Skip to content

Experiment with deploying an A2A server agent on k8s to identify any issues/gaps with being able to scale agents #281

@fjuma

Description

@fjuma

It should be possible to have multiple instances of an A2A server agent. We need to investigate how load balancing/session affinity work in this case.

One possibility might be to define a property in application.properties that specifies the public URL using an environment variable and then have the agent card reference this property.

We could check to see if SmallRye Stork could be helpful for this at all.

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions