OpenVMS halt code = 5

[Latif2002]

Hi,

I have DEC 3000 and always give me error and keeping rebooting

halted CPU 0
halt code = 5
halt instruction executed
PC = ffffffff8004bf84

So please if anyone can help me

Thx.

[Bhadresh]

Hi,

The system has encountered a HALT request while executing in a privileged processor mode.

Have a look at following thread:
http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1228939

Regards,
Bhadresh

[Volker Halle]

Latif2002,

welcome to the OpenVMS ITRC forum.

If your DEC3000 system reboots after a kernel mode halt (halt code 5), the SRM variable AUTO_ACTION is set to BOOT instead of RESTART, otherwise the system would write a HALT crash before booting again. If AUTO_ACTION would be set to HALT, the system would just halt, but you won't be able to further diagnose the problem.

Is this a new behaviour for this machine ? Did it work/boot correctly before ? What changed ?

To further diagnose this problem, >>> SET AUTO_ACTION RESTART and let the system write a crash. Then boot from the OpenVMS Operating System CD - if that works - and have a look at the crash by mounting the system disk and looking at the crash with ANAL/CRASH DKAx:[SYS0.SYSEXE]SYSDUMP.DMP (assuming the system boots from SYS0).

Issue a SDA> CLUE CRASH command once you've successfully opened the SYSDUMP.DMP file and post the output.

Note that OpenVMS itself should NOT issue a HALT instruction in kernel mode under normal circumstances.

Volker.

[Latif2002]

Volker,

When I type
$ ANAL/CRASH DKA0:[SYS0.SYSEXE]SYSDUMP.DMP
I get this
%SDA-F-OPEN, ERROR OPENING DKA0:[SYS0.SYSEXE]SYSDUMP.DMP ; AS INPUT
-RMS-E-DNR, DEVICE NOT READY, NOT MOUNTED, OR UNAVAILABLE.

[Bhadresh]

The error indicates physical device is not ready.

Is this your system disk is dka0 and mounted properly?

Regards,
Bhadresh

[Latif2002]

Bhadresh,

DKA0 is my disk and do I need to mount the disk and how

Thx.

[Bhadresh]

Use the following command to MOUNT an disk device "dka0:" and determine its volume label:

$ MOUNT/OVERRIDE=ID dka0:

Dismount the disk device:

$ DISMOUNT dka0:

MOUNT the disk device with appropriate volume label:

$ MOUNT dka0: volume-label [logical-name]

Regards,
Bhadresh

[Latif2002]

Bhadresh,

I t is ok but after I run the command below
SDA> CLUE CRASH
I saw many pages with numbers so how I can know the problems

Thx.

[Volker Halle]

So you seem to have a valid crash. Capture the output on the screen (connect a laptop to the serial console port, use Putty and session logging). Then post the output as an attached .TXT file to your next reply.

SDA> CLUE CRASH
SDA> CLUE CONFIG

Volker.

[Latif2002]

Volker,

I donâ t have a laptop now so if there is a way to dump it to floppy

Thx.

[Volker Halle]

If this DEC 3000 system has a floppy, you could store the output file on that floppy. But you need an OpenVMS system to read that floppy afterwards !

Let's start with some basic analysis. You have to write down and provide the output exaclty as it occurs on the screen. Or you can take a picture of the screen output and post the .JPG

SDA> CLUE CRASH
Provide the data for the first few lines (from Bugcheck Type until offset)

SDA> EXA/INS 8004bf84

Volker.

[RBrown_1]

Don't forget that you can

SDA> SET OUTPUT filespec
SDA> CLUE CRASH
SDA> CLUE CONFIG

to put the CLUE output into the specified file. You can then transfer this text file to someplace from where you can post to here.

[Bob Blunt]

The capture solution doesn't *need* to be a laptop. Any PC with a spare COM port and terminal emulator, another VMS server with a spare MMJ or RS-232 (DB25), probably even a Unix/Linux system could be setup to capture the console output. Some of the more intelligent dumb terminals (VT4xx, VT5xx) can save some data but pasting it back out can be difficult. Or, if all else fails, a hardcopy terminal or old school dot matrix printer can at least save the console data so you don't have to sit there waiting for it to fail. Or you could connect the console port to a terminal server... There are ways. Not all are simple, but there are ways.

I can't remember if you gave a clear description of when this problem happens? At power-up? During boot? Does the system still boot at all? What OpenVMS version? Patches? The usual suspects...

bob

[Latif2002]

Bob,

It booting until
⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦..
Completed

And I have OpenVMS version 6.2

Thx.

[Volker Halle]

It booting until
⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦⠦..
Completed


These are not OpenVMS native messages, are they ? In the initial problem description, you correctly typed a CPU halt message. Does this problem still exist and show up in the same way ? Or do you now have a different problem ?

If you want us to help you, you need to provide the data from the crash (CLUE CRASH output any maybe more).

Volker.