Follow
@feld
The result of doing so was a multi-hour outage, probably a few hours of work, and loss of something like an hour or two of comments. If any other component that was shared between both servers had the same issue (e.g. CPU, any part of the motherboard, ...) the result could have been very similar (modulo the data loss). Would you also suggest CPU/motherboard/NIC/... supplier diversity between primary and backup? If not, is there a reason why it's more important for storage (or maybe SSDs in particular)?
