This was reported to be caused by severed cables in two different places. And I read some speculation/claims that this cut the primary and the backup systems simultaneously, which would explain why this lead to a complete outage in a system that should have quite some redundancy.

Of course we don't really have enough reliable information right now, but it looks a bit like this would require quite a bit of detailed internal knowledge to know which two cables you need to cut to shut this down entirely. Which would suggest more sophisticated actors might be responsible for this.