Google Plus
Google SearchReuters/Francois Lenoir

Google is the go-to source for any web search - so much so that the term "google" is often referred to as doing an internet search. But what happens when the primary search engine goes bonkers. If you think that's impossible, well, think again.

On Monday afternoon, Google Search delivered unusual results that did not match the search term. While many thought it was some sort of an update, it was realised later that Google messed up, albeit briefly.

Google acknowledged the problem and patched the bugs soon after. But for the short while that Google Search went crazy, people couldn't help but call out the glitch.

"On Monday we detected an issue with our indexing systems that affected Google search results. Once the issue was identified, it was promptly fixed by our Site Reliability Engineers and by now it has been mitigated.

BELGIUM-EU-INTERNET-ANTITRUST-BUSINESS-GOOGLE
Google’s indexing system Caffeine suffers massive glitch: Here’s what happenedGEORGES GOBET/AFP/Getty Images

What happened?

Google has a web crawling and indexing system called Caffeine that processes data transfer. The system helps Google continually index the entire web in real-time. Users pointed out how bad results plagued web search. Recipe Blog SEO Casey Markee shared screenshots of examples, which showed irrelevant results for "baked beans."

Google's Gary Illyes explained the role of Caffeine and how it affects web search.

  • Ingests fetchlogs,
  • Renders and converts fetched data
  • Extracts links, meta and structured data
  • Extracts and computes some signals
  • Schedules new crawls
  • Builds the index that is pushed to serving

"If something goes wrong with most of the things that it's supposed to do, that will show downstream in some way. If scheduling goes awry, crawling may slow down. If rendering goes wrong, we may misunderstand the pages. If index building goes bad, ranking & serving may be affected," Illyes explained.

Explained

In order to fix the glitch, Google had to rebuild the whole serving index in a matter of hours.

Thankfully, the issue has been resolved and there's no chance it could occur again (hopefully).