ArkanoiD wrote:
> I installed this patch (said to improve crash image saving performance
> or something).
> Now it recognizes every reboot as crash! It syncs disks, shuts the
> machine down, LCD displays state change - and then "Unexpected HPMC"
> message is shown and core is being saved.
>

Somebody is thinking the crash dump has not been saved.
Try clearing the dump flag (# savecrash -c) and see
what happens when you reboot the system.

--vishwas.

It says it was already saved (as expected - as savecrash is being run
at system startup)

Here is what tombstone says:

'9000/785 B,C,J Workstation HPMC PIM Analysis (per-CPU)', rev 0, 1304
bytes:

A Data I/O Fetch Timeout occurred while CPU 0 was
requesting information from a device at the path 10/0/14/2 (built-in
PCI device).


Memory/IO Controller Error Analysis Information:

The Memory/IO Controller only observed the Broadcast Error. It did
not log
any additional information about the HPMC.

On 26 мар, 08:37, Vishwas Pai wrote:
> ArkanoiD wrote:
> > I installed this patch (said to improve crash image saving performance
> > or something).
> > Now it recognizes every reboot as crash! It syncs disks, shuts the
> > machine down, LCD displays state change - and then "Unexpected HPMC"
> > message is shown and core is being saved.
>
> Somebody is thinking the crash dump has not been saved.
> Try clearing the dump flag (# savecrash -c) and see
> what happens when you reboot the system.
>
> --vishwas.

...looking into crash dump:
system messages:
---
....
sync'ing disks (0 buffers to flush):
0 buffers not flushed
0 buffers still dirty

Closing open logical volumes...
Done
---

Appear to shut down clean.

further by Q4:

+--------------------------------------------+
| Processor activity |
+--------------------------------------------+
Processor 0 started it by HPMC'ing. Here is the stack trace:
stack trace for event 0
crash event was an HPMC
can not find unwind or stub descriptor for
pc==0x0`01c9e2b4

+--------------------------------------------+
| HPMC Registers for HPA 0xfffffffffffa0000 |
+--------------------------------------------+
type: 20000000 state: 9E000004
cache: 00000000 tlb: 00000000 bus: 0030103B assists: 00000000
slave: F400400C mastr: FFFA0000

This is most likely a hardware problem, for further HPMC
troubleshooting,
please see: http://hphwec07.cup.hp.com/cpu.html


Processor 0: servicing interrupt
-------------
can not find unwind or stub descriptor for pc==0x0`01c9e2b4
-------------



+--------------------------------------------+
| Installed kernel patches |
+--------------------------------------------+
Kernel file is: vmunix
PHKL_31500 PHKL_31502 PHKL_31503 PHKL_31504 PHKL_31507
PHKL_31508
PHKL_31510 PHKL_31511 PHKL_31512 PHKL_31517 PHKL_32262
PHKL_32376
PHKL_32430 PHKL_32541 PHKL_32613 PHKL_32629 PHKL_32635
PHKL_32636
PHKL_32637 PHKL_32645 PHKL_32653 PHKL_32708 PHKL_32714
PHKL_32747
PHKL_32798 PHKL_32799 PHKL_32803 PHKL_32853 PHKL_32997
PHKL_33024
PHKL_33025 PHKL_33049 PHKL_33052 PHKL_33069 PHKL_33114
PHKL_33118
PHKL_33312 PHKL_33323 PHKL_33361 PHKL_33463 PHKL_33551
PHKL_33562
PHKL_33583 PHKL_33604 PHKL_33605 PHKL_33623 PHKL_33650
PHKL_33665
PHKL_33714 PHKL_33807 PHKL_33810 PHKL_33812 PHKL_33813
PHKL_33818
PHKL_33820 PHKL_33827 PHKL_33854 PHKL_33859 PHKL_33900
PHKL_33929
PHKL_33930 PHKL_33990 PHKL_34010 PHKL_34032 PHKL_34033
PHKL_34035
PHKL_34051 PHKL_34057 PHKL_34062 PHKL_34094 PHKL_34095
PHKL_34103
PHKL_34179 PHKL_34194 PHKL_34213 PHKL_34297 PHKL_34335
PHKL_34357
PHKL_34358 PHKL_34359 PHKL_34360 PHKL_34361 PHKL_34408
PHKL_34432
PHKL_34460 PHKL_34479 PHKL_34589 PHKL_34596 PHKL_34795
PHKL_34816
PHKL_34817 PHKL_34818 PHKL_34881 PHKL_34882 PHKL_34901
PHKL_34905
PHKL_34907 PHKL_34909 PHKL_34911 PHKL_34912 PHKL_34917
PHKL_34933
PHKL_34941 PHKL_35114 PHKL_35181 PHKL_35229 PHKL_35246
PHKL_35414
PHKL_35479 PHKL_35503 PHKL_35531 PHKL_35532 PHKL_35562
PHKL_35583
PHKL_35709 PHKL_35731 PHKL_35767 PHKL_35848 PHKL_35851
PHKL_35870
PHKL_35901 PHKL_35926 PHKL_35957 PHKL_36134 PHNE_31735
PHNE_31738
PHNE_32054 PHNE_32055 PHNE_32057 PHNE_32058 PHNE_32201
PHNE_32277
PHNE_33099 PHNE_33225 PHNE_33429 PHNE_33717 PHNE_33732
PHNE_33899
PHNE_33981 PHNE_33982 PHNE_34550 PHNE_35118 PHNE_35182
PHNE_35512
PHSS_32101
---

the link target on HP site seems to be down.

On 26 мар, 08:37, Vishwas Pai wrote:
> ArkanoiD wrote:
> > I installed this patch (said to improve crash image saving performance
> > or something).
> > Now it recognizes every reboot as crash! It syncs disks, shuts the
> > machine down, LCD displays state change - and then "Unexpected HPMC"
> > message is shown and core is being saved.
>
> Somebody is thinking the crash dump has not been saved.
> Try clearing the dump flag (# savecrash -c) and see
> what happens when you reboot the system.
>
> --vishwas.

ArkanoiD wrote:
>
> +--------------------------------------------+
> | Processor activity |
> +--------------------------------------------+
> Processor 0 started it by HPMC'ing. Here is the stack trace:
> stack trace for event 0
> crash event was an HPMC
> can not find unwind or stub descriptor for
> pc==0x0`01c9e2b4
>
> +--------------------------------------------+

That link is HP internal. CPU or some hardware is
causing the system to log a machine check. This
could be a H/W fault (most probable) or a software
issue (least probability).

==========
A Data I/O Fetch Timeout occurred while CPU 0 was
requesting information from a device at the path 10/0/14/2 (built-in
PCI device).


Memory/IO Controller Error Analysis Information:

The Memory/IO Controller only observed the Broadcast Error. It did
not log
any additional information about the HPMC.
==========

The above log shows that there is some dead card/
slot/I/O backplane etc. What card exists on 10/0/14/2 ?
Are using it ? If not - try shutdown-remove-reboot.

If that does not solve the problem, you should contact
HP-support and ask them to have look at the crash.
They may also do a HPMC analysis using the tombstone
files.

--vishwas.

Well, it is not likely to be a hardware problem as it did not appear
before i installed a patch set (can't remember exactly, but looks like
that one caused this behavior as i added it manually later since there
wre problem with authomatic routine).

10/0/14/2 is USB and it is functioning properly.

On 26 мар, 16:10, Vishwas Pai wrote:
> ArkanoiD wrote:
>
> > +--------------------------------------------+
> > | Processor activity |
> > +--------------------------------------------+
> > Processor 0 started it by HPMC'ing. Here is the stack trace:
> > stack trace for event 0
> > crash event was an HPMC
> > can not find unwind or stub descriptor for
> > pc==0x0`01c9e2b4
>
> > +--------------------------------------------+
>
> That link is HP internal. CPU or some hardware is
> causing the system to log a machine check. This
> could be a H/W fault (most probable) or a software
> issue (least probability).
>
> ==========
> A Data I/O Fetch Timeout occurred while CPU 0 was
> requesting information from a device at the path 10/0/14/2 (built-in
> PCI device).
>
> Memory/IO Controller Error Analysis Information:
>
> The Memory/IO Controller only observed the Broadcast Error. It did
> not log
> any additional information about the HPMC.
> ==========
>
> The above log shows that there is some dead card/
> slot/I/O backplane etc. What card exists on 10/0/14/2 ?
> Are using it ? If not - try shutdown-remove-reboot.
>
> If that does not solve the problem, you should contact
> HP-support and ask them to have look at the crash.
> They may also do a HPMC analysis using the tombstone
> files.
>
> --vishwas.

...a quick googling reveals similar problem exposed on 11.11, without
any explanations, though.

I use non-HP (a4tech) 7-button mouse and Genius USB keyboard -
original ones were lost years ago, btw..

http://www.openeyet.nl/scc/examples/scc.hpovou.html

On 26 мар, 16:10, Vishwas Pai wrote:
> ArkanoiD wrote:
>
> > +--------------------------------------------+
> > | Processor activity |
> > +--------------------------------------------+
> > Processor 0 started it by HPMC'ing. Here is the stack trace:
> > stack trace for event 0
> > crash event was an HPMC
> > can not find unwind or stub descriptor for
> > pc==0x0`01c9e2b4
>
> > +--------------------------------------------+
>
> That link is HP internal. CPU or some hardware is
> causing the system to log a machine check. This
> could be a H/W fault (most probable) or a software
> issue (least probability).
>
> ==========
> A Data I/O Fetch Timeout occurred while CPU 0 was
> requesting information from a device at the path 10/0/14/2 (built-in
> PCI device).
>
> Memory/IO Controller Error Analysis Information:
>
> The Memory/IO Controller only observed the Broadcast Error. It did
> not log
> any additional information about the HPMC.
> ==========
>
> The above log shows that there is some dead card/
> slot/I/O backplane etc. What card exists on 10/0/14/2 ?
> Are using it ? If not - try shutdown-remove-reboot.
>
> If that does not solve the problem, you should contact
> HP-support and ask them to have look at the crash.
> They may also do a HPMC analysis using the tombstone
> files.
>
> --vishwas.