Is Autoscaling not possible when DO enters a limited availability state?

October 12, 2019 91 views
DigitalOcean Accounts

I am evaluating Digital Ocean as a possible platform to deploy auto scaling web applications.

I am new to deploying this type of application and had always assumed I could consume additional droplets from a substantial pool where machines would always be available. I think I got this idea from the fact the AWS leases idle machines to users at a bargain that they reserve the right to take back instantly if a customer willing to pay standard pricing needs them. (At least that was my understanding.)

However, when familiarizing myself with the DO dashboard and services, I hit the following note on the “create droplet” page:

“Due to limited availability, we have temporarily disabled droplet creation in this region.”

Does this mean my original assumption was invalid with DO? Does my application need to automatically scale across data centers or suffer an inability to scale?

1 Answer

As far as I’ve seen there have always been available resources for a specific city, so for example if there are not enough resources in region New York 1, you could spin up more droplets in region New York 2.

  • But wouldn’t that mean an infrastructure that depended on droplets communicating with each other on private ip addresses would be invalid when the droplets are in different data centers?

Have another answer? Share your knowledge.