New RunLLM response generated: >>>*# Question* I have updated my flyteadmin GRPC size limit and can see it in the configmap on K8: flyteadmin: ... server: grpc: maxMessageSizeBytes: 52428800 However, even after this change (and restart of the admin), for large map tasks (3000 tasks or so), we are getting Request Too Large errors from flytepropeller. I am wondering if this is due to the GRPC limit not taking affect, or if this request is not going to FlyteAdmin and somewhere else, so perhaps I need to increase this limit elsewhere. Would appreciate your help. # Answer The issue might be due to the gRPC size limit not being updated in all necessary components. According to this discussion, the gRPC size limit needs to be configured in both FlyteAdmin and FlytePropeller. To address this, ensure that the

maxMessageSizeBytes

parameter is also set in the FlytePropeller configuration. You can do this by updating the FlytePropeller configmap similarly to how you updated FlyteAdmin. Here's an example of how to set it in the FlytePropeller config:

apiVersion: v1
kind: ConfigMap
metadata:
  name: flyte-propeller-config
  namespace: <your-namespace>
data:
  config.yaml: |
    propeller:
      ...
      maxMessageSizeBytes: 52428800

After updating the configmap, restart the FlytePropeller pods to apply the changes. Would you like more detailed steps on how to update the FlytePropeller configuration? Sources: - Flyte Discuss: Increasing gRPC size limit - Flyte Discuss: Large fan outs and gRPC size limit Slack Link https://flyte-org.slack.com/archives/CP2HDHKE1/p1721233277.429179 👍 0 button 👎 0 button 💡 Hint: Mention @runllm in the thread for followups.