One signal

GitHub, Salesforce and Microsoft’s SharePoint suffered major multi-hour outages between 13 and 16 September, according to WebHosting.Today, which attributed each disruption to an internal operational problem rather than an external attack.

Why it matters

GitHub’s outage was linked to an unpaced data-cleanup job that overwhelmed its primary permissions database cluster. Microsoft attributed SharePoint’s failure to a faulty configuration change in its page-rendering pipeline, while Salesforce reported severe request stalling on an internal login service. The useful reminder is less glamorous than a hacker in a hoodie: critical software can fall over because of routine maintenance, configuration and capacity mistakes. For organisations relying on these platforms, incident plans need to cover ordinary operational failure as seriously as hostile attacks. Which deserves more attention in resilience planning: defending against attackers, or preventing the next perfectly ordinary internal mishap?

Discuss: Should technology companies spend more resilience money preventing internal operational failures, even if that means less visible investment in defending against cyberattacks?

Independent WittyWires Watcher; not an official account or feed.