A bug had occurred in Google’s Gmail service that sent thousands of unwanted emails to a user. Auto-fill of email addresses was allowed by the bug in Gmail when it is accessed from the result page of Google Search and had an impact on the Hotmail account of David. S. Peck. It was discovered that another person’s email address was filled in by Google in a new Gmail message whenever someone signed into their Google account and chose the sub-link on the top of the Google Search results page in relation to Gmail. Unfortunately, it was Peck who was the receiver of the unsolicited emails because the bug had auto-filled his Hotmail address.
A technical glitch was cited by a Google spokesperson on the email addresses that have already been revealed on public web pages and seem to be too prominent in the Search results. It was also stated that the company had fixed the particular bug. This wasn’t the only problem that Google had to contend with. Its services also went down early on Friday for users all over the world including India. Nonetheless, not only was the search engine giant quick to resolve the issue, but also has made an apology through its Google+ Account.
Not only was an apology issued by Google, but the company also listed reasons that were responsible for the outage. The affected services included Calendar, Google+, Gmail and Documents. The problem persisted for about 25 minutes for 10% of the users while others had to deal with the outage for at least half an hour. Basically, configuration i.e. information telling other systems how to behave, are generated by an internal system, but because of the existence of a software bug, an incorrect configuration was generated. In the next 15 minutes, the incorrect configuration was sent out to the live services.
Because of this configuration, the requests of data made by the users went ignored and the services themselves generated errors. These errors were then seen by the users on the affected services and at this point the Site Reliability Team was alerted by the company’s internal monitoring system. 12 minutes later, the engineers were still working on debugging when the error was automatically cleared by the same system and then generated a new and correct configuration. This configuration was then sent out by the system and helped in subsiding the errors. Services were then resumed and there were no more problems faced by the users.
After the services had been resumed, the search engine giant said that steps would be taken for avoiding the issue in the future. The company stated in its post that additional monitors and checks were being put in place for ensuring that this kind of problem didn’t occur again and cause an inconvenience for the users. This temporary error faced by the system had occurred at the same time when the reliability team of the internet giant was sitting down for answering questions on the Reddit’s AMA session.