Proxmox VE very high load average due to process "z_wr_iss"

[Kingneutron]

Buying two broken highend SSDs in a row is highly unlikely, plus I'm using the previous SSD in a different machine since a month without any issues.
smartctl never reported any errors for any of the SSDs, neither did the intensive UEFI checks.
The SSDs are both fine. This is not related to hardware.

Seems like a firmware compatibility issue between Proxmox/Debian and Samsung SSD 990 PRO reported by multiple Proxmox users in this thread.

Time to get away from the 990 "Pro" and invest in a different brand/model with good firmware and a high TBW rating.

Personally I can recommend the Lexar NM790, it's sitting at 1% wearout and has been running 24/7 since Feb

 

[dennis1024]

Samsung 990 PRO SSDs run smooth and reliably 24/7 on my eight Intel NUCs without any issues since 18 months now, so I'm sticking with them on that machines. There is no reason to replace them.

However, since it appears that, for whatever reason, Proxmox VE has issues with Samsung 990 PRO SSDs, I'll purchase alternative SSDs for my Proxmox VE only. So yeah, you're right, it's time to replace them.

Thank you. I'm glad you found a reliable SSD that works for you.
Unfortunately Lexar NM790 has (or at least had) similar issues resulting in system crashes on Proxmox VE as discussed here.

I purchased two Sabrent Rocket 4 yesterday. I consider them, like Samsung 990 PRO, highend within the consumer market, but not enterprise.
That being said, I'll give highend consumer SSDs this very last chance. If support screws up again, I'll purchase enterprise SSDs.
At the end of the day, reliability is all that matters to me.