Twitter was unavailable for more than one hour on Thursday evening UK time. While hackers claim to have caused the trouble, the micro-bloggin services blames a technical fault.

The failure was the biggest outage on Twitter for some years, clocked at two hours by statistics site Royal Pingdom, and was quickly claimed by hacker group UGNazi. However, a blog post on Twitter’s site specifically denied the outage was caused by hackers.

Twitter-uption

“This wasn’t due to a hack,” said Twitter’s own blog, confiming that from 9am on the US West Coast, “Twitter was inaccessible for all web users, and mobile clients were not showing new Tweets”.

Twitter put the problem down to “a cascading bug in one of our infrastructure components,” explaining that a problem in one software element had knock-on effects, cascading into other elements. The problem was fixed by rolling the service back to a previous stable version.

The firm denied the problem was due to its move to new offices or to an overload caused by Euro 2012 (previous sporting events have sometimes caused trouble). Twitter’s blog also denied the problem was caused by annoying animated GIF avatars and images. These are prohibited on Twitter, but several sites have published ways around the ban.

UGNazi, a group affili claimed to have caused the failure with a denial of service attack, apparently as a protest against Twitter’s support for the US government’s CISPA cyber security bill.

During the outages, TechWeekEurope‘s US partner eWEEK was contacted via email by a person using the alias “Cosmo lol,” who claimed to be a member of the UGNazi hacktivist group.

“Cosmo lol” took credit for taking down Twitter and, when asked why, emailed: “Just to show what we really are capable of.”

When asked for proof of the deed, “Cosmo lol” emailed: “I can follow you on @CosmoTheGod and @UG for proof.” There were lively discussions on those two Twitter threads immediately after the site came back into action.

After Twitter emailed eWEEK about the “cascaded bug” problem, “Cosmo lol” emailed eWEEK with this response: “They moved to multiple servers in the past couple hours trying to migrate our attack, there (sic) is no bug in there (sic) system.'”

Twitter claims that its normal availability is at least 99.96 percent which works out at about half a minute a day of outage. Royal Pingdom confirms these figures, with the site scoring 100 percent availability in April.

It is Alan Turing’s centenary: Try our Turing Test!

Peter Judge

Peter Judge has been involved with tech B2B publishing in the UK for many years, working at Ziff-Davis, ZDNet, IDG and Reed. His main interests are networking security, mobility and cloud

Recent Posts

Tesla Fires Software, Service, Engineering Staff

Tesla lays off software, service, engineering staff after disbanding Supercharger team, as major cull continues

9 hours ago

Grayscale Bitcoin Shares Surge On First Inflow Since January

Dominant Bitcoin ETF Grayscale Bitcoin Trust shows first net inflow since January as investors flock…

10 hours ago

US Crypto Campaign Funding Groups Raise $102m

US campaign funding groups backed by cryptocurrency sector raise more than $102m as firms seek…

10 hours ago

Robinhood Served With SEC Crypto Enforcement Notice

Robinhood Markets says it received SEC enforcement notice over cryptocurrency trading platform amidst ongoing crackdown

11 hours ago

Synopsys Spins Off App Security Unit In $2.1bn Deal

Chip designer Synopsys to sell software integrity unit to private investors to create new independent…

11 hours ago