Unraid writes 500G to my SSD for no reason midnight

[JeffSuperb]

Hello, 

I am new to unraid, I just set up my unraid server last week, and everything goes smooth except one thing...

 

Unraid writes 500G to my SSD for no reason midnight. It already happened two times in the last three days. I notice this because unraid server sends me email tell me my SSD is hot and almost full. (But my SSD is almost empty). When I wake up in the morning, the SSD is almost empty. But I am sure it did write 500G data to my SSD, because I did buy new SSD for my unraid build, and my SSD has 700GB write from smart data after it happened the first time and 1.18TB write after it happened 2nd times. 

 

I am very worried about it because it might eat up my SSD fast, but since it happened in late night, I cannot debug while it happens. I attache the diagnostics log below, please let me know what should I do. Thank you so much!!!

 

 

 

 

 

 

 

 

Email I received:

I receive those email 7/16 morning (1st time write 500G to my SSD)
[7/16/2010 5:35 AM] Unraid Status: Warning [JEFF-UNRAID] - Cache disk is hot (49 C)
[7/16/2010 5:40 AM] Unraid Status: Alert [JEFF-UNRAID] - Cache disk is low on space (96%)
[7/16/2010 5:41 AM]  Unraid Status: Notice [JEFF-UNRAID] - Cache disk returned to normal utilization level

 

I receive those email 7/18 morning (2nd time write 500G to my SSD)
[7/18/2010 4:11 AM] Unraid Status: Warning [JEFF-UNRAID] - Cache disk is hot (49 C)
[7/18/2010 4:11 AM] Unraid Status: Warning [JEFF-UNRAID] - Cache disk is high on usage (72%)
[7/18/2010 4:20 AM] Unraid Status: Alert [JEFF-UNRAID] - Cache disk is low on space (91%)
[7/18/2010 4:21 AM] Unraid Status: Alert [JEFF-UNRAID] - Cache disk is low on space (92%)
[7/18/2010 4:22 AM] Unraid Status: Alert [JEFF-UNRAID] - Cache disk is low on space (98%)
[7/18/2010 423 AM] Unraid Status: Notice [JEFF-UNRAID] - Cache disk returned to normal utilization level
[7/18/2010 4:41 AM] Unraid Status: Notice [JEFF-UNRAID] - Cache disk returned to normal temperature

 

Screen shot and Diagnostics logs zip:

 

 

 

 

 

Edited July 21, 2020 by JeffSuperb

[JeffSuperb]

OK. I found the reason. I create a Vdisk for windows VM for 500G and manually put in shares domains on HDD. But domain share set to 
"Prefer : Cache," so it is trying to move to SSD last midnight. But my SSD not big enough for the Vdisk, so mover failed. But it still writes about 500G data before it fails.

That can be a bug, and I believe unraid mover should compare the size on SSD left with the size it will transfer to SSD. If SSD not big enough it should throw a warning not try to copy 500G of the file to SSD

[BRiT]

Look at minfree settings for that share and for cache. Also is that VM image sparse, if so its size may report differently so it thinks it has room, but when copied may no longer be sparse.