We want to establish disaster recovery for Qlik replicate. The plan we have is to unmount and mount the EBS volumes from unhealthy EC2 server to a healthy server. (We are using VMs).
Is this plan feasible? If yes, then what shall be the process for it? If not, then what are the alternatives?
The scope of the master user key is bound to the original machine where the product is installed. There is always the possibility to do a hard copy of the data directory and reset the repository but this implies that you will need to manually enter the product license and endpoint passwords, which is definitely not recommended for DR scenarios.
The best solution for this purpose would be to work with a cluster instead. This is documented here:
In a cluster installation, the concept is to have 1 cluster composed of 2 or more nodes running the same service initiated by the cluster manager, and the task repository files being isolated on a block device (physical, virtual or iSCSI-based).
In this scenario, let's say you have 2 VMs in AWS, you need to mount a cluster, then install the software on each node. During the installation, you need to point the data directory to your block device, EBS or any preferred/suggested by the vendor that is compatible with our product.
"Run the Qlik Replicate setup wizard on the first node that is part of your cluster. Install Qlik Replicate to a local folder. The default is: C:\Program Files\Attunity\Replicate." The first default folder for the product binaries could be the same drive holding the operating system for instance.
"In the setup wizard, specify a shared storage location for the "data" folder; for example, F:\Replicate\data." Then it points you to the documentation "Setting up Qlik Replicate on Windows."
The scope of the master user key is bound to the original machine where the product is installed. There is always the possibility to do a hard copy of the data directory and reset the repository but this implies that you will need to manually enter the product license and endpoint passwords, which is definitely not recommended for DR scenarios.
The best solution for this purpose would be to work with a cluster instead. This is documented here:
In a cluster installation, the concept is to have 1 cluster composed of 2 or more nodes running the same service initiated by the cluster manager, and the task repository files being isolated on a block device (physical, virtual or iSCSI-based).
In this scenario, let's say you have 2 VMs in AWS, you need to mount a cluster, then install the software on each node. During the installation, you need to point the data directory to your block device, EBS or any preferred/suggested by the vendor that is compatible with our product.
"Run the Qlik Replicate setup wizard on the first node that is part of your cluster. Install Qlik Replicate to a local folder. The default is: C:\Program Files\Attunity\Replicate." The first default folder for the product binaries could be the same drive holding the operating system for instance.
"In the setup wizard, specify a shared storage location for the "data" folder; for example, F:\Replicate\data." Then it points you to the documentation "Setting up Qlik Replicate on Windows."
