HA/DR solution for standalone Alteryx Server

[tcheng]

Our Alteryx Server is a virtual instance running from one data center. We do have another data center for DR. How do you set up your Alteryx Server for HA/DR purpose? 

[fmvizcaino]

Hi @tcheng ,

Here you can find more information about Disaster Recovery and Alteryx: https://knowledge.alteryx.com/index/s/article/Alteryx-Server-High-Availability-Environment

It is important to mention that this will require a specific license for HA/DR.

Best,

Fernando Vizcaino

[tcheng]

Thank you both. This is very helpful.

[Charliebitson]

Setting up Alteryx Server for High Availability and Disaster Recovery (DR) purposes involves creating a failover architecture between your primary and secondary data centers. You’ll want to implement load balancing and replication for the Alteryx Server to ensure that if one data center goes down, the other can take over seamlessly. This setup typically includes configuring multiple worker nodes and a controller in the primary data center and then setting up replication and a failover mechanism in the DR data center. It’s crucial to ensure that your Alteryx Server configurations are consistent across both environments and that your data sources are replicated as well. Using a cloud-based approach for storage or employing virtual machines with high availability features can further reduce downtime. Additionally, if you're ever unsure about the best configuration or need assistance in setting up complex systems, utilizing quality assignment help or consulting with experts could guide you through the process more efficiently. This approach ensures that you maintain business continuity without worrying about critical service interruptions.

[TheCoffeeDude]

One of these days this year, i'll be moving my multi-node environment to a "proper" HA/DR environment. In the meantime, here's my current set up.

• MongoDB Atlas hosts the backend database in a multi-region configuration.
• Set-up hot servers in a different region but the Alteryx services are turned off.
• Performed a DR to those hot servers. This sets up the controller token so it is the same as the primary servers.

As long as the controller token is not rotated, you don't have to perform a DR. If and when you rotate the token, you need to perform a DR to transfer the token to the hot servers.

I can talk more about this if you'd like.