RAM: total 16 GB, but only 8 GB available

parkejunior

Hello everyone, how are you?

This is the first time i've posted on a forum because i've always tried to resolve things on my own, but this time I don't know the way. I would be grateful for any help you can give me.

Recently i noticed that my PC has been freezing when it reaches 8GB of RAM usage, but i have 2 x 8 GB RAM modules in dual channel totaling 16 GB. I don't really know when this started, but i think it's been happening for the last month at least.

I'm using SolusOS 4.5 with Budgie 10.9.1.

The first thing i did was check the available RAM in the BIOS and it shows 16384Mb, even when i open the Budgie Control Center, 16 GB appears:

However, when i check the RAM availability in System Monitor, only 8 GB are available:

Searching a little i found some commands that show my memory data like "sudo dmidecode -t memory":

`
dmidecode 3.5
Getting SMBIOS data from sysfs.
SMBIOS 3.2.1 present.

Handle 0x0009, DMI type 16, 23 bytes
Physical Memory Array
Location: System Board Or Motherboard
Use: System Memory
Error Correction Type: None
Maximum Capacity: 256 GB
Error Information Handle: 0x0008
Number Of Devices: 4

Handle 0x0010, DMI type 17, 40 bytes
Memory Device
Array Handle: 0x0009
Error Information Handle: 0x000F
Total Width: Unknown
Data Width: Unknown
Size: No Module Installed
Form Factor: Unknown
Set: None
Locator: DIMM 0
Bank Locator: P0 CHANNEL A
Type: Unknown
Type Detail: Unknown

Handle 0x0012, DMI type 17, 40 bytes
Memory Device
Array Handle: 0x0009
Error Information Handle: 0x0011
Total Width: 64 bits
Data Width: 64 bits
Size: 8 GB
Form Factor: DIMM
Set: None
Locator: DIMM 1
Bank Locator: P0 CHANNEL A
Type: DDR4
Type Detail: Synchronous Unbuffered (Unregistered)
Speed: 2666 MT/s
Manufacturer: Unknown
Serial Number: 00000018
Asset Tag: Not Specified
Part Number: MSD48G26Q3
Rank: 1
Configured Memory Speed: 2666 MT/s
Minimum Voltage: 1.2 V
Maximum Voltage: 1.2 V
Configured Voltage: 1.2 V

Handle 0x0015, DMI type 17, 40 bytes
Memory Device
Array Handle: 0x0009
Error Information Handle: 0x0014
Total Width: Unknown
Data Width: Unknown
Size: No Module Installed
Form Factor: Unknown
Set: None
Locator: DIMM 0
Bank Locator: P0 CHANNEL B
Type: Unknown
Type Detail: Unknown

Handle 0x0017, DMI type 17, 40 bytes
Memory Device
Array Handle: 0x0009
Error Information Handle: 0x0016
Total Width: 64 bits
Data Width: 64 bits
Size: 8 GB
Form Factor: DIMM
Set: None
Locator: DIMM 1
Bank Locator: P0 CHANNEL B
Type: DDR4
Type Detail: Synchronous Unbuffered (Unregistered)
Speed: 2666 MT/s
Manufacturer: Unknown
Serial Number: 00000000
Asset Tag: Not Specified
Part Number: MSD48G26Q3
Rank: 1
Configured Memory Speed: 2666 MT/s
Minimum Voltage: 1.2 V
Maximum Voltage: 1.2 V
Configured Voltage: 1.2 V
`

The information correctly returns the position of my RAMs, but when i use the "free -h" and "sudo inxi -m" commands, only 8 GB are shown as available:

total used free shared buff/cache available Mem.: 7,7Gi 3,0Gi 3,1Gi 311Mi 1,6Gi 4,2Gi Swap: 3,7Gi 0B 3,7Gi

Memory: System RAM: total: 16 GiB available: 7.71 GiB used: 3.55 GiB (46.1%) Array-1: capacity: 256 GiB note: check slots: 4 modules: 2 EC: None Device-1: Channel-A DIMM 0 type: no module installed Device-2: Channel-A DIMM 1 type: DDR4 size: 8 GiB speed: 2666 MT/s Device-3: Channel-B DIMM 0 type: no module installed Device-4: Channel-B DIMM 1 type: DDR4 size: 8 GiB speed: 2666 MT/s

I'm sorry if someone has already asked this on the forum, i looked but couldn't find it. I would be very grateful if you could show me a way 🙂

ReillyBrogan

This sort of issue often happens when one memory stick is going bad. I'd recommend running memtest on the hardware.

parkejunior

ReillyBrogan I hadn't thought about that possibility yet, i'll test them and tell the results. Thanks for the answer 🙂

parkejunior

I carried out the tests and found the problem!

First I started Memtest86 as recommended, and with the memories in the same position that were causing the problem. My motherboard has 4 DDR4 slots, which are:

Channel A-0 (Black)
Channel A-1 (Red): Memory 1
Channel B-0 (Black)
Channel B-1 (Red): Memory 2

Only 8GB appeared in memtest86, just like in the OS, but the test passed.
I decided to test each memory individually, so I keep memory 1 in the same position and removed memory 2, and the test also passed:

Channel A-0 (Black)
Channel A-1 (Red): Memory 1
Channel B-0 (Black)
Channel B-1 (Red)

I noticed that channel A-1 was working normally, so I put memory 2 on that channel:

Channel A-0 (Black)
Channel A-1 (Red): Memory 2
Channel B-0 (Black)
Channel B-1 (Red)

However, the bios didn't start, and as I don't have a buzzer I couldn't know if it was beeping, but I think so. At that moment I thought my memory had died, but I still wanted to perform the test using the black channels, so I put memory 2 on channel A-0:

Channel A-0 (Black): Memory 2
Channel A-1 (Red)
Channel B-0 (Black)
Channel B-1 (Red)

For some reason, the CMOS reset, but the bios started normally. I put it to the test and it passed. I placed memory 1 on the other black channel and started Solus to check if both memories were available, and indeed both were working:

It turns out that in the first week of the year I had a problem with my SSD and I was forced to reinstall Solus. That day I took the opportunity and cleaned the components, and in doing so I switched the memories from the black channels to the red channels, but I hadn't noticed any problems until then.

When I noticed this problem, I even tried to remove the memories and clean them, this even resolved it for a few days, but then it happened again, that's when I thought it was a software problem and not a hardware problem.

Thank you very much for the answer @ReillyBrogan .

ReillyBrogan

FYI the key thing that made me think it was a hardware issue was the fact that it started having issues as soon as you passed 8GB of usage. IE, once you started using more than one RAM stick of memory and the OS started to use the next RAM stick (the bad one, though bad by way of motherboard rather than the stick itself being bad).

OS-level memory issues tend to be random and erratic. Hardware-level memory issues tend to be very defined and happen at the point almost every time (bad sectors could have issues as soon as those sectors were being written to, bad RAM/mobo traces when you start using a different RAM stick).

Happy to hear you were able to get it figured out!

SilverBird775

Each CPU memory channel probes sticks for actual supported timings.
Chances are, the RAM sticks have different access timings and when both put on the same channel one of them could pick up the other stick unsupported timings. Splitting them off to different channels should always be safe.

brent

I wish you could flag an entire thread 'thumbs up' or 'heart' because I found all this information fascinating, especially a reader with older hardware like me who always has an antennae up for those subtle hardware signs.

parkejunior

@ReillyBrogan , that makes a lot of sense, now I know that generally hardware issues happen in more specific and defined situations than software ones. This is very useful, thanks for the info

Thank you all for the answers!