How 1and1 failed me

1and1 fail

I manage this blog myself. Not just the content, but also the technology it runs on. The systems & servers are from a hosting company called 1and1.com. And recently I had some serious problems.

Join 31,000 others and follow Sean Hull on twitter @hullsean.

The publishing platform wordpress, as a few versions out of date. Because of that some vulnerabilities surfaced.

1. Malware from Odessa

While my eyes were on content, some russian hackers managed to scan my server & due to the older version of wordpress, found a way to install some malware onto the box. This would be invisible to most users, but was nevertheless dangerous. As a domain name with a fifteen year life, it has some credibility among the algorithms & search engines. There’s some trust there.

Google identified the malware, and emailed me about it. That was the first I was alerted in mid-August. That was a few days before I left for vacation, but given the severity of it, I jumped on the problem right away.

Also: Why I say Always be publishing

2. Heading off a lockout

I ordered up a new server from 1and1.com to rebuild. I then set to work moving over content, and completely reinstalled the latest version of wordpress.

Since it was within the old theme that the malware files had been hidden, I eliminated that whole directory & all files, and configured the blog with the newest wordpress theme.

Around that time I got some communication from 1and1. As it turns out they had been notified by google as well. Makes sense.

Given the shortage of time, and my imminent vacation, I quickly called 1and1. As always their support team was there & easy to reach. This felt reassuring. I explained the issue, how it occurred and all the details of how the server & publishing system had been rebuillt from the ground up.

This was August 24th timeframe. As I had received emails about a potential lockout, I was reassured by the support specialist that the problem had been resolved to their satisfaction.

Read: Do managers underestimate operational cost?

3. Vacation implosion

I happily left for vacation knowing that all my hard work had been well spent.

Meantime around August 25th, 1and1.com sent me further emails asking me for “additional details”. Apparently the “I’m going on vacation” note had not made it to their security division. Another day goes by and since they received no email from me the server was locked!

Being locked, means it is completely unreachable. Totally offline. No bueno! That’s certainly frustrating, but websites do go down. What happened next was worse.

Since I use Mailchimp to host my newsletter, I write that well in advance each month. Just like clockwork the emails go out to my 1100 subscribers on September 1st. Many of those are opened & hundreds click on the link. And there they are faced with a blank screen & browser. Nothing. Zilch! Offline!

Also: Why I use Airbnb chat even when texting is easier

4. The aftermath

As I return to connectivity, I begin sifting through my emails. I receive quite a few from friends in colleagues explaining that they couldn’t view my newsletter. I immediately remember my conversation with 1and1, their assurances that the server won’t be locked out, and that all is well. I’m thinking “I bet that server got locked out anyway”. Damn it, I’m angry.

Taking a deep breath, I call up 1and1 and get on the line with a support tech. Being careful not to show my frustration, I explain the situation again. I also explain how my server was down for two weeks and how it was offline during a key moment when my newsletter goes out.

The tech is able to reach out to the security department & explain things again. Without any additional changes to my server or technical configuration they are then able to unlock the server. Sad proof of a beurocratic mixup if there ever was one.

Also: Is Amazon too big to fail?

5. Reflections on complexity

For me this example illustrates the complexity in modern systems. As the internet gets more & more complex, some argue that we are building a sort of house of cards. So many moving parts, so many vendors, so many layers of software & so many pieces to patch & update.

As things get more complex, their are more cracks for the hackers to exploit. And patching those up becomes ever more daunting.

Related: Are we fast approaching cloud-mageddon?

Get more. Grab our exclusive monthly Scalable Startups. We share tips and special content. Our latest Why I don’t work with recruiters

  • niico100

    Er – I have little sympathy for you. You went on vacation at such a crucial moment – you had no backup – you couldn’t even check in once or twice a day to make sure it was working.

    This is primarily your fail – not 1&1s

    • http://www.iheavy.com/blog/ Sean Hull

      Thank you Niico. You’re absolutely right.