I ran into a task being scheduled but never starting due to a system problem in k8s blocking it (a failing mount) -- I can see there's the

timout

arg for

@task

decorators, but just making sure there isn't a system-wide default that can be configured?

there is a systemwide default, per node

@bland-dress-83134 it should be under

propeller:
  node-config:
    default-deadlines:
      node-active-deadline: 0s

there are other parameters you can adjust: https://www.union.ai/docs/flyte/deployment/configuration-reference/flytepropeller-config/#node-config-confignodeconfig The actual location of the config depends on the Helm chart you're using

this is just a total time limit for tasks including queueing right? My situation was that the task was considering "running" from Flyte's perspective since the pod was created, but the pod just stayed in Pending indefinitely: I think what I was really looking for was a timeout from pod created -> pod running, but Flyte probably doesn't keep track this closely. 🤔

Aah yes, sad kubernetes problem