HP-UX 11.00 system panic

[Adil_4]

Hello!
I have HP 9000 Class D (model D220). I have just installed an operating system on it. After errorless boot i have logged into the serial console. Then I opened remote X session. In several minutes the X terminal hanged and i saw the following on the serial console:

........
........
normal boot
........
........
Start NFS server subsystem ...........
Start X print server(s) ................................... N/A
Start CDE login server .................................... OK

The system is ready.

GenericSysName [HP Release B.11.00] (see /etc/issue)
Console Login:

~~~~~~~~~~~~~~~~~~~~~~~~~~
In a few minutes
~~~~~~~~~~~~~~~~~~~~~~~~~~
User initiated TOC, please wait.


******* Unexpected HPMC/TOC. Processor HPA 0xfffbe000 *******
GENERAL REGISTERS:
r00/07 00000000 006219D0 002D6074 EF000800 00B01A58 EF000800 002D5EDC 00000000
r08/15 00000000 00000000 00000100 00009DAC 00A92DC0 01811F80
r16/23 005E5D88 017AC000 00163AEC 00000005 003126C8 0000001E 0000001F 00000060
r24/31 FFFFFFFF FFD05800 0000001F 006239D0 0000001E 00000000 005C45B0 FFFBE000
CONTROL REGISTERS:
sr0/7 00000000 000079D8 00002161 00000000 00000000 00002161 00006DEF 00000000
pcq 00000000.002D6084 00000000.002D6088 isr = 00000000 ior = 002D6084
pid1/4 C667 4604 F34A 0000 ipsw = 0004001F iva = 00027000 sar = 15 ccr = C0
tr0/7 005E5D88 00033000 0000001F 005C4940 0000001F 391AFC7C C067DE3B 005C4618
rctr = 7FFFFC0D itmr = 446C3200 eiem = EF000800 eirr = 00800480 iir = E81F1FF7
cr1/7 00000000 00000000 00000000 00000000 00000000 00000000 00000000
MACHINE CHECK PARAMETERS:
Check Type = 00000000 CPU STATE = 9E000001 Cache Check = 00000000
TLB Check = 00000000 Bus Check = 00000000 PIM State = ? SIU Status = ????????
Detected by 00000000 Assists = 00000000 Processor = 00000000
Assist ID = 00000000 Slave Addr = 00000000 Master Addr = 00000000


TOC, pcsq.pcoq = 0.2d6084, isr.ior = 0.2d6084
@(#)B2352B/9245XB HP-UX (B.11.00) #1: Thu Nov 6 01:58:21 PST 1997
Transfer of control: (display==0xd904, flags==0x0)



*** A system crash has occurred. (See the above messages for details.)
*** The system is now preparing to dump physical memory to disk, for use
*** in debugging the crash.

*** The dump will be a SELECTIVE dump: 19 of 128 megabytes.
*** To change this dump type, press any key within 10 seconds.
*** Proceeding with selective dump.

*** The dump may be aborted at any time by pressing ESC.
*** Dumping: 100% complete (19 of 19 MB).
*** System rebooting.



This problem doesn't happen periodically. But I noticed that this problem occures when I press TCO button on the back of the machine. What can be the reason of this?

Thanks for your answers in advance.

Yeah, one more question. TCO(Transfer Of Control) - What is this for???

[RAC_1 · Accepted solution]

What your are getting is what your sould get when you do TOC. (not TCO)

This TOC will reboot the system and dump the crash to ensure data integrety. This is useful when systems hang and crash helps in that.

[Adisuria Wangsadinata_1]

Hi,

ToC (Transfer of Control) is a kind of reset button for your system; but this ToC not just reset the system, the ToC also will generate the crashdump on your system.

So with ToC, the system will reset by itself and create a crashdump as a log file for troubleshooting for HP (as a fingerprint). But if there's nothing wrong with the system, the crashdump will tell exactly what's happened on the system.

If the system rebooted by itself without ToC, it's better to check PIM information on BCH mode ... check the timestamp, if the timestamp is updated, call HP and refer this info, they will analyse from this information. Also check the warning/error message on LCD panel on your D220.

Hope this information can help you.

Cheers,
AW

[AwadheshPandey]

go to:
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=760601&admit=-682735245+1129792971136+28353475

Awadhesh

[Adil_4]

Here is the output of PIM

Service Menu: Enter command > pim

PROCESSOR PIM INFORMATION

----------------- Processor HPMC Information ------------------

No valid timestamp

HPMC Chassis Codes = No chassis codes logged

General Registers 0 - 31
0 - 3 0x00000000 0x00000000 0x00000000 0x00000000
4 - 7 0x00000000 0x00000000 0x00000000 0x00000000
8 - 11 0x00000000 0x00000000 0x00000000 0x00000000
12 - 15 0x00000000 0x00000000 0x00000000 0x00000000
16 - 19 0x00000000 0x00000000 0x00000000 0x00000000
20 - 23 0x00000000 0x00000000 0x00000000 0x00000000
24 - 27 0x00000000 0x00000000 0x00000000 0x00000000
28 - 31 0x00000000 0x00000000 0x00000000 0x00000000




Control Registers 0 - 31
0 - 3 0x00000000 0x00000000 0x00000000 0x00000000
4 - 7 0x00000000 0x00000000 0x00000000 0x00000000
8 - 11 0x00000000 0x00000000 0x00000000 0x00000000
12 - 15 0x00000000 0x00000000 0x00000000 0x00000000
16 - 19 0x00000000 0x00000000 0x00000000 0x00000000
20 - 23 0x00000000 0x00000000 0x00000000 0x00000000
24 - 27 0x00000000 0x00000000 0x00000000 0x00000000
28 - 31 0x00000000 0x00000000 0x00000000 0x00000000

Space Registers 0 - 7
0 - 3 0x00000000 0x00000000 0x00000000 0x00000000
4 - 7 0x00000000 0x00000000 0x00000000 0x00000000




IIA Space = 0x00000000
IIA Offset = 0x00000000
Check Type = 0x00000000
CPU State = 0x00000000
Cache Check = 0x00000000
TLB Check = 0x00000000
Bus Check = 0x00000000
Assists Check = 0x00000000
Assist State = 0x00000000
System Responder Address = 0x00000000
System Requestor Address = 0x00000000
Path Info = 0x00000000
Physical Memory = 0x00000000
Available Memory = 0x00000000




CPU Diagnose Register 0 = 0x00000000
MIOC Status Register = 0x00000000
MIOC Control Register = 0x00000000
MIOC MDERRADD Error Address Register = 0x00000000
MIOC DMAERR DMA Error Register = 0x00000000
MIOC DIOERR IO Error Register = 0x00000000




----------------- Processor TOC Information -------------------

General Registers 0 - 31
0 - 3 0x00000000 0x006219d0 0x00000000 0x00000001
4 - 7 0x00000000 0xffffffff 0x005e5d88 0x005e63e4
8 - 11 0x00000001 0x00000000 0x00000100 0x0007dac5
12 - 15 0x00a917c0 0x0176ad80 0x0007dac5 0x00000000
16 - 19 0x005e5d88 0x01642000 0x00000002 0x00000001
20 - 23 0x0007dacc 0x005e5d88 0x00000001 0x00002a40
24 - 27 0x00002a40 0x00000000 0x00000000 0x006239d0
28 - 31 0x00000000 0x005e8780 0x00d3a170 0x00000001




Control Registers 0 - 31
0 - 3 0x7ffff795 0x00000000 0x00000000 0x00000000
4 - 7 0x00000000 0x00000000 0x00000000 0x00000000
8 - 11 0x000035a7 0x00008f0c 0x000000c0 0x00000005
12 - 15 0x0000e035 0x00000000 0x00027000 0xfffffff0
16 - 19 0x64d63ad0 0x00000000 0x000e3238 0x4bd03ef1
20 - 23 0x00000000 0x00d3a0e8 0x0004ff0f 0x00000000
24 - 27 0x005e5d88 0x00033000 0x00000000 0x7f65b000
28 - 31 0x00000000 0x42ad3cd3 0x00000010 0x00d3a170

Space Registers 0 - 7
0 - 3 0x00000000 0x00003625 0x000032d4 0x00000000
4 - 7 0x00000000 0x000032d4 0x0000544c 0x00000000

IIA Space = 0x00000000
IIA Offset = 0x000e323c
CPU State = 0x9e000001





Memory Error Log Information:

No valid timestamp

No memory errors logged


I/O Module Error Log Information:

No valid timestamp

No I/O module errors logged




IO Bus Converter Log
--------------------

GBOA HPA = 0x00000000
GBOA UBC_IO_CONTROL = 0x00000000
NIO HPA = 0x00000000
GBOA IO_IO_LO = 0x00000000
GBOA IO_IO_HI = 0x00000000

Service Menu: Enter command >


The LCD displays the following:

RUN F01F
HP D220

[Adisuria Wangsadinata_1]

Hi,

From PIM, there's no valid timestamp and all register with zeros. From LCD display, everything looks up & running fine.

From the first info :
User initiated TOC

This mean that someone try to press TOC and this will make the system reseted.

Do not try to press TOC unless there's no other choice to reboot the system.

From all information that you gave, the machine indicate up & running. And since you mention the system has just installed, please do not forget to apply the necessary patch bundles into the system. This will help the systems run smoothly.

FYI, the EISA card is forbidden on D220 if the OS is HP-UX 11.x 64bit.

Hope this information can help you.

Cheers,
AW

[Adil_4]

Well, but how to fix the problem with timestamp?

And the main thing: Where did you get information about usage of EISA card on HP-UX-11 on D220? Could you please give me a link to that document (if possible)?

Thanks in advance!

[RAC_1]

You seem to have problem with cpu1.
pdcinfo -nobanner > /tmp/cpu.info
Time to call HP.

[Torsten.]

you wrote:

"But I noticed that this problem occures when I press TCO button on the back of the machine"

This happens *ONLY* when you press the button, right?

Nobody is in panic. You only initiate a reset and a memory dump. That's all. There won't be any PIM information. But you have now a complete memory dump in /var/adm/crash.

Don't press the RED BUTTON without a reason.

Press this one instead: http://www.turnofftheinternet.com/

;-))

[Sameer_Nirmal]

It sounds like you only noticed the problem when you press the TOC button. So those messages are quite obvious as part of TOC process.

D220 provide only 32-bit support. As you are installing HP-UX 11.00 , make sure it is
11.00 32-bit.

In the current scenario, I would remove the graphics card and attach a ASCII terminal and see what happens. Moreover, I would check, which graphics card it is ( Product No and if supported ) and in which slot the card was installed .

Be-aware : As mentioned by Adisuria, you can't use EISA card on HP-UX 64-bit environment as the their device driver is not 64-bit capable. If you you need to use EISA cards, then stick to 32-bit OS environment only.

[Adisuria Wangsadinata_1]

Hi Adil,

Here is the information from HP about EISA card on 64bit environment (docID : F27124709) :

http://www1.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000080256021

Please do not forget to apply the latest patch bundles for your system, and press TOC for emergency only (eg. system hang).

Hope this information can help you.

Cheers,
AW

PS : please assign the points to the people that spend the time to help you, thanks

[Adil_4]

I said that the system crash occures with no reason and not periodically without any intervention. The system just works and suddenly crashes... and this also happens when the TOC button is pressed.


How can i know what version of OS (32 or 64 bit) was installed to be sure that there is no conflict between EISA drivers and system?

P.S. All your answers deserve high points :)

[Arunvijai_4]

# getconf KERNEL_BITS
# getconf HW_CPU_SUPP_BITS
# getconf HW_32_64_CAPABLE
# file /stand/vmunix

Also, you can find via SAM.

-Arun

[Adisuria Wangsadinata_1]

Hi Adil,

To check whether your OS run on 64 bit :

# getconf KERNEL_BITS

To check if your system is pure 64 bit system :

# getconf HW_CPU_SUPP_BITS

Hope this can help you.

Cheers,
AW

[Arunvijai_4]

From SAM, # sam--> Performance Monitors--> System Properties and it should be some thing like,

Processors: â â â
â â â â Active: 2 â â â
â â â â Total: 2 â â â
â â â â CPU Version: PA 8800 CPU Module 3.1 â â â
â â â â Clock Frequency: 800 MHz â â â
â â â â Machine Identification: 383457852 â â â â
â â â â Hardware Model: 9000/800/rp3410 â â â â
â â â â Kernel Width Support: 64

and in Operating system tab,

OS Identification: B.11.11 U â â â
â â â â OS Kernel Width: 64

-Arun

[Muthukumar_5]

chatr /stand/vmunix givies bit information.

hth.

[Adil_4]

These commands show that my system is 32bit.

[Arunvijai_4]

Thats correct. You have to go with 32 bit only.

-Arun