Yesterday, on what was probably one of the biggest online shopping days this season, a few web hosting services went down and with them an estimated 2 to 6 million websites and mail servers were inoperable for over 12 hours. This website was affected, as well as many of our clients’ sites.
The hosts affected were Bluehost, Hostgator, and Unifiedlayer. Dedicated servers which are supposed to have RAID redundancy didn’t, shared servers had web up but no mail, no one could login to Bluehost or Hostgator, and many were left with unanswered questions for a few hours. Many have suggested it was a DDOS attack, others have mentioned it was a hardware upgrade that crashed the entire system. I’d believe the latter as Bluehost mentioned in it’s tweets about a network looping problem, particularly on dedicated servers.
What do you do in this instance? Your site may go down for a few minutes periodically for maintenance, etc. It’s usually not a big deal. But when you can’t send or receive important emails, or receive orders, or even reach your own webpage for hours? Here’s what I learned yesterday. If your mail, web commerce or business site needs to be up, redundancy in your host as well as your data is the logical answer. That simply means finding a host that has mirrored servers in many locations, cloud based services, or hosting your own if you are a small operation. But hosting your own doesn’t guarantee that your server won’t go down at some point either, as they all do at some point.
While being very patient as I waited for tweets from Bluehost, I began researching into all of the areas mentioned above. So in my next article I will be writing about the who what wheres of alternative hosting when your primary webhost is down.
I need to rest and recover from a cold that has kicked me to the curb since Wednesday!