Users from around the world reported difficulties logging on to the site after problems started at about 17:00 BST.
The service was affected for around two hours while engineers sought to solve the issue.
In a blog post, Twitter dismissed reports it had been hacked, and offered “sincere apologies” to its users.
“A cascading bug is one with an effect that isn’t confined to a particular software element, but rather its effect “cascades” into other elements as well,” explained Mazen Rawashdeh, the site’s vice president of engineering.
“One of the characteristics of such a bug is that it can have a significant impact on all users, worldwide, which was the case today.
“As soon as we discovered it, we took corrective actions, which included rolling back to a previous stable version of Twitter.”
Mr Rawashdeh described the situation as “turbulence”.
“We began recovery at around 10:10am PDT [18:10 BST], dropped again around 10:40am PDT [18:40 BST], and then began full recovery at 11:08am PDT [19:08 BST].”
In its earlier days, Twitter was famed for its flaky reliability, with its “fail whale” cartoon – which was displayed when the site was overloaded – becoming a regular sight for users.
However, added investment into the service has resulted in better stability, Mr Rawashdeh wrote.
“For the past six months, we’ve enjoyed our highest marks for site reliability and stability ever: at least 99.96% and often 99.99%.
“In simpler terms, this means that in an average 24-hour period, twitter.com has been stable and available to everyone for roughly 23 hours, 59 minutes and 40-ish seconds.”