What first appeared to be a minor Dropbox outage on Thursday turned into a major outage that kept users offline and unable to sync using their desktop apps or the website for more than 15 hours.
Not even half a month into the first month of the year, and now we've seen the first major public cloud outage of the year.
On Thursday, Jan. 10, Dropbox announced some users were experience issues connecting to the service and that it would take about an hour before service was resumed. More than 15 hours later, Dropbox was finally back online, but no explanation has yet to surface.
The problem first came to light on Thursday afternoon (Pacific time), when Dropbox acknowledged that syncing via the desktop client and uploading to the website were going to be affected for about an hour. Affected users must have been wondering if their clocks were off when, an hour later, they were still waiting for service to return.
It wasn't until after 7 a.m. PT on Friday that Dropbox tweeted it had officially fixed the problem and service had been restored. Still, users are still waiting on an official explanation as to what happened. In fact, @DropboxOps has been dead silent since it announced service had been restored.
During the downtime, there was much complaining, and users ended up creating a new hashtag specific to the occasion. Make sure to check out #DropboxDown for the bits of conversation that continued through to the end of the downtime period.
The last time Dropbox experienced an outage was when Amazon Web Services went down in late October 2012. The AWS crash took down not only Dropbox, but also Netflix, Foursquare, Pinterest, Reddit and Salesforce.com's Heroku platform.
Before the AWS cloud outage, cloud monitoring startup Boundary saw evidence of upcoming meltdown, but with the Dropbox outage last week, there was no such luck. Users suddenly found themselves without access to the popular cloud storage and syncing platform.
Some Dropbox users have taken this as evidence that the cloud still isn't reliable enough for business usage, but that's going a bit too far. Compared to internal downtime, the odd public cloud crash every so many months isn't all that bad. Still, cloud providers need to take care to keep their services up all the time if they expect wary business users to make the shift.