16 Comments

Summary:

It may have only lasted 100 minutes, but the Great Gmail Outage of last week generated discussion that endured for days. It started with panic attacks on Twitter and took some bizarre turns, such as sparking talk of lawsuits. But in the end, the fact of […]

gfailIt may have only lasted 100 minutes, but the Great Gmail Outage of last week generated discussion that endured for days. It started with panic attacks on Twitter and took some bizarre turns, such as sparking talk of lawsuits. But in the end, the fact of the outage wasn’t nearly as interesting as what it said about Google, about email and about us. Here are five lessons that Gfail had to offer:

  • 1. Get used to outages. Why? Scale forces history to repeat. As the Internet matures, we expect it to operate more smoothly, so outages make it look like you’re falling behind. But outages can also be a sign of that very maturation. Companies will learn to avoid them, then as the whole thing scales up and grows more complex, it will happen again. There will always be outages, inside the cloud and out.
  • 2. If you can’t stop a crisis, you can control how you respond. Gfail may have caused some companies to think twice about paying Google for its apps, which include Gmail. But when all was said and done, Google’s swift, forthright reaction to a PR nightmare won it praise. In a way, the company’s response outweighed the bad publicity of the outage itself.
  • 3. Big is bad for Google users. Gmail has become a crucial part of many workdays, but many people had no problem finding alternatives such as Twitter — to the point that Twitter itself felt the strain. It reminded me why I would hate for Google to buy Twitter or Facebook or any other communication channels. A supersized Google may help Google, but it doesn’t really help me. As Robert Cringely aptly put it, “[A]t the scale Google operates, even a hangnail can look like a fatal condition.”
  • 4. Email is finally a utility. Email downtime isn’t a life-or-death matter, as a power outage can be. Still, we have come to expect it to be there 99.9 percent of the time. This may seem obvious, but consider that most corporate email systems suffer an hour of unplanned outages a month (in addition to planned downtime).
  • 5. Chain reactions are just a part of networks. Google’s explanation of the Gmail outage made it sound a lot like the way a power outage happens: An overloaded transformer goes down, shifts its load to another prompting that one to go down, and so on. That kind of domino effect also played a big role in last fall’s financial meltdown. That’s how networks work: United you stand, united you fall.
    1. Isn’t hotmail or yahoo mail still a bit larger than gmails? We don’t seem to get the same hype when they are down, or do they just run a better network?

      Share
      1. I used to use yahoo mail, and I remember it being down once for almost a full 24 hours. And that was the Plus service. People need to chill out. From the uproar you’d think Gmail had been down for a week.

        Share
    2. [...] 5 cosas que aprendimos de la caída de GMail [en]gigaom.com/2009/09/05/5-things-we-learned-from-the-gmail-out… por repapaz hace pocos segundos [...]

      Share
    3. The biggest lesson learned from Gfail: keep an IMAP client running in parallel (and more or less up to date), because you’ll need it when the web interface bites the dust.

      This last one bit me hard when I wasn’t able to lead a conference call with the notes in my sent mail. My IMAP client hadn’t been refreshed in months, too.

      Share
      1. Or just enable POP access, even if you aren’t going to use it. POP didn’t stop working.

        Share
    4. [...] 5 Things We Learned From the Gmail Outage [...]

      Share
    5. [...] PDRTJS_settings_48057_post_192 = { "id" : "48057", "unique_id" : "wp-post-192", "title" : "5+Things+We+Learned+From+the+Gmail+Outage", "item_id" : "_post_192", "permalink" : "http%3A%2F%2Fregulapati.wordpress.com%2F2009%2F09%2F07%2F5-things-we-learned-from-the-gmail-outage%2F" } http://gigaom.com/2009/09/05/5-things-we-learned-from-the-gmail-outage/ [...]

      Share
    6. Don’t depend on a web client for email, because if you do, a GFail like the one described leaves you without access to any of your email, or calendar etc. Instead, use a rich client, do local backups that are robust against hard drive failure (much cheaper in the long run than doing it to the “cloud”), and access your rich client remotely.

      Share
    7. I believe it’s just that people need some thing to talk and discuss for long. People remember It’s a machine!!!!!!
      It will break down sometime, So for me the big part is how long it took to fix it back, and honestly it is fine as it didn’t took that long. A lesson for the Google in future though.

      Sonal Maheshwari
      USourceIT your single source for all IT needs

      Share
    8. Some reasons why it wasn’t so bad was because the actually mail protocol was not down just the interface to the mail server. I was still able to use imap just fine on my iPhone

      Share
    9. Let’s remember that email is a business tool – and so telling people that they need to maintain POP access or use IMAP on their mobile phone is kinda irrelevant – unless they are the technical staff entrusted with keeping this going in the first place.

      BUSINESS people understand that “glitches” like this happen (whether it be with their email, their electricity supply, their telephone or their favorite lunch spot) and deal with it. It doesn’t mean that they have to like it – but they can deal with it – especially when its handled as well as it was by Google.

      Share

    Comments have been disabled for this post