Social Enterprise

Using Twitter as a server notification tool: The pros and the cons

Twitter is breaking out of its "look at me" shackles and has real business value. Scott Lowe suggests using Twitter for server notifications.

I'm quickly begininng to understand that Twitter can be used for a whole lot more than just telling the world what I ate for breakfast. This summer, Rick Vanover (@rickvanover) and Jason Hiner (@jasonhiner) talked me (@scottdlowe) into giving it a shot as a way to help promote my articles. Tonight, as I was driving home, I had the thought that Twitter could be a good way to get server exception notifications to administrators -- and maybe even customers -- quickly and easily. (I'm not the first person to think of this, but it seemed interesting.) You probably already have a mechanism in place to notify people when servers have problems, but the more channels on which you can notify people, the more likely it is that at least one of the notices will get to the right person.

You might not want anyone and everyone to be able to start following your server notification Twitter account. Twitter allows you to create a new account that requires approval before people can follow it. Unless you plan to allow your customers and/or the general public to follow your server tweets, you should look into this possibility.

There are a number of pros and cons to consider if you decide to go this route for server notification.

Pros

  • All your friends are doing it. Ok, not really... that said, finding creative ways to make use of social networking tools does help to increase your understanding of those tools. Most companies are at least considering ways to benefit from the use of social media tools. Don't ignore the power of differentiation, either; if your company is doing something creative and different, use that as a point of distinction.
  • Twitter is available from anywhere. Let's say that you enter a no-service zone with your cell phone, your cell battery dies, or you simply left the cell at home -- no more notifications. You can always hit a library or other public place with Internet access, log in to Twitter, and see if anything bad has happened to your servers.
  • Your users and customers can sign up, too. Many companies have services on which users and customers depend. Through the appropriate use of Twitter, you can turn an appropriately worded server exception notification into an instant customer notification channel. As soon as you're notified about a problem, so are your customers. Once you resolve the issue, send out another tweet and let everyone know that the service is available again. Even if your entire infrastructure is on the fritz, you can send out a Twitter update from just about any mobile device, so it becomes a powerful communication tools. This can work against you if you have a lot of alerts, so be careful.
  • Twitter's API makes it possible to integrate. You may be able to integrate Twitter updates into an existing notification system through the use of Twitter's API.

Cons

  • Twitter is not known for uptime. Twitter used to be famous for its Fail Whale, but this was before my time using the service. The Fail Whale appears whenever Twitter is down and, apparently, that used to be pretty common. Since I've started using Twitter, I haven't seen the Fail Whale, but I will admit that I'm not exactly a heavy user.
  • Possible PR issues. If you don't mark your Twitter profile as private, anyone can follow you and receive updates. Since, in theory, your updates are automated, if your systems experience extended periods of trouble, and a lot of alerts are generated, everyone will see the alerts. Don't forget that allowing anyone to see error messages can sometimes be a security issue as well.
  • What if the network is down? If your connection to the Internet is not working, you won't be able to get to Twitter. I would not recommend using Twitter as a sole means for notification at this point. Instead, I'd look at Twitter sort of like how many of us look at security -- as simply one component of an overall notification strategy.

Will you be the first server admin on your block to tweet your troubles using Twitter? Share your thoughts on this approach to server notifications.

Additional TechRepublic resources about Twittter

Want to keep up with Scott Lowe's posts on TechRepublic?

About

Since 1994, Scott Lowe has been providing technology solutions to a variety of organizations. After spending 10 years in multiple CIO roles, Scott is now an independent consultant, blogger, author, owner of The 1610 Group, and a Senior IT Executive w...

6 comments
bfsmith
bfsmith

I simply HAD to thank you for information previously posted about the "transpose" selection in excel. I worked on the same spreadsheet at another person and it took her almost 5 hours to comnplete the project. It took me less than 3. I could have never done it without you.

CharlieSpencer
CharlieSpencer

McAfee VirusScan Enterprise 8.7, .DAT version 5728 of Sept. 2nd, has given me a warning every time I open the original article or the second comment. "Script executed by firefox.exe. JS/Exploit-Packed.c.gen. Trojan." McAfee is successfully blocking the script. I've notified Tammy.

Luis Cerezo
Luis Cerezo

I have been doing this for a while. it seems to be a good supplemental way to be alerted of problems. As you mentioned, service dependency is an issue... (or a whale of an issue)

maclovin
maclovin

Don't think I want to have my server failures, when they happen put up on someone else's server. Also relying on someone else's server.

tcavadias
tcavadias

Having it looked into :-) -Tammy [_]3

Editor's Picks