If you can believe the current online chatter, we’re all destined to embrace the cloud, leaving all or most of our stuff online so we can get to it regardless of computing device or location. So you should be able to go about your digital life without worrying about where the files are located, and how you get them.
Well, at least that’s the theory, but it hasn’t quite worked out that way.
Although Apple has frequently gotten dinged for occasional failures with their various online services, including MobileMe and, lately, iCloud, trouble doesn’t just target that company. Google has occasional failures with Gmail, Amazon has had cloud-related issues, and, just this week there was a major failure at Microsoft.
According to a published report, Microsoft’s outlook.com email service suffered a 16-hour outage, resulting in the inability to use several services that included Hotmail and SkyDrive. The excuse? Some sort of firmware upgrade that went out of control and caused “a rapid and substantial temperature spike in the datacenter.”
Only one datacenter?
Microsoft blames the need for “human intervention” in fixing such problems for the delay in restoring services. The apology and the explanation are refreshing, however. Apple tends to admit the existence of problems only reluctantly, and with the briefest of details, if they admit anything at all.
Indeed, coming with the release of the OS X 10.8.3 update, some iCloud users encountered authentication errors for a while. Curiously, I had no difficulties checking my iCloud email on a Mac running the newly-installed 10.8.3, whereas that Samsung Galaxy S3 I’ve had under extensive evaluation couldn’t retrieve email from that account for a while. I am not, of course, going to suggest it was an act of revenge or part of Apple’s PR offensive against Android ahead of the release of the Galaxy S4.
On the other hand, I get occasional Google sync errors on the Samsung. Stuff happens.
The long and short of it is that it’s hard to trust these services when they fail from time to time. You almost think that these companies just haven’t nailed down all the problems just yet, and maybe we are all beta testers for using them. In saying that, most of these failures don’t actually result in lost data. Once the outage is resolved, things usually work normally once again. But not always.
Some years back, during the early stages of the MobileMe rollout, some email users lost email, at least briefly, during service outages. One of the worst failures occurred in 2009, afflicting Microsoft’s Sidekick service, which impacted T-Mobile users. It took days for the lost data to be at least mostly restored. But even BlackBerry, which was once the gold standard of smartphones, has suffered from occasional email service problems over the years.
But there’s more.
What about the place that hosts your Web site? When it comes to a cheap plan known as shared hosting, where a number of accounts are on a single server, what happens if the hardware fails? Yes, a server may be a more robust computing device than a Mac or a PC, but logic boards, power supplies and even hard drives go bad. If the drive fails, where’s the backup? Well, some Web hosts use redundant services, such as RAID drives. One is touting the ability to store your sites on two separate servers in different locations, so if one fails, you still have a fallback. Cloud-based services do not restrict your content to a single server, so you shouldn’t have to suffer the consequences of a single point of failure.
So what about a backup? Well, if you’re paying $5 a month for a shared service, backups are apt to be few and far between when they happen at all. Even a host that promises backups won’t guarantee them. The only way to be sure your data is safe is to keep your own copy of everything, not just your files, but the databases that come with such blogging systems as WordPress.
Our sites are run on a single dedicated server, but I’ve made sure there are regular nightly backups, not just to a drive in the same datacenter, but offsite. Just in case. All of the data on my Macs is backed up a pair of drives, one using Time Machine, the other using Carbon Copy Cloner. There’s also a cloud backup, so if something bad were to happen to my home, I’d still be able to retrieve my stuff and get back to work.
My son, Grayson, was able to restore his MacBook after a hard drive failure by retrieving his cloud-based backup. Safety comes in numbers.
So, yes, I realize that Apple, Google, Microsoft and Amazon are heavily pushing their cloud-based services and signing up hundreds of millions of customers. A number of independent companies have gotten into the game as well, and the cloud makes sense when you have a smartphone or tablet, where local storage may be limited and you aren’t always able to rush off to a desktop computer to back up your stuff.
These systems are not quite perfect, however. You can only hope that these huge tech companies are working hard to provide better reliability, and a way to ensure that you will never, ever lose your data. But I’m not one to take chances yet. For now, I plan to make extra copies of almost everything just in case, including this article.