Skip to content
View in the app

A better way to browse. Learn more.

Unraid

A full-screen app on your home screen with push notifications, badges and more.

To install this app on iOS and iPadOS
  1. Tap the Share icon in Safari
  2. Scroll the menu and tap Add to Home Screen.
  3. Tap Add in the top-right corner.
To install this app on Android
  1. Tap the 3-dot menu (⋮) in the top-right corner of the browser.
  2. Tap Add to Home screen or Install app.
  3. Confirm by tapping Install.

NVMe overheating

Featured Replies

Interresting

  • Replies 59
  • Views 13.7k
  • Created
  • Last Reply

Top Posters In This Topic

Most Popular Posts

  • As will everyone using a Samsung 980 NVMe device, until they fix it.

  • I have a heatsink on mine and it makes a HUGE difference

  • Tested with the above command whilst extracting sensor data. It does indeed report increasing temperatures:   Composite: +42.9°C (low = -273.1°C, high = +81.8°C) Sensor 1: +42.9°C (l

Posted Images

Would this indicate that I should revert back to rc1?

  • Author

I have reverted and the problem has gone away. If there are any other ideas to try to troubleshoot this I'm very happy to help.

 

-Brian

I'm having the exact same issue with a z490 and Samsung 980. I have a massive heatsink with thermal bad and the temps go from 39C to 84C randomly. The actually drive is not very hot, so this seems to be an software issue reading the temps. 

  • Author

Hi all. For fun I took the cover off my server and then installed rc2. As soon as I got a notice that the NVMe went to 84C I went to go touch it. I can tell you that it was NOT at 84 (otherwise I'd be in pain!)

 

This seems to point to a sensor read error in rc2. 

I also noticed it stays at 84c for exactly 1 hour. I have Pushbullet send me notifications and it's consistently going from 39c to 84c for 1 hour before dropping back to 39c. 

31 minutes ago, macieksoft said:

I also noticed it stays at 84c for exactly 1 hour. I have Pushbullet send me notifications and it's consistently going from 39c to 84c for 1 hour before dropping back to 39c. 

Samsung 980 ?

1 minute ago, Vr2Io said:

Samsung 980 ?

Ya, the regular 980 (not the pro). 

On 11/5/2021 at 10:23 AM, MaxwellHouse said:

Well, after reverting I had no issues. So just to make sure I went back to rc2 but this time I ran "Update Assistant" before going from rc1 to rc2. It's been up for about 90 minutes without any issues.

 

Told you :)

  • Author
30 minutes ago, johnwhicker said:

 

Told you :)

 

Yeah, I went back to rc2 and I'm getting the same problem. But, we've determined it's a sensor read error on Samsung 980 NVMe drives. So, for now, I'm ignoring the warning!

41 minutes ago, MaxwellHouse said:

 

Yeah, I went back to rc2 and I'm getting the same problem. But, we've determined it's a sensor read error on Samsung 980 NVMe drives. So, for now, I'm ignoring the warning!

You could always cook same eggs :)  Kidding 

Just a weigh in that I am having exactly the same issue. RC2, Asus x570 MB, 5700g, NVMe is a Samsung 980.

 

Drive is cool when it is reporting 84 degrees. Drive logs show around 36-38 during this period too.

 

Two (perhaps) differences from my setup to the others in the thread: (1) Have a heatsink on the NVME SSD already. (2) NVMe is being used as a passthrough drive for a VM - it is not assigned to an array or pool.

 

So definitely seems to be an issue with RC2.

  • Author
On 11/10/2021 at 11:35 PM, macieksoft said:

I also noticed it stays at 84c for exactly 1 hour. I have Pushbullet send me notifications and it's consistently going from 39c to 84c for 1 hour before dropping back to 39c. 

 

What size is your 980? I have a 500 GB drive and noticed that it reports 84C for exactly 30 minutes!

9 minutes ago, MaxwellHouse said:

 

What size is your 980? I have a 500 GB drive and noticed that it reports 84C for exactly 30 minutes!

I have the 1TB version. 

1TB - about 30 mins for the issue sounds about right.

  • 2 weeks later...

I have a solve, or at least can point to the cause of this issue. It's something to do with this specific NVMe drive. Not sure if it is the lack of DRAM or the new controller on the 980, but whatever it is, that is what is causing this 30 min period of reported overheating.

 

I just switched out NVME drives (in this case to the 980 Pro) and the issue is gone. No other change in configuration at all. Even using the same M2 slot on the Motherboard. Been running solidly like this over 48 hours now.

 

I'm not sure if this is something that Unraid can fix, or if it will get better with newer releases that deal with this new style of NVMe drive more correctly. 

 

One parting thought on this. I don't believe the 980 actually ever got to 84 degrees or that it overheated at all. I think it was just misreporting. I made sure to be there a couple of times when it was reporting the overheating and the drive felt cool to the touch. So I don't think over-time, this problem will cause any issues to the physical drive, but since this was a mission critical drive for me, I didn't like the issue and wanted to be sure to get onto something which works as it should. 

Edited by V1per5h0t

  • Author
2 hours ago, V1per5h0t said:

I just switched out NVME drives (in this case to the 980 Pro) and the issue is gone. No other change in configuration at all. Even using the same M2 slot on the Motherboard. Been running solidly like this over 48 hours now.

 

Ok, so you've talked me into it... :D I just bought a 980 Pro 1TB NVMe. So, what's the best way to replace my 500 Gb NVMe drive? Should I do an rsync to another mount, switch out the drive, let Unraid do it's formatting thing, and then rsync back? I use this guy for all of my Docker containers.

 

EDIT: Never mind, I found this wiki. 

 

Thanks!

Brian

Edited by MaxwellHouse

  • 1 month later...

Wanted to add I have two of the samsung 980 nvme SSDs in an NVME pool on an asus hyper m.2 card and am running the current RC2. The 980s read 84C for 30 minutes then go back to normal. The other two NVME drives on the card are usually at 25-35C even under load. This was never a problem in the several previous versions of unraid. Hopefully this error is fixed in the future. 

  • 2 months later...

Just to add, I'm also experiencing this bug in RC2/3/4 with a Samsung 980 1TB. Getting the same as others where it reports 84C for 30 minutes. I'm glad I searched and found this thread, I'm less concerned now and assume it's mis-reporting the value.

Same here,

I have 4 1TV Samsung 980 nvme in an ASUS Hyper M.2 x16 Gen 4 Card.
Massive cooling with fan, so there's no way it could spike to 84°C. I run this set as a btrfs-raid 10 so the likeness that if there's overheating only 1 of the 4 would overheat is very, very low.

It must be a software error.


I run 6.10rc4

 

If I switch back to 6.9 no issues.

 

Any suggestions from unraid ?

  • 2 months later...

I am having the exact issue.  Exact

  • Community Expert
8 hours ago, vyral said:

I am having the exact issue.

As will everyone using a Samsung 980 NVMe device, until they fix it.

  • 1 month later...

Reporting same issue , glad to see it seems to be a bug 

  • 4 weeks later...

Here to report that I am having the same issue and config.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

Account

Navigation

Search

Search

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions → Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.