More than enough has been written about the Spectre & Meltdown CPU vulnerabilities so I won’t go over them directly here. If you’d like more info, I’ve included some links at the end.
My interest is more with the trouble caused by the intended fixes. Despite CPU manufacturers testing the fixes for some months, in certain circumstances they are causing random reboots, and other performance problems.
To “Safely Remove” a disk is standard practice to avoid filesystem corruption. In simple terms, the computer stops writing to the disk and then you’re safe to unplug it. In the case of a random reboot, the computer doesn’t get the chance to finish with the disk. A crash at a critical moment can damage the filesystem and render the machine un-bootable. Even if the system can recover itself, if you crash enough times your chances of trouble increase. It’s hard to trust a system that could randomly reboot at any moment.
Until we get fixes from the manufacturers, there’s not much else we can do. Now would be a really good time to make sure your backups are working and up-to-date. These fixes could take a while to stabilise, and in the meantime you could be one random reboot away from data loss.