An entire building of the French society was destroyed by the flames. It happened on OVH’s Strasbourg campus.
What happened?
Around 2 a.m. on March 10, one of OVH’s four data center buildings
, SBG2 was engulfed in flames. The entire site was cordoned off and this implied a shutdown of data centers SGB1 to 4.
SGB1 and SGB4 are scheduled to reopen by March 15, while SGB3 will be restored by Friday, March 19.
The company, through a
press release
, is progressively updating its customers on the status of the incident.
The intervention of the fire brigade was unable to contain the damage, but it seems that there were no victims. The cause of the fire has not yet been established, meanwhile an investigation has been launched on behalf of the authorities.
The sudden interruption of activities has caused quite a few problems for the customers of the cloud service provider, which has more than one and a half million applications on its data centers.
Down of thousands of sites, inactive mailboxes, along with the numerous complaints from users on the company’s official social profiles, who ask for greater technical assistance to buffer the various problems. The company has shown itself ready to intervene with direct assistance to customers, but the situation still seems to be unmanageable.
What is the Disaster Recovery Plan?
Through his twitter profile, founder and CEO Octave Kabla urged his users to activate the Disaster Recovery Plan as soon as possible to deal with the sudden emergency.
The
Disaster Recovery Plan
It is a process related to preparing for the recovery and continuity of a business’s vital services after a natural event or human error. It consists of a set of phases including:
- Testing: After installing the DR solution, you need to test it. “Game day” is when you perform a failover to your DR environment.
- Monitoring and Alerting: You need to have regular audits and sufficient monitoring to alert you in case your DR environment has been affected by server failures, connectivity issues, and application issues.
- Backups: Once the DR environment is implemented, you should continue to perform regular backups. Periodic backup and recovery testing is essential as a fallback solution.
- User Access: You can secure access to resources in your DR environment using AWS Identity and Access Management (IAM).
- Automation: You can automate the deployment of applications to AWS-based servers and on-premises servers using configuration management software.
Data redundancy
Also of concern and amazement is the apparent absence of data redundancy, i.e. a design of the server architecture that replicates its content, ensuring the continuous provision of a service even if a system becomes inaccessible.
In fact, all the data that was hosted on the data centers that went up in flames had not been backed up beforehand.
Why is your data safe with AWS?
One of the strengths of AWS it’s its global cloud infrastructure. AWS has the largest, most dynamic, and most secure ecosystem with millions of active customers and tens of thousands of partners worldwide.
The AWS network is organized into Regions, which are defined as the physical location in the world where data centers are clustered. Each Region is made up of a series of Availability Zones that are isolated and physically separated within a geographic region.
Availability Zones enable customers to run applications and databases in production environments with high availability, fault tolerance, and scalability that would otherwise be impossible to achieve within a single data center.
All Availability Zones in an AWS Region are interconnected via a high-bandwidth, low-latency network, over a fully redundant dedicated metropolitan fiber that distributes high-throughput, low-latency networks between them.
All traffic between Availability Zones is encrypted. The network performance is sufficient to achieve synchronous replication across Availability Zones. Partitioning an application into different Availability Zones allows for the isolation of businesses and protects them from issues such as blackouts, lightning, tornadoes, earthquakes, and more.
Availability Zones are physically separated from each other by a significant distance of many kilometers, while remaining within 100 km of each other.
High Availability
Unlike other technology infrastructure providers, each AWS Region has different Availability Zones. Availability Zones are connected to each other with lightning-fast private fiber-optic networks, allowing customers to design applications that fail over to multiple zones without interruption.
The AWS Control Plane and AWS Management Console are deployed across AWS Regions and use a multi-AZ architecture within each Region to provide resiliency and ensure continuous availability.
This ensures that customers avoid critical service dependencies on a single data center. It can conduct maintenance activities without making a customer-critical service temporarily unavailable.
Improved continuity
In addition to replicating applications and data across different data centers within a single Region using Availability Zones, you can also achieve greater redundancy and fault tolerance by replicating data across multiple AWS Regions.
You can use both high-speed private networks and public internet connections, further improving business continuity and keeping low latency in check around the world.
What does VMEngine do to protect its customers?
In the aftermath of such a dramatic event, there are many unanswered questions about how much an event of this magnitude could have been predicted.
The fact is that the (unexpected) consequences have made people talk much more about the event itself, just think of the amount of users who will have to wait until next week to return to normal.
Could it have been avoided? Can you protect your infrastructure from accidents or human error? Is the Cloud always a certainty?
Below are some of the VMEngine Case Studies that have exploited the potential of one of the world’s leading cloud providers and, with the help of AWS Specialized Architects, have created ad hoc infrastructures while also adopting all the solutions to safeguard their data and architectures.