rp4440 /11.11/ crashed and rebooted

[moonchild]

shutdown log shows reboot

attached is the ts99

the mp sl log shows:

375 BMC 2 0x2048A6D9360228A0 FFFF027000120300 Type-02 127002 1208322
16 Aug 2008 13:42:14
376 BMC 2 0x2048A6D93E0228B0 FFFF0103FDC00300 Type-02 c00301 12583681
16 Aug 2008 13:42:22
377 SFW 3 *7 0xE880035C03E028C0 0000000000000000 ERR_CHECK_HPMC
26 Aug 2008 05:53:48
378 3 2 0x57800F7303E028E0 8040004000000000 IPMI Type-E0 Event
26 Aug 2008 05:53:48
379 SFW 3 *7 0xF680105E03E02900 000000003F901000 MC_HPMC_MONARCH_SELECTED
26 Aug 2008 05:53:48
380 SFW 3 2 0x5680100B03E02920 000000000000003A MC_OS_HPMC_LEN_ERR
26 Aug 2008 05:53:49
381 SFW 3 2 0x5680106403E02940 FFFFFFF0F043FD40 MC_BR_TO_OS_HPMC_FAILED
26 Aug 2008 05:53:49
382 BMC 2 0x2048B3A08D022960 0783A37000120300 Type-02 127003 1208323
26 Aug 2008 06:19:57
383 BMC 2 0x2048B3A08D022970 FFFF027000120300 Type-02 127002 1208322
26 Aug 2008 06:19:57
384 BMC 2 0x2048B3A08D022980 FFFF027000120300 Type-02 127002 1208322
26 Aug 2008 06:19:57
385 BMC 2 0x2048B3A0AA022990 FFFF0103FDC00300 Type-02 c00301 12583681
26 Aug 2008 06:20:26
386 BMC 2 0x2048B3A0AC0229A0 FFFF027000120300 Type-02 127002 1208322
26 Aug 2008 06:20:28
387 BMC 2 0x2048B3A0B00229B0 FFFF0103FDC00300 Type-02 c00301 12583681
26 Aug 2008 06:20:32

thank you in advance

[Vihang.]

Its an HPMC. Better open a call with HP & let them check.


Regards,
Vihang.

[Phil uk · Accepted solution]

Moonchild,

Unfortunately the HPMC looks like its invalid:

Inconsistent timestamps detected.
HPMC data is probably corrupt.
- Analysis possibly meaningless
Time Span: 31 day(s), 08 hour(s), 20 minute(s), 38 second(s)

Having looked at it anyway, it does seem to be inconclusive as it points to various areas within the system and hence can't be relied upon.

Sorry

Regards,
Phil

[Stefan Stechemesser]

Hi Moonchild,

a small tip: assign points to people who help you, then they might be more responsive next time ;-)

The HPMC seems to be a problem with an access to hw path 0/0 (USB, serial ports and IDE controller). A lookup to an I/O address during a DMA transaction from this path failed.
Possible causes can be software (OS drivers) or hardware (everything that could corrupt an address => CPUs, ZX1 chip (CPU board), IO backplane, PCI card etc.).
I would recomend to first install the latest OS patches and firmware and / or run offline diagnostics for CPU, memory and IO if the same problem happens again.

best regards

Stefan

[moonchild]

Stefan, I always assign points. If i forgot once i will go through my questions and assign where i missed.

thank you for this note and thanks for the helpful answer to my question.

best regards
moon

[Phil uk]

Moon,

As an afterthought, it may be worth clearing the PIM and PDT from the BCH menu when you next have downtime.
This will clear out all the old stuff so that any future crashes (if indeed it re-occurs) will hopefully dump some better information to work from.

From the BCH> Service Menu, you can clear the PIM and PDT.

Regards,
Phil