Facebook Chief Executive Officer and founder Mark Zuckerberg in Dublin in 2019

This morning.

Via Mashable:

In an oddly short update on its engineering blog, Facebook explained the root cause of the outage that saw half the world suddenly post memes on Twitter, and the other half realize WhatsApp (another Facebook service that was down) isn’t a good backup for Messenger.

“Our engineering teams have learned that configuration changes on the backbone routers that coordinate network traffic between our data centers caused issues that interrupted this communication. This disruption to network traffic had a cascading effect on the way our data centers communicate, bringing our services to a halt,” the post said.

The outage also had an effect on the company’s internal systems, which is why it took so long to fix it, Facebook said.

Meanwhile..

The timing of the outage was somewhat suspicious, given that it happened right after a whistleblower went public with damning info on how Facebook handles misinformation on its platform. However, Facebook is adamant that the root cause of the outage is the misconfiguration issue detailed above.

Finally, Facebook said it has no evidence that user data was compromised as a result of this downtime.

Facebook apologizes for that massive outage, says no user data was compromised (Mashable)

RollingNews

Sponsored Link

21 thoughts on “Soz

      1. Noel Browne

        Care to expand sometime Mr Bodger because if reading & understanding correctly then a dystopian tsunami of technological global dictatorship(s) is/are rapidly evolving above & below us in plain sight, which will make the 2008 grand-theft financial crisis seem like a minor hindrance. Askin for a friend thanks..

  1. Cian

    I heard from my brother’s friend’s sister that the whole thing was triggered when Zuckerberg ordered a pizza!!

  2. ian-oG

    Not a chance in hell this was just a technical issue, a massive platform like this should have multiple redundancies ready to kick in in the event of a failover which clearly didn’t kick in.

    Zuckerberg is a pox on the planet and the sooner his nauseating platforms are consigned to the internet recycle bin the better. Whatsapp has literally one thing going for it – everyone uses it. I use it myself begrudgingly but never had a FB or Instagram account.

    Zuckerberg looks like the guy in zombie movies who gets bitten but doesn’t tell anyone – he’s probably also the one who will push a child in front of the horde to save himself.

    1. scottser

      i simply don’t trust anyone that rich who looks like he cuts his own hair and dresses from a charity shop.

    2. Ronan

      Resiliency doesn’t stop stupid. You’re protected against single mistakes, single failures of data centres, single failures of power circuits, single failures of power supplies on devices, single failures of core, distribution, access switches etc. You’re not protected against human error during a core change. If someone messes up your routes under eBGP it doesn’t matter how many different backbone providers and links you have because your routes are now being advertised incorrectly everywhere.

      That said, for the reasons I’ve outlined above it’s incomprehensible that anyone would make a backbone routing change during normal US working hours – so that’s what makes the story suspicious, not the credibility of the incompetent network engineer narrative.

      So this says to me that either:
      1. There was an attack that deliberately took down their backbone or
      2. There was an attack going on and they decided to isolate a circuit, but made a bonehead play and took too much offline or
      3. There was an attack going on that was serious enough to isolate their entire network
      4. Some bad AI-driven SIEM software detected an attack and started blocking devices everywhere
      5. Misconfiguration of SIEM software started locking things down due to a false-positive attack

      Someone, either a machine or a human, saw a justification to block traffic in my opinion. You just don’t make the kind of config change that can mess stuff up to this extent just before 12 eastern time. You do it at 0300 EST.

      1. SOQ

        But as its a global operation across all time zones, and that it runs 24/7- would it really be so EST centric?

        Also, 3 hours is quite a long time for that sort of business- which would suggest some sort of DR was invoked.

  3. Nigel

    Amazing that something which started as a way of rating women’s appearances to get revenge after a break-up turned out to be so toxic.

Comments are closed.

Broadsheet.ie