Two cloud services, Google Cloud and Oracle Cloud, stop at about the same time due to record heat wave



Since the third week of July 2022, Europe has been hit by unprecedented levels of severe heat waves,

killing more than 1,700 people in Spain and Portugal. This heat wave is affecting not only people but also digital devices, and it has evolved into a situation where cloud services such as Google Cloud and Oracle Cloud are stopped.

Google Cloud Service Health
https://status.cloud.google.com/incidents/fmEL9i2fArADKawkZAa2



Multiple Services | UK South (London) | CN-764094 --OCIC Status
https://ocistatus.oraclecloud.com/#/incidents/ocid1.oraclecloudincident.oc1.phx.amaaaaaavwew44aa7zoskanlspjh4ll6wxhwxrbkbed4d4cnupxexzqzvlyq



UK heat wave causes Google and Oracle cloud outages
https://www.bleepingcomputer.com/news/security/uk-heat-wave-causes-google-and-oracle-cloud-outages/

Google, Oracle cloud servers suffer outage in UK heatwave • The Register
https://www.theregister.com/2022/07/19/google_oracle_cloud/

The heat wave that hit Europe has hit record-breaking heat over the past week, recording the first ever temperature of 40.2 degrees Celsius on July 19, 2022 local time. The heat wave caused the data center cooling system to host Google and Oracle's cloud infrastructure to fail, causing Google Cloud and Oracle Cloud to shut down.

Specifically, Google and Oracle have decided to shut down the data center to prevent long-term service outages due to permanent damage to hardware components, instead causing cloud services to go down. It is a flow that I decided to do.

Oracle was the first to shut down its cloud service, which reported a cooling system failure in a data center around 11:30 local time on the 19th, explaining that it turned off 'non-essential hardware.' doing. 'The unseasonable temperatures in this region (UK) have caused problems with a subset of the cooling infrastructure in our data centers in South London, which has led to service infrastructure to prevent uncontrolled hardware failures. I had to power off a subset of the structure, 'explains why the cloud service went down. Regarding this response, Oracle also states that 'this measure was taken to limit the potential long-term impact on our customers.'



Approximately two hours after Oracle reported, the cooling system failed in the building hosting Google's two European West servers, and the cloud service also went down. Regarding the failure, Google said, 'One of the buildings hosting Zone europe-west2-a in Region europe-west2 had a cooling-related failure. This caused a partial failure in the capacity of the target zone and was virtual. The machine (VM) has shut down and some customers' machines have been lost. '

In addition, Google said, 'We are working hard to bring the cooling system back online and create capacity in that zone. No further impact in zone europe-west2-a is expected and is currently being implemented. The VMs inside should be unaffected. A small portion of the replicated persistent disk devices are running in single redundant mode. '

In addition, Google also announced that it turned off some machines in the zone and restricted the preemptive startup of Google Compute Engine (GCE) in order to prevent machine damage and prolonged outage.

in Web Service, Posted by logu_ii