768Mbytes


>>>show console
console graphics

Which, I assume, is why it's expecting to see
a keyboard, but, as I expected, it's willing
to fall back to the serial console when the
keyboard is not connected.


If I yank the denser memory:

*** keyboard not plugged in ...

ff.fe.fd.fc.fb.fa.f9.f8.f7.f6.f5.CPU 0 speed is 2.00 ns (500MHz)

ef.ee.ed.ec.eb.f4.ea.e9.e8.e7.e6.Auto Negotiation restart failed
Auto Negotiation restart failed
Auto Negotiation restart failed

ewb0: link failed : Using 100BaseTX: full duplex
Auto Negotiation restart failed
Auto Negotiation restart failed
Auto Negotiation restart failed

ewa0: link failed : Using 100BaseTX: full duplex
Digital Personal WorkStation 500au
Console V7.2-1 Mar 6 2000 14:47:02


>>>show memory

Memory Size = 256Mb


So, 256MB seems to be plenty (for the
console).


If I then connect some network cables:

*** keyboard not plugged in ...

ff.fe.fd.fc.fb.fa.f9.f8.f7.f6.f5.CPU 0 speed is 2.00 ns (500MHz)

ef.ee.ed.ec.eb.f4.ea.e9.e8.e7.e6.
ewb0: link up : Negotiated 100BaseTX: full duplex

ewa0: link up : Negotiated 100BaseTX: full duplex
Digital Personal WorkStation 500au
Console V7.2-1 Mar 6 2000 14:47:02


>>>show config

Firmware
SRM Console: V7.2-1 Mar 6 2000 14:47:02
ARC Console: 5.70
PALcode: OpenVMS PALcode V1.20-16, Tru64 UNIX PALcode V1.22-18
SROM Version: v5.90

Processor
DECchip (tm) 21164A-2 Pass 500 MHz 96 KBytes SCache
2 MB BCache
PYXIS ASIC Pass 257

MEMORY


Memory Size = 256Mb

Bank Size/Sets Base Addr
------ ---------- ---------
1 128Mb 00000000
2 128Mb 08000000


BCache Size = 2Mb

Tested Memory = 256Mbytes

PCI Bus
Bus 00 Slot 03: Digital Semiconductor 21143 Network Controller
ewa0.0.0.3.0 00-00-F8-75-99-EB

Bus 00 Slot 04: PCI IDE
pqb0.0.1.4.0 PCI EIDE

Bus 00 Slot 07: Intel SIO 82378


Bus 00 Slot 20: DECchip 21052 PCI to PCI Bridge
Bus 01 Slot 08: DECchip 21152 PCI to PCI Bridge
Bus 02 Slot 00: NCR 875 Scsi Controller
pka0.7.0.2000.0 SCSI Bus ID 7
dka0.0.0.2000.0 SEAGATE ST118202LC
dka400.4.0.2000.0 PLEXTOR CD-ROM PX-40

Bus 02 Slot 02: DE500A Network Controller
ewb0.0.0.2002.0 00-06-2B-00-07-B5

Bus 01 Slot 09: ELSA GLoria Synergy


Bus 01 Slot 10: ISP1040 Scsi Controller
pkb0.7.0.1010.0 SCSI Bus ID 7

ISA
Slot Device Name Type Enabled BaseAddr IRQ DMA
0
0 MOUSE Embedded Yes 60 12
1 KBD Embedded Yes 60 1
2 COM1 Embedded Yes 3f8 4
3 COM2 Embedded Yes 2f8 3
4 LPT1 Embedded Yes 3bc 7
5 FLOPPY Embedded Yes 3f0 6 2
6 EIDE Embedded Yes 1f0 14
3f6 15
170
376
7 ES1888 Embedded Yes 220 5 1
>>>


It appears that my SRM console version,
"V7.2-1 Mar 6 2000 14:47:02", is newer than
your "SRM 7.1-3 *", so I'd vote for an
upgrade to the new one. It's been a long
time since I updated this thing, but at:

http://h18002.www1.hp.com/alphaserver/firmware/readmes/digitalpw.html

I see this:

[...]
Note: V7.2-1 was the last firmware
release.
[...]

According to:

ftp://ftp.hp.com/pub/alphaserver/firmware/retired_platforms/workstations/digital_pers_wrksta/digitalpw_v72_fw_relnote.pdf

2.1 Console Firmware Components ­ Revision Information
The latest Firmware Update Bootfile "DIGITALPW_V72_1.EXE" on the Alpha Systems Firmware
Update CD V5.7, contains the SRM Console V7.2-1. [...]


So an Alpha Firmware CD-ROM of that vintage
(or another floppy kit) should be ok.

ftp://ftp.hp.com/pub/alphaserver/firmware/iso_images/
ftp://ftp.hp.com/pub/alphaserver/firmware/iso_images/v57.iso

[Elroy_Starcke]

Hello Steven!

I've downloaded the DIGITALPW_V72_1.EXE and copyed it to a diskette. The PWS does'nt recognize it, or the installed version is the same as the version from diskette. The installed Alpha-BIOS is 5.70 and the version from the DIGITALPW_V72_1.EXE is 5.70 also. That means I'm not able do upgrade the SRM-version. Now it's very triky. What can I do??? Is it possible to reset the firmware? On the ISO-file I've downloaded is no firmwareversion for the PWS 433a and I thinks that it is not possible to use the firmware of an other alphamachine?! Maybe I should try to write the diskette on a windows-pc?

[Steven Schweda]

> I've downloaded the DIGITALPW_V72_1.EXE

As I said, I'd make and use a CD-ROM.

> and copyed it to a diskette.

There may be more than one way to do that.
How did _you_ do it? On what?

> The PWS does'nt recognize it, or the
> installed version is the same as the
> version from diskette.

I'd know more if you explained exactly what
you did, and what happened then, instead of
providing your analysis and interpretation of
what happened when you did I-don't-know-what.

I assume that there are precise instructions
somewhere which explain how to create a
suitable floppy, holding a suitable file,
using the tools available on one or more
different operating systems, but I haven't
seen them for a long time (because I simply
use a CD-ROM).

[Bob Blunt]

I'm pretty sure that the .EXE is intended to be booted from a hard disk. A .SYS file is downline loaded from an Infoserver or a suitably setup VMS-based MOP server (LANCP, Phase IV or V). I don't think you can boot the firmware bootable file from a floppy at all.

The "insufficient dynamic memory" error sounds very familiar, though. It's not related to the actual memory on the system. Something about needing to find a certain amount of contiguous memory in the first "so much" of memory. It'd be interesting to know if the physical memory in the system is properly setup with the SIMMs/DIMMs in the right order, etc. There's always the option of taking out the peripheral cards, using a serial console and trying without any add-on cards.

Nothing that you've done seems like it should have corrupted anything permanently especially if you got no errors during the firmware upgrade. The sticking point is that error. If you can't get SRM to pass that point then you can't get output from SHOW CONFIG or SHOW DEVICE for diagnostics. If it were a machine in my lab, I'd pull any PCI and EISA/ISA cards and try to get it to pass POST. It seems like you've got a device no longer supported by the SRM giving you a wedgie that AlphaBIOS allows.

bob

[Elroy_Starcke]

Sorry guys, it was a very bloddy failure of mine. I did'nt rename the file on the diskette to "fwupdate.exe"!!! The machine did now the firmwareupdate. I did a romtest and it reports that the srm-rom had a error. It seems that by the first firmwareupdate something went wrong. Now the SRM-Console starts normaly.

Thank you very much for your patience and your support!!!

[Elroy_Starcke]

The first firmwareupdate seems to cause a error in the SRM-rom. A new firmwareupdate corrected this error. SRM-Console starts normaly again.

Thanks a lot!!!