This is essentially a re-post (with a little editing and updating) from a six year old thread that has not had any activity in a couple years, so starting a new thread rather than rehash the old one.
I'm going to bring this up (again), because we are seriously into sequencing season, and every year at least one person shows up on the forum after having lost all or part of their sequences due to a hard drive crash, home burglary, fire, accidental deletion, computer virus or ransomeware, or some other reason. I'm hoping this reminder will keep at least one person from being in this awful situation. The summary is: Backup! Backup! and Off-site Backup!
I know that some of you have seen parts of this before, but take this as a reminder. I am a bit unusual in that I have a domain controller / file server and all computers log into the domain. Therefore, most of you can't do exactly what I do, but it should give ideas. On both the show computer and primary sequencing computer, there is a mapped network drive of L: that is really D:\Shared\LOR on the server. All the LOR files are stored on the server. My normal routine goes something like this. While I am sequencing, the file being worked on is saved with a new filename regularly with the date and a revision letter as part of the file name so it's easy to tell what is newer than what. For example a file might be:
Jingle Bells 2020-08-03a.loredit
I'll bet most of you can figure out what song it is, and what date that I edited it. The "a" after the date indicates that it was the "a" revision for that date. Every time I save the sequence, I save with a new filename and increment the revision letter. BTW, if I were to get to a z" revision, I figure I have spent too much time sequencing and it's time to quit for the day. I think I got to a "s" revision once. Usually when I am completed sequencing for an evening (and happy with it) or during the next session, I will delete the intermediate versions leaving only the last version of the night. The next step is to upload the last version of the night onto the cloud. In my case I use Google Drive.
The D drive on the server is currently two 1 TeraByte hard drives in a RAID-1 configuration. This protects against a single drive failure. There is also a third identical drive as a hot spare (the server has room for eight drives), and when I need more space, I will add two more drives to the RAID array. Note that in the event that the RAID controller detects a problem, it will immediately send me an E-Mail. The hot spare drive will protect against a second drive failure (provided it happens long enough after the first failure that the RAID controller has moved it into an active role and fully rebuilt the array). The data on the server is backed up weekly to an external hard drive and taken into my office at work. I have two external hard drives that swap places as the off-site backup. In other words, there is always a backup drive at work that is no more than 8 or 9 days old. And as noted above, important files are copied to the cloud right away. In the event of a problem that takes out the server completely (house burns down, major theft, Sax's RansonWare, etc), the data is sitting somewhere else - usually more than one other place. One note on cloud storage. Some cloud storage solutions will create a drive letter on your computer. That way you can save or copy files to the cloud just as easily as a local hard drive. This is very convenient, but does come with one major gotcha. Because it appears as just another hard drive on your computer, malware (such as a virus or RansomWare) can access it just as easily as a local hard drive. In other words, your cloud storage could get wiped out at the same time as your local hard drive. Keep that in mind.
You will note that I have made several references to having backed up data off site. I can not stress the importance of this. There are several things that are quite capable of removing or destroying your backup if it's sitting on the shelf right next to the computer. At least one person here on the forum had their home burglarized, and the bad guy took both the computer and the backup disks that were right next to it. A fire can easily destroy both as well. And yes, we had someone on the forum a few years ago that lost their house in a large brush fire that took out their backups that were in another building on their property.
Just for good measure, this computer (what I normally use for sequencing), the LAN switch in this room, and internet modems are protected by a UPS that has about an hour of battery life. The server, show computer, routers, and LAN switch in the data cabinet, are protected by redundant UPSes that currently are good for about 90 minutes (each UPS feeds a different power supply in the server), and when that project is completed, will have about 8 hours of battery (unless it's during the day when the solar will extend that). I also get sent E-Mails in the event of loss of primary power for any of the UPSes. During Christmas show season, all the active LOR files are also copied onto a backup folder on the show computer. In case of a failure of the server, I simply change the mapped network drive of L: on the show computer so that it points to C:\LOR_Local, and all my files are there. I did have to use that in 2014 when my ancient server was dying and kept re-booting. In the event that the show computer dies, I can move the three LOR networks back to the server and run the show from there - which is how I run the year round landscape lighting show.
In case you are wondering why I am so detailed in making sure stuff stays working, it comes from what I do for a living. I run a regional public safety two way radio system with the understanding that under the wrong set of circumstances, failure of the radio system could result in someone dying. We take system reliability VERY seriously at work! That mentality rubs off at home...