@ancient-wolf-19325 what should I update in my flyte settings to increase my limits?

Request rejected by the API, due to Invalid input.
RPC Failed, with Status: StatusCode.INVALID_ARGUMENT
	Details: Requested MEMORY limit [2Gi] is greater than current limit set in the platform configuration [1Gi]. Please contact Flyte Admins to change these limits or consult the configuration

To increase your task resource limits (e.g., memory) in Flyte, you need to update the platform configuration that sets the maximum allowed resources. For most deployments, this means editing the

task_resource_defaults.yaml

in your Flyte Admin config (often managed via Helm values or a Kubernetes ConfigMap) and increasing the

limits.memory

value. For example, set:

task_resources:
  defaults:
    cpu: 100m
    memory: 500Mi
    storage: 500Mi
  limits:
    cpu: 2
    memory: 2Gi   # <-- increase this value
    storage: 20Mi

After updating, apply the new config and restart the Flyte Admin service for changes to take effect. If you are using project/domain-specific overrides, ensure those are also updated as needed. For more details, see the Flyte task resource configuration guide and customizing task resources. Would you like more detail on how to apply these changes in your specific deployment (e.g., Helm, Flyte sandbox, or manual Kubernetes setup)? Sources: - Customizing task resources - Flyte task resource configuration guide 👍 0 button 👎 0 button 💡 Hint: Mention @runllm in the thread for followups.

@ancient-wolf-19325 what about _cluster_resource_manager.config.cluster_resources.customData_

You can also increase resource limits by updating the

cluster_resource_manager.config.cluster_resources.customData

section in your Flyte Helm values or config. This section sets per-domain or per-environment quotas, such as:

cluster_resource_manager:
  config:
    cluster_resources:
      customData:
        - production:
            - projectQuotaCpu:
                value: "5"
            - projectQuotaMemory:
                value: "4000Mi"
        - staging:
            - projectQuotaCpu:
                value: "2"
            - projectQuotaMemory:
                value: "3000Mi"
        - development:
            - projectQuotaCpu:
                value: "4"
            - projectQuotaMemory:
                value: "3000Mi"

Adjust the

projectQuotaMemory

value for your environment to raise the memory quota. After editing, apply the config and restart the relevant Flyte components. For reference, see flyte-core values.yaml and cluster resource manager docs. Would you like more detail on how this interacts with task resource limits or how to apply these changes? Sources: - flyte-core values.yaml - flyte-core/README.md 👍 0 button 👎 0 button 💡 Hint: Mention @runllm in the thread for followups.

@ancient-wolf-19325 what happens if I don't set

cluster_resources

If you don't set

cluster_resources

in your Flyte configuration, the cluster resource manager will not enforce project/domain-level resource quotas, and Kubernetes namespaces created for Flyte projects/domains will not have resource quotas applied by default. This means resource usage is only limited by global Kubernetes cluster limits and any task-level resource settings, not by Flyte-managed quotas per project or domain. See flyte-core values.yaml and flyte-core/README.md. Would you like more detail on the implications for multi-tenant environments or resource isolation? Sources: - flyte-core values.yaml - flyte-core/README.md 👍 0 button 👎 0 button 💡 Hint: Mention @runllm in the thread for followups.