rp5470 GSP infomation

[Hu Songyun]

All Dear Experts.

rp5470 Atten light on,and sel GSP log:

Log Entry # 0 :
SYSTEM NAME: ygapp1
DATE: 07/01/2010 TIME: 23:28:09
ALERT LEVEL: 2 = Non-Urgent operator attention required

SOURCE: 0 = unknown, no source stated
SOURCE DETAIL: 0 = unknown, no source stated SOURCE ID: FF
PROBLEM DETAIL: 0 = no problem detail

CALLER ACTIVITY: 6 = machine check STATUS: 2
CALLER SUBACTIVITY: 74 = implementation dependent
REPORTING ENTITY TYPE: 0 = system firmware REPORTING ENTITY ID: 01

0x1800102000FF6742 00000000 00000040 type 3 = Actual Data
0x5800182000FF6742 00006E06 01171C09 type 11 = Timestamp 07/01/2010 23:28:09
Type CR for next entry, Q CR to quit.



Log Entry # 1 :
SYSTEM NAME: ygapp1
DATE: 07/01/2010 TIME: 23:28:09
ALERT LEVEL: 2 = Non-Urgent operator attention required

SOURCE: 0 = unknown, no source stated
SOURCE DETAIL: 0 = unknown, no source stated SOURCE ID: FF
PROBLEM DETAIL: 0 = no problem detail

CALLER ACTIVITY: 6 = machine check STATUS: 2
CALLER SUBACTIVITY: 72 = implementation dependent
REPORTING ENTITY TYPE: 0 = system firmware REPORTING ENTITY ID: 01

0x0000102000FF6722 00000000 00000001 type 0 = Data Field Unused
0x5800182000FF6722 00006E06 01171C09 type 11 = Timestamp 07/01/2010 23:28:09
Type CR for next entry, - CR for previous entry, Q CR to quit.

there is no system error log in hp-UX.

so what is the information refer?

[Matti_Kurkela]

> CALLER ACTIVITY: 6 = machine check STATUS: 2

This says the problem was a "machine check". If the system did not reboot as a result, it was a Low-Priority Machine Check (LPMC).

LPMCs are parity errors in a processor or memory, which the ECC circuitry could successfully correct automatically, so that your data is not corrupted. An uncorrectable error would cause a HPMC, which would cause the system to crash and reboot.

Use the STM diagnostics (cstm/mstm/xstm). Run the Information tool on memory and CPUs; you'll probably find something in the "Memory Error Log Information" part of the CPU information and/or that the Page Deallocation Table's "PDT Entries Used" value is non-zero.

If the system detects that a RAM memory page produces errors, it will use the PDT to mark it as unusable. A small number of used PDT entries is relatively normal in an old server (within the limits of "expected wear & tear); but if the number of used PDT entries is large and/or keeps increasing, there might be a reason for concern.

If you have a support agreement, you might want to send the STM output to your hardware support contact for evaluation.

MK

[Hu Songyun]

o i see,thank you,MK.

[Prashanth.D.S]

Hi,

MCA indicates that the issue is due to a HW fault with the box, you should have a latest ts99 file under /var/tombstones. Attach the same here...

Best Regards,
Prashanth