Is my 960 Evo dying? (becomes RAW but comes back after restart)

[Acquire]

So, my PC is setup with two samsung 960 evo ssds on a asus maximus IX hero motherboard.  I have the motherboard set to have both m.2 slots operate at pcie 3.0 x4 mode.  First drive is a 250GB.  This is the boot drive, has windows on it, along with all software except games and game launchers.  Game launchers and games are on the second drive, which is a 1TB drive.  The drives are on their latest firmware.  Latest drivers all around.  I've had this setup for almost two years.

 

Recently I've been having the problem where after like 40 hours of the PC being on, any program running off the second drive (so a game while I'm playing it) crashes.  Then the game launcher crashes.  I've been playing a lot of assassin's creed odyssey, so the game will crash, then uplay crashes, then steam crashes.  All these are on the second drive.  I can never get any of these to launch again successfully.  I can still see the drive on windows explorer and it shows up as ntfs.  I can navigate through most of the folder structure, but some give me some "invalid function" error when I try to access them.  In rare cases steam stays alive, but when I try to start a game it says "invalid platform" in the error box.  Windows' Disk Management utility shows the drive still there and as NTFS. However, when I try to do a chkdsk on it, it says it's of file type RAW.  If I restart windows the drive is back and everything's working again.  I performed a full chkdsk scan on it and it says there's nothing wrong with it.  Samsung's Magician software's SMART monitoring thing also says nothing's wrong.  It'll continue to work for about two days if that until it fails again. Restarting windows seems to always fix it.

 

So, seems like I should just call samsung and have it replaced? Still is under their warranty.  Just wondering if anyone's encountered a similar problem.  The confusing part is how it "fixes" itself after a restart, chkdsk says nothing's wrong.   I have nothing of importance in this drive.  It's just there for game install data.  This is purely a gaming machine so even on the main drive I also have nothing of importance to lose since all game saves are backed up to some or many clouds.

[GirlFromYonder]

any bclk overclocking?

[Acquire]

11 minutes ago, GirlFromYonder said:

any bclk overclocking?

No.  All I have setup in terms of overclocking is my ram's xmp profile.  Had that on for years though.

[redteam4ever]

I was waiting to see some other replies and I forgot about your topic a bit, I'm sorry. I have a theory - your board have M.2 slots positioned one on top of another. Because of that, I think that the drive with games is in the bottom slot and it's overheating - there is another drive that is heating it up and blocking air from cooling it. That's why it happens only when you play games and takes some time for the drive to heat up so that it malfunctions. When you restart Windows, there is no access to the second drive, so it cools off.

 

You also mentioned that Samsung Magician Software doesn't report any problems. Temperature is monitored by S.M.A.R.T., so either Samsung Magician doesn't read the temperature, or doesn't see it as a problem. I would try running other diagnostic software, such as CrystalDiskMark and get access to all the data (I don't know Samsung Magician but manufacturer utilities often don't show all or raw data). Other solution is to swap the drives between the 2 slots, to see if anything changes.

[Acquire]

On 1/19/2019 at 6:18 PM, redteam4ever said:

I was waiting to see some other replies and I forgot about your topic a bit, I'm sorry. I have a theory - your board have M.2 slots positioned one on top of another. Because of that, I think that the drive with games is in the bottom slot and it's overheating - there is another drive that is heating it up and blocking air from cooling it. That's why it happens only when you play games and takes some time for the drive to heat up so that it malfunctions. When you restart Windows, there is no access to the second drive, so it cools off.

 

You also mentioned that Samsung Magician Software doesn't report any problems. Temperature is monitored by S.M.A.R.T., so either Samsung Magician doesn't read the temperature, or doesn't see it as a problem. I would try running other diagnostic software, such as CrystalDiskMark and get access to all the data (I don't know Samsung Magician but manufacturer utilities often don't show all or raw data). Other solution is to swap the drives between the 2 slots, to see if anything changes.

My motherboard does not have that layout and I'm not aware of any that do.  It's an Asus Maximus IX Hero. The m.2 slot where this drive is failing is the bottom one under the GPU.  If anything it's the best placement as it gets plenty of airflow from the front intake fans and the intake fans of the GPU sucking in air.  At most it would throttle down and not entirely die.  I had samsung look at it and they said nothing was wrong with it.   Still dies after a few hours.   I'm thinking at this point it's a motherboard problem.

[redteam4ever]

2 hours ago, Acquire said:

My motherboard does not have that layout and I'm not aware of any that do.  It's an Asus Maximus IX Hero. The m.2 slot where this drive is failing is the bottom one under the GPU.  If anything it's the best placement as it gets plenty of airflow from the front intake fans and the intake fans of the GPU sucking in air.  At most it would throttle down and not entirely die.  I had samsung look at it and they said nothing was wrong with it.   Still dies after a few hours.   I'm thinking at this point it's a motherboard problem.

Oh, sorry then. I was looking at the board schematic on the ASUS page and there was a caption on the top M.2 slot that says "Dual M.2 slots". Now I see that there is another slot near the bottom of the board, I didn't notice the other slot. It seemed really weird to me why would they place the slots on top of each other, but I couldn't find the second slot back then.

 

I was following this topic to find out what's the problem with your SSD, because I also have a Samsung SSD that was in my laptop when I bought it. It's a Samsung PM841 128GB  (according to some discussions it's an OEM version of 840 PRO) and it was also acting up. Mine is out of warranty and it's slowly dying (SMART shows 70% remaining life), but it's been acting up similarly. I relocated it to my PC when I bought a new SSD for my laptop and then my PC wouldn't want to boot into Windows at random times with blue screens that some system files were missing or I/O errors. Sometimes even while the PC was running, the blue screens would pop up with random error codes that were undefined. I finally solved it by replacing the SSD, but most of the time it works fine...

 

For now, I'm out of ideas on what the problem is

[Acquire]

8 hours ago, redteam4ever said:

Oh, sorry then. I was looking at the board schematic on the ASUS page and there was a caption on the top M.2 slot that says "Dual M.2 slots". Now I see that there is another slot near the bottom of the board, I didn't notice the other slot. It seemed really weird to me why would they place the slots on top of each other, but I couldn't find the second slot back then.

 

I was following this topic to find out what's the problem with your SSD, because I also have a Samsung SSD that was in my laptop when I bought it. It's a Samsung PM841 128GB  (according to some discussions it's an OEM version of 840 PRO) and it was also acting up. Mine is out of warranty and it's slowly dying (SMART shows 70% remaining life), but it's been acting up similarly. I relocated it to my PC when I bought a new SSD for my laptop and then my PC wouldn't want to boot into Windows at random times with blue screens that some system files were missing or I/O errors. Sometimes even while the PC was running, the blue screens would pop up with random error codes that were undefined. I finally solved it by replacing the SSD, but most of the time it works fine...

 

For now, I'm out of ideas on what the problem is

 

I don't want to deal with the mess of replacing the motherboard so I dusted off a Samsung 1TB 860 Evo SSD I had lying around and am using that now.  Seems to be working.  As long as it doesn't crash