Scott, I noticed your Drobo post come up on my Android as a "hot post" or whatever Google calls it. Honestly, I do not have the time to read the 70+ comments (and growing) from other people giving you advice to see if I would duplicate it or not, but I will say what I have to say and be done with it.
The main issue here that you and consumers of any tech products out there is that there is 1 thing you can count on: hardware will eventually fail. Does not matter if it is the small CPU in the NAS appliances you get from Drobo, the HDs you have installed in the NAS app, or one of the I/O ASICs on the motherboard. You simply have to first admit that this is the nature of electronics today and secondly plan around that detrimental inevitable scenario.
When enterprise-level companies have mission-critical business apps that they need to run, they run them on servers that have multiple redundancies (CPUs, DIMMs, NICs, FC HBAs, disks, PSUs, etc.). However, even with such servers that are engineered for redundancy, there will almost always be 1 or a few components that can be considered a "single point of failure" in the system (i.e., if that type of part fails, the whole server fails to function).
This is why software companies have created server clustering software. In the event that a whole server goes tits up, the backup cluster node (a secondary server) immediately steps in to serve the mission-critical business app (e.g., a huge Oracle database). With such midrange and high-end servers that are clustered together, that should enable the cluster to continue to serve whatever business app is running with 0 downtime to internal employees or external customers.
With the above in mind, I am not going to offer any specific solution but offer you a framework to implement multiple redundant solutions so that even if 1 piece of hardware fails, you can continue to do your business without interruption from lack of access to your photos.
One of the 1st and probably easiest suggestions I can make is simply plan on buying 2 NAS appliances (Drobo, Synology, or whichever manufacturer with whom you feel comfortable using) and simply make sure that you save copies of your photos to 2 or 3 mirrored NAS boxes. Yes, think of it as clustering your data in case your hardware does fail again in the future.
Now you may be balking at the thought of buying 2 NAS boxes (or more) just to mirror your data, but this is what enterprise-level businesses do with mission-critical apps and data. They cluster and mirror the hell out of the hardware and their business apps so that at least 1 system is hopefully always operational, feeding the needs of internal employees or external customers.
Yes, this will require a bigger financial investment on your part to mirror your data by buying and using more hardware, but ask yourself this question: how much is it truly worth to you to have multiple copies the same photos at any particular site in case you have a catastrophic hardware failure that prevents access to your precious photos? Every business asks this question of themselves, determining for themselves if they can afford to buy more hardware to ensure maximum uptime or can they afford to be down for a while waiting for replacement hardware to arrive, be configured, and data up and running.
In the end, I would recommend buying double the NAS boxes at any site, and also make use of some online/off-site storage (either using a data syncing service such as Dropbox, Google Drive, MS SkyDrive, etc. (maybe even 2 or 3 of these services for even more redundancy) or buy a web hosting account that has a ton of space (say 250 GB or more) so that you can use ftp/sftp to upload files to your web space / FTP space). There are tons of web hosting providers out there, and the one I currently use is 1and1.com
since it is cheap and offers both Windows and Linux hosting.
Also consider that you are using consumer-level (i.e., cheaper) NAS appliances by using Drobo's products. If you have the money, there are enterprise-class NAS appliances that can not only provide you more disk space, more speed, and more redundancy, but you could potentially have those NAS appliances scripted to auto-sync to your web hosting account's FTP space so that you can have data synced off-site on a regular basis.
To summarize, the moral of the story is make sure your hardware units/systems are mirrored, clustered, or otherwise made redundant in case 1 of your NAS appliances fails again. In the end, you may not have thought that you should have been expected to buy extra hardware, but now you know what businesses with mission-critical apps and data must do all the time. With that same mindset, set aside some extra money for yourself to invest in more redundancy.
In the end, yes, it was your fault for not doing this. Yes this is a learning experience for you and a primer of real-world issues that us IT guys have to deal with on a daily basis. However, now that you know what needs to be done, that should enable you to better protect the data that is precious to you.
Keep your $100. I have a feeling you will need to reinvest that in buying a little more hardware soon. :-) Have a good day.