Me, training an IT staff prior to migrating their Exchange migration: “Remember that in the server policy view, don’t press the red X button next to a server because, counterintuitively, it deletes the server entirely from the system rather than just removing it from that policy!”

Me, on the first day after the migration: (presses the red X and deletes the production server entirely from the system)

All mail flow ceased and began queuing up in the file system. But, I designed the server disk layouts to be ready for an extended period of queuing by sizing and isolating the transport queues from the os, xaction logs, and mailbox databases.

I did scramble to get the server object restored but as soon as it was, the mail queues began draining and I mean fast! Those brand new disk arrays proved their IOPS that day, I can assure you.

In the end it just delayed mail delivery a bit but nobody’s mailbox went down, so it was just a little “day 1 hiccup” of the migration that we laughed about