Google Maps Platform services went missing for a few hours on Friday as various APIs fell over.
Around 0847 am PDT (1347 UTC), users of Google Maps Platform services began reporting problems. These surfaced on crowdsourced reporting sites like DownDetector.com and on the Maps Platform Status Page.
“We’re seeing reports of difficulties accessing some Google Maps and Google Maps Platform services,” a Google spokesperson told The Register via email. “Our team is investigating and working to resolve the issue as quickly as possible.”
The outage interfered with the function of third-party apps that integrate Google Maps Platform APIs. For example, FlightRadar24, which offers flight tracking apps and uses Google data to render geographic details, warned that users might experience issues. In this case, that meant maps without any details on them.
Yes, also issues on local installations (my Radarcape device) pic.twitter.com/N6rbpAzTVZ
— Volkskamera (@volkskamera) March 18, 2022
The IssueTracker for the various APIs indicates that the outages affected countries around the globe. Reports of trouble with websites and apps using Maps arrived from Brazil, Canada, Ecuador, Guatemala, Lithuania, Japan, the UK, and various US states.
These reports underscore the seriousness of losing access to geo data: “Google maps outage is affecting GPS Tracking and Address forwarding for EMS services in Texas currently,” says one trouble ticket.
Meanwhile, Manitoba, Canada’s Click Before You Dig service – which relies on Google Maps data to check whether planned excavation might disrupt buried utilities – limited responses to emergency phone inquiries only for about an hour due to the Maps outage.
Google Cloud appears not to have been involved in the snafu. Apart from a notice about issues with Horizontal Pod Autoscaler (HPA) and limited GKE validation problems, the GCP status page remained all green for the duration.
“Customers may experience performance issues with Horizontal Pod Autoscaler (HPA) cluster versions 1.21 and 1.20 resulting in delays for cluster scale up and scale down actions. The issue is only impacting clusters with more than 100 HPA objects,” reads one banner on the cloud status page.
Another warns: “Unexpected rapid scale down in Horizontal Pod Autoscaler (HPA) controlled workloads resulting in potential overload and subsequent outages. HPA versions 1.21 or above are affected by the issue. Customers can increase minReplicas or set scaleDown policy with large periodSeconds to increase stabilization periods to 15+ minutes as a workaround for the issue.”
Shortly after 1300 PDT, (1900 UTC), a Google spokesperson said the problem with the maps service has been resolved: “We can confirm that the technical issue from earlier today is now fixed. All impacted Google Maps and Google Maps Platform services are now back up and running globally.”
The Register asked Google if anyone would provide a technical explanation for what happened, but, alas, we’ve received no reply to that particular question.
It may be that Google doesn’t have all the answers yet. In the IssueTracker thread, an unidentified Googler says, “We will publish an analysis of this incident once we have completed our internal investigation.” ®