Christmas meltdowns – they happen to the best of us. And without the right provision and preparation, they can hit your website too. It often comes down to 3 things – your server’s ability to handle large amounts of traffic, system security and, when CPU usage hits the fan, how well-equipped you are to recover and keep the cogs turning.
We’re not scaring you into caring (Halloween was so last season), but it’s important to highlight that even the big players aren't safe when it comes to this stuff. And the consequences can be disastrous. All the more reason to make sure your systems can keep cool when the heat is on this Christmas.
So, we’ve put together 5 chilling moments to learn from, when huge server failures wreaked havoc for their owners and made ripples across the globe.
When AWS went AWOL
‘Twas the night before Christmas when disaster struck at Amazon Web Services (AWS), with grave consequences for Netflix. According to AWS, a combination of human error and flawed access controls caused a developer to accidentally delete a key data set in the AWS Elastic Load Balancing Service, meaning a major server outage – whoops…
As a result, Netflix went down for hours on Christmas Eve, leaving millions of users without access to arguably the most popular night in the film calendar. Meanwhile, Amazon Prime coasted along nicely like nothing happened (awkward).
Lesson – keep your eyes on the ball and prepare for busy periods around holidays.
What happened to Whatsapp?
Back in 2017, WhatsApp suffered a worldwide outage lasting for hours. Sure, WhatsApp has experienced a number of outages since then, but this one still stands tall in their back catalogue for its bizarre absence of reasoning.
The messaging service acknowledged the outage in a vague media statement after the event, offering no details on the cause. During the outage, users were left in the dark. As angry as they were perplexed by the lack of explanation, leaving one user tweeting “WhatsAppening”. Top marks to them for creativity, and no marks to WhatsApp for their lack of communication.
Lesson – it’s a good idea to keep your audience informed during outages to avoid further reputational damage.
Stuck on the runway at British Airways
After an outage that physically damaged servers at its data centre, British Airways experienced an “uncontrolled return of power” that led to a wider IT shutdown. Naturally, this led to – you guessed it – flight delays and cancellations for a whopping 75,000 passengers on a busy Saturday morning.
As far as outages go, this one was particularly expensive, with over £100m in compensation costs and a 4% share drop on the first day of trading after the event for their parent company, IAG.
Lesson – if your operations can't function without a working IT system, be prepared with a just-in-case compensation process for your customers.
2020 – the worst year for Google
As the lockdowns of 2020 sought to keep us indoors, one of the largest outages ever actually stopped one person from accessing their own house! The culprit was the mighty Google Services, whose major global outage caused the shutdown of free services (like Gmail, YouTube, Google Docs, and Google Photos) and paid services (like YouTube Premium and Google Suite).
Most users received generic error messages when trying to access a Google service, while one user was left out in the cold (literally!) after the outage caused Nest smart door locks to fail.
Lesson – test (test and test again!) your disaster recovery programmes, especially if outages could affect multiple vital sites.
Facebook faced the music
At precisely 4:15 PM GMT on 13th March 2019, the largest ever recorded outage at the time was just warming up. Facebook, Instagram, and WhatsApp were the services to take a dive, with approximately 7.5 million problems reported from end users throughout the incident. This lasted a whopping 12 hours. That’s a lot of disgruntled scrollers and chatters!
And the global outage surpassed the previous record by almost 200%, held by YouTube in 2018, which generated 2.7 million problem reports.
Lesson – plan for the worst and hope for the best so you can limit downtime when disaster strikes.
Secure your servers with Fasthosts
If you rely on servers to power your business, you can rely on us to host them, with a wide range of solutions to suit your needs. From Dedicated and Cloud to VPS, we can take care of everything for you (and plan for disaster!). So you can focus on wrapping your Christmas presents, safely in the knowledge that your servers are in good hands.
Get in touch to find out more about how Fasthosts can help, as well as our latest Dedicated Server deals, with up to 40% off!
