rp3440 boot problems

[Ranjith_5]

Hi Gurus,
The error is 0x438010EF00E00180 00000000FFFFFFFF MEM_UNEXPECTED_HPMC

One of my friend faces problems in one of the rp3440 servers. It fails to boot intermittenly. and once it is booted works perfectly. The system is brought to shutdown every evening and when morning when trys to start it gives boot problem. he system event log shows 0x438010EF00E00180 00000000FFFFFFFF MEM_UNEXPECTED_HPMC
and i told my friend that this could be a processor related issue. Can some one tell me what the exact problem could be.

Pls see the attachment as well.

Regards,
Syam

[Eric Antunes]

Hi Syam,

It seems to me a hardware issue: your friend should consider a call to HP...

Best Regards,

Eric Antunes

[Pete Randall]

HPMC means High Priority Machine Check - most likely hardware related - you should have HP take a look at it.


Pete

[HGN]

Hi

The HPMC errors are mostly hardware related,need to open a case with HP and send them the details and they would be able to debug and would let you know if anything needs to be replaced.

Rgds

HGN

[Kent Ostby]

Syam --

It means he has a hardware problem. Could very possibly be a CPU, but you would need to get the /var/tombstones/ts99 file and have an HP engineer look at it to verify.

Oz

[Sameer_Nirmal · Accepted solution]

Hi,

The event "0x438010EF00E00180 00000000FFFFFFFF MEM_UNEXPECTED_HPMC" indicates an unexpected High Priority Machine Check (HPMC) has encountered during memory intialising routine and reported by SFW (system firmware).

It is also better to look for FPL ( Forward Progress Log ) with keyword/text format to get more information. Saying that the exact problem/cause needs to be decoded from the info which could be done by HP only.

Ask your friend to log a H/W call to HP , provide them all MP logs ,syslogs , dmesg, /var/tombstones/ts99 ( assuming HPMC occured sometime under OS OE )


I believe the event "CPU STOP" is just a informal log as part of system intialisation and indication of forward progress.

I do suspect the problem with hardware , no doubt about it as mostly the cause of HPMC is malfuncationing hardware. It may be one of the memory module or other memory related board hardware.

[Arunvijai_4]

Syam, Better to call HP and log a support call. It seems to be a hardware issue.

-Arun

[Ranjith_5]

Hi Friends

Thanks for all your replys. I understand that it is an HPMC error, which can be due to a procesor or a component which is related to processor. My question was, is there any method to find out which component has failed. Is there any way to get a hint about the malfunctioning component?

Regards,
Syam

[Ranjith_5]

the memory modules were already replaced and tested but the problem exists.

[Sivakumar TS]

Dear Syam,

Since this is a HPMC, mostly it would be a HW issue only.

Could you please collect the following file from the server? This will be created during each HPMC. you can also find the old HPMC logs ts98,ts97 etc. The ts99 file is the latest.

/var/tombtones/ts99

We can decode to some extent if we have this file.

With Regards,

Siva