Several network links between two datacenters in West Europe became unavailable, due to a fiber cut caused by severe weather conditions in the Netherlands. This primarily impacted network connectivity between datacenters within the region, not traffic to and from the region itself.
The cloud engineers from Group K successfully transferred critical core services to alternate Azure Datacenters to overcome the network connectivity issues.
[Update]
We have repaired the impacted links, restoring networking capacity between datacenters. Many internal and external services saw signs of recovery from 15:30 UTC, but we continued to work on mitigating some latency issues.
[Next steps]
We continue to monitor the network infrastructure and capacity, and will continue to augment the region with additional capacity to de-risk the impact of any future issues like fiber cuts. We will publish a Preliminary Post Incident Review (PIR) within approximately 72 hours, to share more details on what happened and how we responded.