I can't tell you there is no benefit. Under the right circumstances I would imagine that you could just change the boot priority to boot from the other DOM and reboot the system, but for this to be an automatic thing, not very likely. This means that you could do this type of operation via IMPI and not open the system up to yank out the original boot DOM.
If you have been reading some of the threads over the past several months, there is a turn away from DOMs in favor of SSD, more like a single SSD as the boot device because they are inherently more reliable than USB Flash and DOM. Well I shouldn't say that about the DOM, I'm not savvy on your DOM hardware materials so if it were just a small SSD, a single one should be just fine, just they cost more than a normal cheap SSD.
My goal wasn't to tell you not to use the DOM's, but more to educate you on what limits they have. When this redundant boot device thing came out, I was very skeptical while folks were stating how wonderful it was. The testing was crap, meaning you start with a proper system, two boot devices, reboot and yank out the primary boot device. In someones mind that was all it takes to simulate a boot drive failure but that is far from true. In my world I generate very unique hardware faults for software developers and I'll place circuits on extenders, disconnect signals, add extra pulses, pull data lines low or high, or even just make a CPU go catatonic all in the name of producing a realistic failure. For the dual boot device thing here, if a device fails it will likely be corrupt code in which odds are the device just hangs the NAS while booting up. The odds of the device being completely dead and not recognized by the computer hardware is actually slim.
Sorry, I get a bit wordy at times. So I recommend a single boot device and after you make changes to the FreeNAS configuration, make a backup of it. It's just my two cents as they say.
