Hi everyone, I want to ask something about RunConf...
# ask-community
r
Hi everyone, I want to ask something about RunConfig and Executor. Im using KubernetesAgent on GKE. My KubernetesRun config was:
Copy code
KubernetesRun(
    labels=["agent:k8s-dev"],
    cpu_limit=8,
    memory_limit="12Gi",
    cpu_request=0.5,
    memory_request="512Mi",
)
And my executor was:
Copy code
DaskExecutor(adapt_kwargs={"maximum": 16, "minimum": 1})
When I start a flowrun with these config, prefect agent create a new pod with this config: (image attached) The problem is: The dask worker always killed first before the GKE pod going for upscale, and prefect logging says error KilledWorker Any idea how to perfectly adjust between DaskExecutor and KubernetesRun?
d
Hey @Ranu Goldan, Are you able to run successfully when the DaskExecutor is using a static number of workers?
r
@Dylan I just tried with this dask config:
Copy code
executor = DaskExecutor(
    # adapt_kwargs={"maximum": 16, "minimum": 1},
    cluster_kwargs={"num_workers": 16},
)
Then it just stucks while creating dask cluster (image attached)