Both parity drives immediately fail when downloading a large amount of data on 99% full array

I'm trying to pinpoint this issue. Both of my parity drives end up in error state (red x/disabled) after attempting to write a large amount (~50GB) of data to a share. It's happened first about a week ago, but the drives came back as fine after some testing and recreated the parity with no issues, and the large file download was hash checked/resumed with no errors. Small files (~5GB) downloaded just fine in between now and then, until another large download killed them both again today. The only thing I can think, is that there's an issue due to how much free space I have (which is to say none, as most drives are at 99% capacity), because it happened to pop at the exact moment I started to write that large amount each time

Here's my setup

Parity: 2x10TB

Disks: 20x8TB, 3x4TB, 1x10TB (Slowly switching out to 8TB, and now 10TB as drives die)

Cache: 2x1TB SSD (RAID1)

Hardware:

Case: Supermicro SC846E16-R1200B

Motherboard: X8DTE-F

CPU: 2xIntel Xeon L5640

Memory: 48GB

Controllers: 2xSuperMicro AOC-S2308L-L8i (LSI SAS3008-based, one card dedicated to dual channel to the 24-bay chassis, one card dedicated to the 2 parity drives and 2 cache drives)

I know I need to free up some space, or rather get some more, but I'd like to rule out any hardware or software setup issues first. Like I said, the failed drives tested and rebuilt fine, the controller card and cabling seems to test fine as well, and there's been no issues with the cache which live on the exact same controller. I'm hoping it's just a write error due to how low I am on space, and can just put a mental block to not go below a certain threshold until I can add some more. I'm not sure if it's relevant, but because of how low on space I am, I was writing to disk shares rather than to a global share, because I knew said disk had enough storage to store the file (I was trying to make sure the disk had twice the available space than the file size, eg 100GB free for a 50GB file)

Logs attached. My system has been pretty rock solid (minus a brief issue where I was running some mismatched firmware a while back) so I'm not too well versed into what I should be poking around the logs looking for

void-diagnostics-20201124-1539-anon.zip