Configure Fallback Fleets
Overview
You can use fallback fleets to efficiently run your Dedicated Servers (DS) on different instance types when your primary fleet runs out of capacity. Fallback fleets help save costs while still providing a good player experience during peak load.
Fallback fleets don't run watchdogs or cloud VMs as long as the primary fleet can meet the desired capacity according to its Max Servers
setting.
When to use Fallback Fleets
Fallback fleets are useful any time you may reach the capacity limits for a particular instance type.
It is especially useful when you've purchased some Bare Metal capacity to support your base load, but want to burst into the cloud during peak load, without always running a buffer of DS on cloud VMs.
How Fallback Fleets Work
A primary fleet is just a regular fleet with a fallback configured. There is no difference in how it is configured or scales in and out. A fallback fleet also works like a regular fleet, except that its Buffer Value
for regions which are enabled on both are dynamically adjusted based on the primary fleet's target DS count and Max Servers
setting:
- When the target DS count for the primary fleet is below its
Max Servers
setting, theBuffer Value
of the fallback fleet is set to zero. - When the primary fleet exceeds its
Max Servers
capacity, the fallback fleet'sBuffer Value
increases to provide the extra desired buffer of ready servers.
For example:
- If the primary fleet is configured with 5 DS per host, 50
Max Servers
, and aBuffer Value
of 10:- The primary fleet starts with a target of 10 DS. The fallback fleet's
Buffer Value
is zero. - As players claim DS, the target DS count becomes the claimed DS count + 10.
- When the claimed DS count reaches 41, the primary fleet's target DS count becomes 51, exceeding
Max Servers
, so the fallback fleet'sBuffer Value
is increased to 1. - The fallback fleet starts a watchdog on a new VM. Its buffer increases as needed.
- Once the primary fleet reaches 50 claimed DS, the fallback fleet's buffer is set to 10, and it continues to scale out as needed to maintain a buffer of 10 ready DS.
- The primary fleet starts with a target of 10 DS. The fallback fleet's
Set Up a Primary and Fallback Fleet
To set up a primary/fallback fleet pair:
- Create both fleets as usual. (see Create fleets to deploy dedicated servers for details)
- Edit the fleet you want to be the primary.
- Go to the
Configure Fallback Fleet
tab. - Toggle
Fallback Fleet
toEnabled
. - Choose the fallback fleet from the dropdown.
- Save your changes.
Alternatively:
- Create the fallback fleet first.
- While creating the primary fleet, when you reach the
DS Host & Region
tab: 3. At the bottom, enable a fallback fleet. 4. Choose the fallback fleet from the dropdown.
Both the primary and fallback fleets should have claim keys which will match the same claim requests. See Explore different dedicated server claim scenarios for details on setting up claim keys.
Remove a Fallback Fleet
To convert a primary/fallback fleet pair back to independent fleets:
- Edit the primary fleet.
- Go to the
Configure Fallback Fleet
tab. - Toggle
Fallback Fleet
toNot Configured
. - Save your changes.