Memory problem on DL580 G7 512Gb

[user4test]

Hi.

We have some problem with memory on DL580G7.

After suddenly fail our server lose about 32Gb memory.

Our configuration is 8 memory riser x 8x8Gb on each and must be 512Gb.

So we have try mix,remap memory bank,risers. No error,no fail bank. But show only 480Gb memory available.

I attached screenshot of memery test. There writen 491520Mb Available  and 524288 Mb Configured.

What we missed or what we can do in this situation?
tnx

[Suman_1978]

Hi,

I would suggest you to use Memory Configurator to configure DIMM with processors.

http://h22195.www2.hp.com/MemoryTool/Home/Legal

HP ProLiant DL580 G7server technology

http://h20565.www2.hp.com/hpsc/doc/public/display?docId=emr_na-c02899605

Just make sure you have not done any configuration like RAID Memory or SPARE Memory in BIOS.

Thank You!
I am a HP employee
__________________________________________
Was the post useful? Click on the white KUDOS! Thumb below to say Thank You!

Useful Links for ProLiant Servers

[user4test]

Our server had been  in  working configuration before server crashed and rebooted. All slot in use by 8Gb modules and dl580g7 with 4 CPU.
No errors or fail modules. After mixing and relocating memory risers we locate one riser where only 32 gb in use, but we check 8 memory modules form this riser in other server, no error. All modules in working condition.
And we did not set any RAID Memory or SPARE Memory in BIOS.
Any ideas?

[sfora · Accepted solution]

Try booting from the latest SPP: www.hp.com/go/spp

On boot choose "Interactive update" , then you have the option to start Insight Diagnostics.

In that application you can read out all the hardware log files and check for errors. You can also launch a thorough memory test (it will take long!).

[user4test]

We did all test and have located two problem.
First one test failed - SPD Verify test.
Second - one module have error - Correctable error threshold exceeded.
I attach two screenshots.
But failed only one module and this only 8Gb minus not 32Gb.
Is it possible that failed only one module 8Gb and disabled 4x8G 32G for work?
I think that modules work as pair and only 16 Gb can failed from one bad module.

[sfora]

Hello, I think your best chance is to replace the failed dimm.

Clear all logs and launch the test again.

[user4test]

Thank you all for help. We have replaced problem modules and fix the problem. Thanks again!