Twitter went down hard twice in a few hours. It's back to the old days, except this time, there was no Failwhale, just timeouts galore. In IT Blogwatch, bloggers get a life for a short time.
By Richi Jennings: Your humble blogwatcher curated these bloggy bits for your entertainment. Not to mention: The Gregory Brothers are BACK, baby...
Sharon Gaudin goes back to using email:
The microblogging site first crashed around 12:30 p.m. Eastern...[on] all platforms...third-party apps [and] the Twitter apps on the Android and iOS platforms. ... The site came back up around 1:15 p.m. ... .
...[A] hacker group claimed responsibility...but Twitter denied the claim.
Aunty Beeb adds:
The service was affected for around two hours while engineers sought to solve the issue. ... The site's last major downtime...was on New Year's Day when the site crashed after becoming overloaded.
And Julianne Pepitone peps it up a notch:
Cue the collective Internet freakout! ... Service returned intermittently around 1 p.m., but less than an hour later, Twitter crashed again.
...Downtime is a common problem for websites, though Twitter has been far better lately. ... [This] was extensive enough that Twitter didn't even display its famous "Fail Whale".
...[O]utage tracker downforeveryoneorjustme.com [is] coincidentally...the brainchild of a Twitter engineer.
Mazen Rawashdeh, Twitter's VP of engineering, offers "our most sincere apologies":
At approximately 9:00am PDT, we discovered that Twitter was inaccessible...there was a cascading bug in one of our infrastructure components...a bug with an effect that isn’t confined to a particular software element.
...As soon as we discovered it, we took corrective actions, which included rolling back to a previous stable version. ... We are currently conducting a comprehensive review to ensure that we can avoid this chain of events in the future.
...For the past six months...in an average 24-hour period, twitter.com has been stable...for roughly 23 hours, 59 minutes and 40-ish seconds.
Meanwhile, John Herrman ponders the meaning of "cascaded bug":
[If] you read between the lines and, like, realllly squint, this looks more like a low-level failure than a problem in Twitter's own codebase...[it] must be a fairly essential disturbance.
...[A] hacker group...is claiming the outage is a result of a DDoS attack [which] sounds very unlikely. [It's] not the kind of site you easily take down by loading it a bunch of times.
[yes, Joe Biden does drop the F-bomb, but it's bleeped, so have a SFW day]