Spark Stand Alone Mode

What would happen if the resources required by a spark cluster deployment are not available ? By spark cluster deployment I mean starting the master and workers for a single standalone cluster with certain requirements. Will the spark cluster begin with the resources that are available and then get other resources later or will the application return an error ?

These are the few cases that I can think of. I am sure there are many more but I could not find any info on this topic.

1.Spark cluster needs 4 nodes and 3 are available.

2.Deployment requires 3 workers with 2 cores each but only 2 workers are available with 2 cores, the third one can only get one core at the moment.

3.Deployment requires 3 workers with 500 mb each but only two workers can get 500mb and the third one cannot get 500mb.

It would help me if you also provided a source of the answer.