Memtest86 10.4 hangs

[Pompon]

Hi,

Memtest86 10.4 (and 10.3) Free hangs on Ram benchmark with “read” and “per block size”,
(not sure of the relevancy) in spite of whether All/Single core is selected and in spite of ECC enabled/disabled

With “per access step” it works fine though.
Xeon E5 2695 v3, 64Gb, 8xDDR-2400 (72 bit)
Are there any other details I could add?

Thank you

[Richard (PassMark)]

Can you send in the debug log and we can take a look?

https://www.memtest86.com/tech_debug-logs.html

[Pompon]

Thank you!
https://drive.google.com/file/d/1qJO...usp=share_link

[David (PassMark)]

We'll take a look at the logs. It might be a couple of days.

[keith]

Hi,

Memtest86 10.4 (and 10.3) Free hangs on Ram benchmark with “read” and “per block size”,
(not sure of the relevancy) in spite of whether All/Single core is selected and in spite of ECC enabled/disabled

With “per access step” it works fine though.
Xeon E5 2695 v3, 64Gb, 8xDDR-2400 (72 bit)
Are there any other details I could add?

Thank you

We added some debug logging to the RAM benchmarks. Can you give this build a try:
https://www.passmark.com/temp/memtes...-10.4.1003.zip

[Pompon]

Hi,
https://drive.google.com/file/d/1nmo...usp=share_link

Thank you

[keith]

Hi,
https://drive.google.com/file/d/1nmo...usp=share_link

Thank you

Thanks for the logs.

We think we may have fixed the issue. Can you give this build a try:
https://www.passmark.com/temp/memtes...-10.4.1004.zip

[Pompon]

Thanks a lot!
The test doesn't hang

[Rudolf]

Hi! I have a similar problem,
Sysinfo:
CPU: 2x intel Xeon Gold 6248R
RAM: DDR4 16x64Gb Samsung M393A8G40AB2-C (3200MHz), but motherboard supported 2933MHz
I'm attaching the photo and logs.

Thank you!

[David (PassMark)]

Rudolf, your issue seems to be unrelated to the initial post. (i.e. it doesn't seem to be related to the Ram benchmark function).

So it might be a real hardware failure.

We'll take a look at the log.

[keith]

Hi! I have a similar problem,
Sysinfo:
CPU: 2x intel Xeon Gold 6248R
RAM: DDR4 16x64Gb Samsung M393A8G40AB2-C (3200MHz), but motherboard supported 2933MHz
I'm attaching the photo and logs.

Thank you!

The logs don't seem to be consistent with what is shown in the screenshot. As the logs indicate it is in the middle of test #4 but the screenshot is showing that it is running test 13.

Also, the logs indicate that the system is testing 1TB RAM in round robin (ie. single threaded mode). This will take an especially long time to execute.

Has the UI froze or is it just taking a long time?
Have you tried running in parallel (ie. multi-threaded mode) for quicker execution?

[Rudolf]

The logs don't seem to be consistent with what is shown in the screenshot. As the logs indicate it is in the middle of test #4 but the screenshot is showing that it is running test 13.

Also, the logs indicate that the system is testing 1TB RAM in round robin (ie. single threaded mode). This will take an especially long time to execute.

Has the UI froze or is it just taking a long time?
Have you tried running in parallel (ie. multi-threaded mode) for quicker execution?

Sorry these are logs from another identical server, attached the correct logs.

[Rudolf]

The logs don't seem to be consistent with what is shown in the screenshot. As the logs indicate it is in the middle of test #4 but the screenshot is showing that it is running test 13.

Also, the logs indicate that the system is testing 1TB RAM in round robin (ie. single threaded mode). This will take an especially long time to execute.

Has the UI froze or is it just taking a long time?
Have you tried running in parallel (ie. multi-threaded mode) for quicker execution?

UI froze, did not try to run in parallel.

[keith]

Thanks for the logs.

Can you try running Test 13 by itself and see if the issue still occurs.

We are also looking at reducing the test time for Test 13 especially for systems with large amount of RAM.