DataOn updates Back-ups and DR Policy
Like any SaaS provider or software vendor providing hosted solutions, DataOn does everything in its power to eliminate the possibility of unplanned system failures. In case such failure does happen, we follow detailed plans and procedures to make sure servers and customers are back online quickly.
In context with our ISO 27001:2013 certification efforts for Information Security Standard we have updated our security management procedures. Following, I want to share a summarized version of that new procedure:
DataOn attempts to configure the datacenter with substantial redundancy and multiple layers of redundancy in fault prone areas. This means that a single point of failure does not cause downtime and multiple points of failure have limited downtime and little or no data loss.
Data is stored on Storage Array Networks (SAN) with RAID 1+0 configuration providing additional multiple point redundancy and separation of data from less reliable database servers. Database back-ups and database log files are stored on separate equipment from databases allowing point in time recovery even in the event of a SAN failure.
In addition to redundancy of normal systems, DataOn maintains a full Warm Disaster Recovery Center (DRC). In the event of a loss of accessibility of the primary datacenter, the Disaster Recovery Center can be activated to provide ongoing operations based on the latest transferred back-up data.
Sample Faults and Solutions
- A broken hard drive – No impact as all systems use RAID storage
- Power outage – No impact as all systems have UPS and Generator support
- A completely failed server – No Impact as servers are clustered
- A single Internet connection is down – No impact
- All Internet providers are down – System unavailable until DRC is activated
For Indonesian Area Services the Warm Disaster Recovery Center (DRC) is located in South Jakarta and Cold DRC in Texas.
Disaster Recovery Process and Business Continuity
In compliance with DataOn’s ISO 27001 procedures a disaster recovery process is in place and tested periodically. In the event of a primary datacenter failure impacting some or all customers the event is notified automatically to an IT support team who will obtain authorization to activate the disaster recovery center within a recovery time objective of four hours. A first priority is placed on restoration of operation at the primary datacenter in order to minimize the impact for any data loss which may occur due to the existing recovery point objective range of 15 minutes to 8 hours.
Customers are informed during the occurrence of a disaster by email advising the estimated duration of the downtime and any additional information that may be required. Once availability and stability of the primary datacenter is restored, a minimal additional downtime will occur as data is transferred back to the primary datacenter. Provided that other operational concerns permit, the transition back to the primary data center will be performed during non-business hours. [tx_spacer size="16"] Contact us at firstname.lastname@example.org if you have any questions.
[tx_spacer size="16"] by Katrin Enns on Company Policies