Memory Single Bit Error

[Joelmel Roche]

Good Day! I'm enccounterring single bit error on my rp8400 server. Anybody have an idea how can i clear this error. Is this critical? what is the impact of this error on my server/system?

Memory Error Log Summary
Error
CAB/CELL DIMM Error Address Error Type Page Count
------------- ----------------- ---------- --------- -----
0/0 0A 0x000000015526f400 Single-Bit 0x015526f 1
0/3 2D 0x00000001840182c0 Single-Bit 0x0184018 1
0/3 2D 0x00000001810182c0 Single-Bit 0x0181018 1
0/3 2D 0x00000001fb0182c0 Single-Bit 0x01fb018 1
0/3 2D 0x00000001f90182c0 Single-Bit 0x01f9018 1
0/3 2D 0x00000001f80182c0 Single-Bit 0x01f8018 1
0/3 2D 0x00000001f60182c0 Single-Bit 0x01f6018 1
0/3 2D 0x00000001f50182c0 Single-Bit 0x01f5018 1
0/3 2D 0x00000001f40182c0 Single-Bit 0x01f4018 1
0/3 2D 0x00000001f20182c0 Single-Bit 0x01f2018 1
0/3 2D 0x00000001f10182c0 Single-Bit 0x01f1018 1
0/3 2D 0x00000001f00182c0 Single-Bit 0x01f0018 1
0/3 2D 0x00000001ee0182c0 Single-Bit 0x01ee018 1
0/3 2D 0x00000001e70182c0 Single-Bit 0x01e7018 1
0/3 2D 0x00000001e60182c0 Single-Bit 0x01e6018 1
0/3 2D 0x00000001e54182c0 Single-Bit 0x01e5418 1
0/3 2D 0x00000001e08182c0 Single-Bit 0x01e0818 1
0/3 2D 0x00000001dd0182c0 Single-Bit 0x01dd018 1
0/3 2D 0x00000001db0182c0 Single-Bit 0x01db018 1
0/3 2D 0x00000001d80182c0 Single-Bit 0x01d8018 1
0/3 2D 0x00000001d60182c0 Single-Bit 0x01d6018 1
0/3 2D 0x00000001d5c182c0 Single-Bit 0x01d5c18 1
0/3 2D 0x00000001ca0182c0 Single-Bit 0x01ca018 1
0/3 2D 0x00000001c84182c0 Single-Bit 0x01c8418 1
0/3 2D 0x00000001c80182c0 Single-Bit 0x01c8018 1
0/3 2D 0x00000001c60182c0 Single-Bit 0x01c6018 1
0/3 2D 0x00000001c3c182c0 Single-Bit 0x01c3c18 1
0/3 2D 0x00000001c30182c0 Single-Bit 0x01c3018 1
0/3 2D 0x00000001800182c0 Single-Bit 0x0180018 1
0/3 2D 0x00000001c00182c0 Single-Bit 0x01c0018 1
0/0 0B 0x0000000fd295d800 Single-Bit 0x0fd295d 9
0/0 0D 0x0000000edea18000 Single-Bit 0x0edea18 1
0/1 3a/b/c/d 0x0000000000281341 Single-Bit 0x0000281 0
0/1 3a/b/c/d 0x0000000000280341 Single-Bit 0x0000280 0

[Jeeshan]

read this link

http://forums12.itrc.hp.com/service/forums/questionanswer.do?admit=109447627+1212466477894+28353475&threadId=1204819

[Steven Schweda]

> Anybody have an idea how can i clear this
> error.

Replace the bad memory?

> Is this critical?

Depends on what you call critical. How
important are your data?

> what is the impact of this error on my
> server/system?

You're wasting time and disk space handling
and logging the errors, and, while ECC memory
can correct single bit errors, it normally
can only _detect_ two-bit errors, so most
people would elect to replace the failing
memory module before it gets worse (and data
loss becomes more likely).

[Steven Schweda]

> read this link

When posting ITRC forum links, it would be
good to remove the useless bits, like "11" or
"12" from the host name, and all the
(cookie-like) "admit=XXX" stuff, so:

http://forums.itrc.hp.com/service/forums/questionanswer.do?threadId=1204819

[Torsten.]

Read my comments here:

http://forums.itrc.hp.com/service/forums/questionanswer.do?threadId=1049776

Short summery:

- update online diagnostics
- reboot the server
- monitor the errors

[Ron Irving]

Single-bit memory errors are usually fine until the system starts to de-allocate the memory due to too many of them. I would suggest monitoring STM, and eyeballing which set of DIMMs de-allocate.

Just my 4c worth.

[TTr]

Might these errors be the result of

http://forums.itrc.hp.com/service/forums/questionanswer.do?threadId=1235962

It's a lot of them and across the board, most of them on 0/3 DIMM 2D.

If so does that mean the memory was not compatible between the two machines? Or was it bad handling during the transfer?