Document best practices for multinode synchronisation
Explain how to reorganise multinode jobs when porting from the current dispatcher to the pipeline.
0: Organise the test action to have inline for calls to lava-sync, lava-wait, lava-wait-all and lava-network. 1: Do other test tasks in other inline or in external repositories 2: Develop the skeleton of a job independently of the (possibly lengthy) individual tasks.
Provide use case.
Sub task of issue #1
Migrated from https://projects.linaro.org/browse/LAVA-321