HPE 9000 and HPE e3000 Servers
1748209 Members
2796 Online
108759 Solutions
New Discussion юеВ

Re: PIM logs boot error

 
SOLVED
Go to solution
Michael O'brien_1
Regular Advisor

PIM logs boot error

Hi,

I have an N-Class server which fails to boot. It starts to load the kernel and then fails. I've tired to boot the machine from both primary and alt boot disk, also tired the HP-UX install CD and still get the same error.

The machine gets to

alloc_pdc_pages: Relocating PDC from 0xf0f0000000 to 0x7f9ab000.

and then crashes.

In the SL logs I get errors

ALERT LEVEL: 2 = Non-Urgent operator attention required
SOURCE: 8 = I/O
SOURCE DETAIL: 2 = system bus adapter SOURCE ID: 1A
PROBLEM DETAIL: 3 = function failure
CALLER ACTIVITY: 6 = machine check STATUS: 3
CALLER SUBACTIVITY: 34 = implementation dependent
REPORTING ENTITY TYPE: 0 = system firmware REPORTING ENTITY ID: 03
0x70003023821A6343 00000000 00070200 type 14 = Problem Detail
0x58003823821A6343 00006804 0F002413 type 11 = Timestamp 05/15/2004 00:36:19

Type CR for next entry, - CR for previous entry, Q CR to quit.

I've attached the PIM logs and the SL logs to this postings. Could anybody help figure out what part of the server has the problem.

Many Thanks
Michael



12 REPLIES 12
Mel Burslan
Honored Contributor

Re: PIM logs boot error

I am not sure, who in the forums other than HP employees can claim to be hardware experts but I sure am not one of those. Hence, my best suggestion to you is to place a hardware call for HP to figure out the problem and fix it. I am sure you have some sort of support contract for the hardware at the least. If not, it is a very good point to be made to those who declined to have support contract.

To my remotely familiar to the context eye, this looks like a cpu board problem but again take my word with a lot of salt, not just a grain. :)

Good luck
________________________________
UNIX because I majored in cryptology...
Devender Khatana
Honored Contributor

Re: PIM logs boot error

Hi,

This seems to be a hardware failure case. You need to call HP for rectification if you have a support contract.


HTH,
Devender
Impossible itself mentions "I m possible"
melvyn burnard
Honored Contributor

Re: PIM logs boot error

This is normally caused by the hardware PDT (page deallocation table) being full, and while booting the system is trying to mark another page of memory as faulty.
Log a hardware call with your local HP Response Centre
My house is the bank's, my money the wife's, But my opinions belong to me, not HP!
timmy b.
Honored Contributor
Solution

Re: PIM logs boot error

Michael,
Thanks for including the PIM dump.

Analysis of the HPMC points to the most likely cause of the problem to be in slot 6, path 0/2. Try removing this card and booting.

If you want the analysis information, I can provide it for you.

Good Luck - Tim
There are 10 kinds of people in this world: Those who understand Binary, and those who don't.
aliasgar
Valued Contributor

Re: PIM logs boot error

Hi Michael,

what card is there in slot 3 path 0/12 .u can try removing the card in slot 3 and try to boot the server .


Regards,
Aliasgar.
aliasgar
mits
Respected Contributor

Re: PIM logs boot error

Most likely is some IO problem. The kernel initializes the I/Os after diplaying the message "alloc_pdc_pages:" on the console.

You can remove all I/O cards then try the install CD boot to check which I/O card causes your HPMC. HPMC PIM itself should indicate which I/O is most culprit. You can log a HP support call for help.

Or you can run ODE mapper or IOTEST from the ISL prompt. ODE -> run mapper or run iotest.
Michael O'brien_1
Regular Advisor

Re: PIM logs boot error

Thanks for all your help, I removed an unused fddi card and the system came back. I will run some further tests to see if any further errors occur.

Cheers
Michael
Lester Dias
Advisor

Re: PIM logs boot error

Michael,

This may not be related to your current problem but I have seen a very similar fault where the system craches after the 'alloc_pdc_pages' message. Mine was due to the console path not being correct. It looked very much like an I/O error and the fault is re-creatable - just change the console path in BCH and power cycle the system.

Perhaps on this occasion I'm barking up the wrong tree.

Lester