Scheduled runs can now tolerate Kubernetes node taints
Platform administrators can now configure scheduled jobs to tolerate specific node taints, unlocking the ability to route these runs to dedicated, isolated Kubernetes node pools.
Platform administrators can now configure scheduled runs to tolerate specific Kubernetes node taints. By passing a comma-separated list of tolerations, teams can route scheduled jobs onto dedicated, tainted node pools. This isolation ensures that heavy, predictable scheduled runs do not compete with standard workloads for general compute resources. The configuration happens at startup in the supervisor application, parsing the syntax and failing fast if the toleration format is invalid.
View Original GitHub Description
Adds support for taint tolerations for scheduled runs. Useful for selectively tolerating taints on dedicated node pools.
The new KUBERNETES_SCHEDULED_RUN_TOLERATIONS env variable accepts a comma-separated list in the format key=value:effect (or key:effect for the Exists operator).
Drive-by: renames all KUBERNETES_SCHEDULE_* affinity env vars to KUBERNETES_SCHEDULED_RUN_* for clarity — this feature isn't used in production yet or published in a tagged image; the name change is fine.