Remote or internal (preferably both) system event logging services. Sending fault records via email is ok, but has its limitations. If alerts are emailed, chances are they are going to get lost, or filtered, by either the sender's service or the recepient's, or worse, delayed to the point where they are useless. Or the worst case, the SMTP API on the NAS breaks and either refuses to send mail, or floods the destination box with garbage.
If someone's NAS develops a fault, the admin would dump the logs, and send them to the team to deduct what happened and formulate a solution.
All you guys would do to start things rolling is to ask the customer if they have a log of the event that happened and send it to you.
Almost all servers, routers and major NAS arrays have one form of logging or another. Logs don't take up that much room, especially if stored in text format and preset for limited size like 512K to 1MB and then rollover. Even then there would be loads of data to evaluate any faults.