Table of contents of the article:
Yesterday, June 2nd, around 20:50 PM, we were inundated with at least thirty calls complaining of service disruptions and server downtime. After a quick check, we confirmed that all machines were up and running, all services were reachable and usable, and that there were no routing issues.
Simply all these users had seen the number of visitors die on Google Analytics and from that they thought there was a downtime without even bothering to try to navigate the site.
Searching Italian Facebook groups like Fatti di SEO, Fatti di AdSense, and so on, we immediately noticed a lot of messages complaining about the exact same problem: Google Analytics had been down globally for several minutes now.
As the minutes passed, however, further complaints arrived from users complaining about problems at the Cloud Computing level , Cloud instances to be clear and gradually also other services such as Gsuite and other Google services that you can see in the image below.
Not just Google but their customers as well
If the problem is limited only and exclusively to Google services (by Google), one might even be patient considering that at most one will probably be without its services for a few hours but Google does much more with its Google Cloud.
Specifically, Google through Google Cloud computing allows you to allocate resources such as storage, IP, instances of VPS virtual machines and to be able to take advantage of the computing and storage power to offer services in the same way as Amazon AWS or Microsoft Azure could do.
Among their customers there are very important realities, big and very popular online brands that have gone irremediably down after this unexpected event last night which, however serious it may seem, is still rather singular and absolutely sporadic.
The causes of the downtime of Google and Google Cloud
The cause of Google's downtime is not yet known. Officially, Google reports in its Google Cloud Status Dashboard that there is "network congestion" in the eastern United States, a rather curious statement given that Google, as we know, is not the smallest of internet service providers, but it has truly remarkable bandwidth to be so "congested" on the fly.
However, online, and especially on Twitter, there are already those who hypothesize a hacker attack on Google as retaliation by China after Google banned Huawei from continuing to collaborate in the sale of Android, a rather popular news story in recent weeks.
Of course, read in this light it could only and exclusively be a conspiracy consideration without any practical feedback, but nevertheless it could also not be wrong given the reasons described above regarding this phantom network congestion. For now at 1:57 Italian we remain with the benefit of the doubt certain to be able to update this article tomorrow morning.
No one will ever be fired for choosing Google Cloud.
While such a downtime could cause hundreds of millions of dollars in damages per hour, it's clear that at the managerial level, choosing a huge player means what in technical jargon is called making the market's choice . That is, choosing the solution the market would have chosen, placing complete trust in it. Simply put, no company director can in any way be blamed for choosing a partner like Google, even if that downtime may have cost them considerable sums.
Otherwise, if a director had chosen a more modest player, he would have risked having to justify his choice and the consequences of his choices other than those of the market.
To be clear with a practical example, a 30-minute down by our company would be perceived at the top of a large company as more serious than a 4-hour down by a large company such as Google, Amazon or Microsoft.
Precisely for this reason, big companies continue to make the market's choices, even though it is increasingly clear that the Cloud is not the panacea for all ills . The networks of these large players are still susceptible to technical problems and 100% uptime does not exist and perhaps never will.
