Ensure that timeout set on a Script task is propagated to a timeout of a Cloud container (AWS/Azure/Google Batch Script Runners)

[anna-geller]

Feature description

When a user sets a timeout, we should ensure that the configured timeout will be roughly respected.

We say roughly because there can be some significant latency for starting a container (on Fargate, even in a range of minutes), so ideally the timeout should start once the container starts. However, the main use case is matching kestra timeout with Cloud container service timeout (even if not 100% accurate) to avoid unexpected costs caused by not respecting timeout and having a zombie container running for weeks.

id: timeout
namespace: dev

tasks:
  - id: failed
    type: io.kestra.plugin.scripts.shell.Commands
    scriptRunner: ...
    timeout: PT2H