DDR3 RAM failing High Frequency Raw Hammer Flips

[tech.monk]

Hi All,

I have Samsung 1333Mhz 8GB RAM sticks. For some reason it failed the RAW Hammer flips. I intend to trade this with 4gbx2.

However, I'm interested to know the impact this failure will have on ram and whether it should be sold or not.

Thank You.

[Edit]: Read the thread on Tom's Hardware - it answers my questions.

RAM may be vulnerable to high frequency row hammer bit flips

I bought a new(*edit: new for me, but used condition :p) RAM stick for my laptop, identical memory stick (exactly same model number), to get the best dual channel results. It's a 4GB stick of ram. (currently with 8GB) I ran memtest86, (not memtest86+), and I got this note: [Note] RAM may be...

forums.tomshardware.com

 

[salian]

This (row hammering) is a very specific hacking technique that has worked only in labs under controlled conditions and not been observed yet in any actual malware.

I imagine there are hundreds of known and unknown ways your system can be hacked, and unless you are doing something incredible with your system like running nuclear centrifuges for enriching uranium or hosting secret keys for a crypto exchange, I would not consider this as an actual issue to think about.

 

[enthusiast29]

I've read about this as well and the author/developer of memtest86 says that there are system with him in production which fail this test yet continue to run for years just fine.
Can you try and see if v4.3.7 I sent in PM works fine fully? That doesn't have this test.

I think this should be ignorable.

 

[tech.monk]

This (row hammering) is a very specific hacking technique that has worked only in labs under controlled conditions and not been observed yet in any actual malware.

I imagine there are hundreds of known and unknown ways your system can be hacked, and unless you are doing something incredible with your system like running nuclear centrifuges for enriching uranium or hosting secret keys for a crypto exchange, I would not consider this as an actual issue to think about.

Surprising - I have another pair of 2400 Mhz, Crucial DDR4 UDIMM and surprisingly it pass all the tests.

I will check with 4.3.7 version.

Tagging @rsaeon as well for his comment.