Hi,
I just encountered an issue with my HPE 1820 24G PoE+ switch, J9983A, with PT.02.09 firmware.
While provisioning some Cisco 3702I APs I had set DHCP option 43 on my Windows 2019 DHCP Server to point those APs to their controller in another subnet. While that worked without issues, my switch decided to renew its DHCP leased IP. It locked up with all lights on. I rebooted it with everything plugged in, and it worked until it tried to obtain a new lease. It locked up again. So I removed its uplink to my network and it worked for an hour without issues. I replugged that uplink and it crashed within seconds. Reboot, crash, reboot, unplug, no crash. You may continue this as long as you want. It doesn't matter what's plugged in, as long as the uplink is unplugged, it will run for as long as you want. Plugging it back in brings the switch to a crash.
To verify my guess I removed that option 43 from my DHCP server and it works without issues since then.
Is it a known bug? I attached the last crash log below, support file is too big