During their life-cycle, jobs transition among the following states:
The job has been created but hasn't been scheduled for execution yet.
Job is being processed.
The job has been processed and enqueued for execution.
The job execution is starting.
BatchX has provisioned a new machine for running the job container, and is copying the job input data to it from the environment file-system.
Job container is running. From this point container
stdout can be received as job logs.
Container execution has finished and BatchX has started uploading the job output files to the environment file-system.
The job has received a cancellation request, but cancellation has not been confirmed yet.
The job has been successfully executed.
The job has been successfully executed but with warnings.
bx run exit code will still be 0, so the curse of client workflow is not broken.
The job has terminated with errors.
The job has been cancelled.
The job has been killed due an out of memory error. This means that the memory specified for the job was not high enough.
The job has generated an invalid output. This error means that the image is faulty and needs to be fixed.
The job execution has been skipped, and the output has been generated from a past equivalent job. See Execution caching for more details.