Currently having an issue where all of my disks will throw a "Alert [] - Disk X overheated (* C)" (with temperatures between 56 and 70C) with a "Returned to normal temperature" shortly after. My thresholds are set at 75 (warning) and 82 (critical). These are HGST 10K disks that run a bit hot, but I have manual lower thresholds set for my SSDs (and these do not have any issues with errant alerts). Currently running unRAID 6.9-beta30 but I'm not having any issue on my other 2 unRAID servers also running 6.9-beta30 (running the beta for multiple pools of 10K drives).

 

Attached is my diagnostics file. It's not a high priority because the disks are "normal" temperature but the notifications do pile up. I set the notifications to archive to flash yesterday so I could capture some of the notification archives.

 

Thanks in advance!

tryhardcodemaster-diagnostics-20201030-1257.zip

Damn thats hot do you have the ability to put a better fan in front of those drives to keep them from getting that hot?

https://www.buildcomputers.net/hdd-temperature.html

 

Id be concerned unless those drives have some sort of manufacturing tolerances that are out of the norm. 

It's just a case with bad airflow and already has a high speed 120mm industrial fan on the only intake under the drives. I'm not too worried as it's backed up to the other servers that have drives running at 40-50C. These are enterprise 10K drives that should be kept under 55C but they run hot (with a standard 1500RPM 120mm fan they were at 80C). I'm just mainly wondering about the errant notifications when my threshold is set higher. (I also have a box of 10Ks that I can just replace them as they fail. I picked up a few cases totaling 100 for 4 euro per drive).

 

EDIT: Also, before anyone says just copying to another on-site is not backup, there are also 2 off-site copies of all pertinent data on two separate continents. I might be cheap picking up used drives, but I'm thorough with my setup.

Edited October 30, 20205 yr by tryhardcodemaster