HA is an approach that minimizes downtime with redundancy and failover—essentially downtime prevention. DR complements HA by having systems and processes in place to recover quickly with minimal loss from system downtime—essentially downtime mitigation. High availability and disaster recovery plans should be addressed when developing the initial architecture, and it is recommended to leverage the existing platform features that provide these HA and DR capabilities. The following sections demonstrate functionality across the platforms that can be employed to deliver varying levels of high availability and resiliency.
inVia systems operate as an encapsulated component of a larger environment. While the entirety of the inVia software is contained within the proprietary Ubuntu 20.4 VM, it can be recovered by multiple methods in the event of a critical failure.
While all configurations within the VM are proprietary, the VM itself can be snapshot and redeployed using conventional methods detailed elsewhere in this document.
Contained within the inVia system is a Docker stack that maintains multiple database engines. Each database contains portions of the necessary configuration data that is required to operate the system. These database tables are backed up to an inVia cloud repository every 30 minutes through the inVia VPN contained within the Docker stack. In the event of a critical system failure that does not prevent access to the VPN, inVia technical support personal performs a remote connection and recovery operation to confirm that all necessary containers are functioning as well as populating the necessary databases with the most recent backup. After this operation is complete, all functionality is restored.
In the event of a critical systems failure that prevents access to the inVia VPN, on-site support personnel are required to contact the provided inVia support center and work to restore remote access. This can be facilitated by providing remote access to the hypervisor or bare-metal stack to the inVia support staff who can then perform a reinstallation of the proprietary Ubuntu 20.4 image that contains the necessary software stack.
Once this operation is completed and the inVia Docker stack is restored, a recovery of the configuration databases is initiated by inVia support staff. This returns functionality to within 30 minutes of the point of critical failure.