personally i'm just quietly stunned that desktop and laptop CPUs pretty much universally feature failure modes that maintain slow but sustained operation if they start to overheat, even if because of a clogged or completely failed fan, but server CPUs seem to not be able to deal with the same... and that a professional data centre doesn't have the machines set up to alert such failures with beeping and flashing of LEDs, and someone onsite who can look for and respond to such...

just when I thought the world was creeping back towards some kind of sense over the last week or so...