Alpha Server DS25 auto boot

[dschwarz]

Today I replaced an Alpha Server by another one from our stock.

The "new" one refuses to boot automatically (no error messages).

Entering BOOT at the SRM prompt ends in a normal boot without any errors

SRM parameters are (same as on the "old" machine):

auto_action              BOOT

bootdef_dev             dkb0.0.0.101.2 dkb100.1.0.101.2  (both are valid devices)

boot_osflags            0,0

boot_reset              OFF

Any suggestions?

Thanks in advance.

[Mark_Corcoran]

Only just reading this post 3 weeks later...

It's been a while since I've worked with DS25s (we do actually have a couple here, but not ones that our team look after), but I managed to find a copy of the owner's guide .PDF in my stash.

You say that the SRM parameters are "the same" as the old machine, but - probably for reasons of brevity - you only list a couple that are directly related to booting.

If you listed all of the environment variables from the SRM console (show *), are there any differences (that are pertinent - I'd be more concerned with SCSI-related stuff (pk*)).

You don't mention boot_file, so I'm assuming that that is not set?

Presumably neither there's no error beeps on boot, and the diagnostic LEDs don't indicate anything untoward?

You say you don't get any error messages;  is the console environment variable set to serial or graphics?

Presumably the "more el" command doesn't record any errors on the new system?

Presumably also, the "more nvram" command doesn't show a difference between both systems?

I'm not sure that it's relevant (and I don't have a DS25 to test), but certainly on old VAXen, if the system had been unpowered for a significant period of time, the battery backup of the TOY clock could have been lost, and you'd certainly be prompted to enter the system time & date when booting (you didn't mention that you were prompted for this, and I'm not sure how the DS25 stores this for when it is eventually powered back on;  if you were prompted and the system has been up for long enough now, any relevant battery backup might have been restored to a usable level).

Just my $currency 0.02 worth...

Mark

[dschwarz]

Hi Mark,

thank you very much for your reply and the time you spent to write it.

I have solved the problem last week when I had the time to play with the machine.

On page 3-3 of "AlphaServer DS25 Owner’s Guide" (Order Number: EK-DS250-UG. D01) I read:

"The auto_action environment variable specifies the action the console takes any time the system powers up, fails, or resets. The value of auto_action takes effect only after you reset the system by pressing the Reset button or by issuing the init command."

That seems not to be correct, at least for our "new" machine. Entering "Init" did not do the job. After turning off power and waiting for some 20 seocnds before powering on again, everything worked as expected.

Thanks again

[Mark_Corcoran]

Did you find that the init command didn't even reboot the system?   Or that it did, but the change to auto_action hadn't been saved?

Presumably the new system is now in production and you can't do any tests with it?

There's a suggestion in the user guide that the nvram "script" contains the commands necessary to set the various environment variables;  if there had been some error in the file then depending on where it lay in relation to the setting of the auto_action, this might explain why it didn't auto boot (however, I would expect the same problem to persist even after a power-cycle).

It would be interesting to know if on the affected machine, whether or not the reset button worked (instead of the init command), but I suspect that whatever originally caused the problem is now gone, and will be impossible to reproduce.

The user guide doesn't indicate how nvram is stored (I would guess that these days (well, when the DS25s were built), it's more likely to be flash memory than battery-backed memory like PCs of old - I know that for VAXstation 4000 VLCs, the TOY clock information was stored in a Dallas chip that had a button cell inside it, and that if the VLC had lain on DEC's (or a customer's) shelves for a long time, then cell would be dead;  I remember having to replace it on one VLC, and the Managing Director baulking at the cost - I don't think we ordered them from DEC, but whoever we did order them from, there was a minimum order quantity of 5, and they were't cheap - but this was back in the mid 90s).

It's always helpful to post the (seeming) solution, even if you manage to fix it yourself, in case others encounter the same problem in future;  on that note, if you ever get a chance to determine what the version of the SRM console, PAlcode, serial ROM, RMC ROM and RMC flash ROM are ("show config" from the SRM console) and post it here (especially if it is different to the system it replaced), that would be useful for the future :-)

Glad you got it sorted nonetheless!

Mark

[dschwarz]

Hi Mark,

just to complete the story:

The machine is in production agai.

There is no nvram script, I checked for it.

I logged console output during the last session:

P00:nkof32>>>sho config

                           hp AlphaServer DS25

Firmware
SRM Console:    V7.3-2
PALcode:        OpenVMS PALcode V1.98-43, Tru64 UNIX PALcode V1.92-33
Serial ROM:     V1.4-G   
RMC ROM:        V1.0
RMC Flash ROM:  V1.2

Processors
CPU 0           Alpha EV68CB pass 4.0 1000 MHz  8MB Bcache

Core Logic
Cchip           Rev 18
Dchip           Rev 17
PPchip 0        Rev 17
PPchip 1        Rev 17
TIG             Rev 2.6

Memory
  Array       Size       Base Address    Intlv Mode
---------  ----------  ----------------  ----------
    0       4096Mb     0000000000000000    1-Way

     4096 MB of System Memory

 Slot   Option                  Hose 0, Bus 0, PCI
   7    Acer Labs M1543C                                Bridge to Bus 1, ISA
   8    Intel 82559ER Ethern    eia0.0.0.8.0            00-02-A5-20-3D-42
   9    HP Smart Array 5300A    pya0.0.0.9.0        
  12    Yukon PCI Hot-Plug C                        
  16    Acer Labs M1543C IDE    dqa.0.0.16.0        
                                dqb.0.1.16.0        
                                dqa0.0.0.16.0           HL-DT-ST RW/DVD GCC-4482

        Option                  Hose 0, Bus 1, ISA
        Floppy                  dva0.0.0.1000.0     

 Slot   Option                  Hose 1, Bus 0, PCI
   1    Radeon 7500 PCI         vga0.0.0.1.1        

 Slot   Option                  Hose 2, Bus 0, PCI
   1/0  Adaptec AIC-7899        pka0.7.0.1.2            SCSI Bus ID 7
                                mka400.4.0.1.2          HP C7438A
   1/1  Adaptec AIC-7899        pkb0.7.0.101.2          SCSI Bus ID 7
                                dkb0.0.0.101.2          COMPAQ BF0368A4B9
                                dkb100.1.0.101.2        COMPAQ BF03687B54
                                dkb200.2.0.101.2        COMPAQ BD03686223
                                dkb300.3.0.101.2        COMPAQ BD03686223
                                dkb400.4.0.101.2        COMPAQ BD03688272
                                dkb500.5.0.101.2        COMPAQ BF03688575
                                dkb800.8.0.101.2        COMPAQ BF03688575
   5    BCM5703c (LOM)          ega0.0.0.5.2            00-16-35-3E-0B-5D

 Slot   Option                  Hose 3, Bus 0, PCI
   6    Yukon PCI Hot-Plug C                        

P00:nkof32>>>init
Initializing...


OpenVMS PALcode V1.98-43, Tru64 UNIX PALcode V1.92-33

starting console on CPU 0
initialized idle PCB
initializing semaphores
initializing heap
initial heap 240c0
memory low limit = 224000 heap = 240c0, 17fc0
initializing driver structures
initializing idle process PID
initializing file system
initializing timer data structures
lowering IPL
CPU 0 speed is 1000 MHz
create dead_eater
create poll
create timer
create powerup
access NVRAM
4096 MB of System Memory
Testing Memory
...
probe I/O subsystem
Hose 0 - PCI bus running at 33Mhz
entering idle loop
Hose 1 - PCI bus running at 66Mhz
Hose 2 - PCI bus running at 66Mhz
Hose 3 - PCI bus running at 33Mhz
starting drivers
initializing keyboard
initializing GCT/FRU at 25a000
Initializing ega dqa dqb eia pka pkb
Memory Testing and Configuration Status
  Array       Size       Base Address    Intlv Mode
---------  ----------  ----------------  ----------
    0       4096Mb     0000000000000000    1-Way

     4096 MB of System Memory
Testing the System
Testing the Disks (read only)
Testing the Network
AlphaServer DS25 Console V7.3-2, built on Feb 27 2007 at 13:10:18

P00:nkof32>>>

At this point I powered down the system and after powering on again it came up:

RMC - System is Down
RMC - Starting to Test Max Fan Speeds
RMC - Power Supplies OK
RMC - System is Down
RMC - Starting to Test Max Fan Speeds
RMC - Power Supplies OK
RMC - System DC is OK
RMC - System is up
RMC - System Fans OK
RMC - Powerup Complete

 SROM V1.3 CPU # 00 @  1000  MHz
SROM program starting
Reloading SROM

 SROM V1.4-G CPU # 00 @  1000  MHz
 System Bus Speed @ 0125  MHz
SROM program starting
PCI66 bus speed check
Bcache data tests in progress
Bcache address test in progress
CPU parity and ECC detection in progress
Bcache ECC data tests in progress
Bcache TAG lines tests in progress
Memory sizing in progress
Memory configuration in progress
Testing AAR0
Memory data test in progress
Memory address test in progress
Memory pattern test in progress
Memory thrashing test in progress
Memory initialization
Loading console
Code execution complete (transfer control)


OpenVMS PALcode V1.98-43, Tru64 UNIX PALcode V1.92-33

starting console on CPU 0
initialized idle PCB
initializing semaphores
initializing heap
initial heap 240c0
memory low limit = 224000 heap = 240c0, 17fc0
initializing driver structures
initializing idle process PID
initializing file system
initializing timer data structures
lowering IPL
CPU 0 speed is 1000 MHz
create dead_eater
create poll
create timer
create powerup
access NVRAM
4096 MB of System Memory
Testing Memory
...
probe I/O subsystem
Hose 0 - PCI bus running at 33Mhz
entering idle loop
Hose 1 - PCI bus running at 66Mhz
Hose 2 - PCI bus running at 66Mhz
Hose 3 - PCI bus running at 33Mhz
starting drivers
initializing keyboard
initializing GCT/FRU at 25a000
Initializing ega dqa dqb eia pka pkb
Memory Testing and Configuration Status
  Array       Size       Base Address    Intlv Mode
---------  ----------  ----------------  ----------
    0       4096Mb     0000000000000000    1-Way

     4096 MB of System Memory
Testing the System
Testing the Disks (read only)
Testing the Network
AlphaServer DS25 Console V7.3-2, built on Feb 27 2007 at 13:10:18

CPU 0 booting

(boot dkb0.0.0.101.2 -flags 0,0)
block 0 of dkb0.0.0.101.2 is a valid boot block
reading 1143 blocks from dkb0.0.0.101.2
bootstrap code read in
base = 61c000, image_start = 0, image_bytes = 8ee00(585216)
initializing HWRPB at 2000
initializing page table at ffff0000
initializing machine state
setting affinity to the primary CPU
jumping to bootstrap code


    OpenVMS (TM) Alpha Operating System, Version V7.3-2  
    © Copyright 1976-2003 Hewlett-Packard Development Company, L.P.

........

Thanks again

Dieter