Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Flyte

hey everyone, we are seeing a lot of *re-starts* in agent pod that do not happen to any of our other pods, we are curious if you guys have seen this before or have any idea why this might happen?

```yaou@yaou-mn1 ~ % k get pods
NAME                               READY   STATUS    RESTARTS       AGE
ambassador-69d4998797-2rf9r        1/1     Running   0              4d23h
datacatalog-864796d889-ngqx2       1/1     Running   0              4d23h
flyte-pod-webhook-76d88c67-6x7rb   1/1     Running   0              4d23h
flyteadmin-5cf46b95c7-8dnnk        1/1     Running   0              4d23h
flyteagent-56bd59b5d4-9vjcg        1/1     Running   37 (33m ago)   6d3h
flyteconsole-69c4c8795d-hnhg8      1/1     Running   0              67d```

or the flytekit code in agent will some how crash

you have to print logs as much as possible to ensure that you didn't crash  by the flytekit

no worries, we just set it to log as much as possible, will reach back out if there’s findings / difficulty , thanks !

<https://github.com/Future-Outlier/flyte-custom-agent-template>

this can help you create your custom agent more easily