The fun started this morning at around 4:00 AM when one of the big data servers died. The info that’s been relayed is one or more of the disks in the drive array failed. Replacing the drive and rebuilding is expected to take all day - 350 GB of data is going to take quite a while to be read back from tapes.
(What ever happened to RAID and hot-swap drives?) :dubious:
It gets better.
I just got an email from the network people that they’ve had to evacuate their building because of a fire in the server room. :eek: The strange thing is the dead server is in a different building.
Days like this make me glad I support a different facet of our users’ computational needs. Of course, about a third of the tools and databases I use daily live on that dead server, so a lot of my work has been shoved off to tomorrow. At least I can tell callers now that I can’t do whatever because there was a fire in the server room so the system is unavailable.