Mountain View, California — Millions of Google’s free e-mail service users were handicapped for about two-hour from accessing Gmail as it failed again on Tuesday afternoon, when users attempted to log on to find the service is no longer operational, seems to be the result of overloaded servers, a problem that Google acknowledged.
Gmail may be out of beta, but it is still not geared up for prime time Tuesday. (Credit: Google)
This Gmail outage is the fourth occurring this year, and last year Google apologized for a 15-hour outage. Gmail users from the world over announced problems on the Gmail forum, and remarks over the problem flooded Twitter.Problems developed this morning when Google switched off a small number of Gmail servers offline to perform routine upgrades. The Gmail was out of service from about 12:30 p.m. PDT Tuesday to about 2:30 p.m. PDT, disrupting millions of Gmail users who rely on the service for everything from fantasy football roster updates to business-critical information.
Recent modifications to the service that were supposedly designed to improve service availability, however, put stress on request routers, or servers that deliver Web queries to the necessary Gmail servers.
“At around 12:30 pm Pacific a few of the request routers became overloaded and in effect told the rest of the system ‘stop sending us traffic, we are too slow!’ This transferred the load onto the remaining request routers, causing a few more of them to also become overloaded, and within minutes nearly all of the request routers were overloaded,” wrote Ben Treynor, vice president of engineering and site reliability czar.
The Web version of Gmail subsequently was not responding, but IMAP/POP access was still up and running because they use different routers. Gmail engineers have since “brought a LOT of additional request routers online,” Google said, and Gmail is back.
According to research firm comScore, the Gmail service, launched in 2004, has nearly 146 million users worldwide — and is the third most-visited e-mail service in the U.S., and it suffered short outages twice in May. In February Google apologized for a significant outage, and there were other failures last year, including a 15-hour interruption that Google attributed to “a temporary outage in our contacts system.”
Users from Italy, Germany, France, Ireland, Taiwan and the United States reported problems on the Gmail forum. Google’s other Web services appeared to be working normally.
Google, however, rectified the problem by diverting traffic across the rest of its prodigious network, a luxury that it enjoys given the resources it has put in place to operate the world’s leading search engine.
Moreover, Google said it is making sure that “if there is a problem in one data-center, it should not affect servers in another data-center” and that “if many request routers are overloaded simultaneously, they all should just get slower instead of refusing to accept traffic and shifting their load.”
Google assured to “be hard at work over the next few weeks enforcing these and other Gmail reliability improvements.”
“We are aware how many people depend on Gmail for personal and professional communications, and we take it very seriously when there is a problem with the service,” Treynor wrote. “Thus, right up front, I would like to apologize to all of you — today’s outage was a Big Deal, and we are treating it as such.”