Twitter’s been having a rough month keeping the service stable, and the micro-blogging service is willing to admit it.
For the last few weeks, Twitter has been experiencing constant errors, fail whales and downtime, culminating in yesterday’s extended downtime and today’s high error rate.
In a short blog post, Twitter’s Sean Garrett didn’t mince words: “From a site stability and service outage perspective, it’s been Twitter’s worst month since last October.”
Twitter has been working on improving its system in anticipation of the World Cup, but according to Garrett, the company has uncovered deeper issues with its architecture — some so deep-rooted that they’ve caused unanticipated downtime. Several days ago, Twitter engineer Jean-Paul Cozzatti explained that the issues stem from several critical mistakes setting up and maintaining its internal network.
I’ve spoken to members of the Twitter team about the issue, and for them the downtime really boils down to the tidal wave of Twitter activity surrounding the World Cup. Having issues from growth that surpasses all of your expectations is a good problem to have, but perhaps Twitter should have expected this level of activity given its large international userbase.
To fix the issue, Twitter will likely take the service down during the next two weeks to make repairs and perform maintenance. The time frame — two weeks — makes us think that these issues aren’t simple fixes. Expect the high rate of errors and unexpected downtime to continue for a while longer.
No comments:
Post a Comment