Improper cleanup between runs causes erroneous timeouts to be triggered

[ivoadf]

These two variables are not properly reset when a run completes

    private void checkSampleTimeout() {
        long previousSampleCount = 0;
        long previousSampleTime = 0;

while noResultsReturned = 0; is reset in the endOfRun method.

This means that once the first run completes and noResultsReturned is reset to 0 this if condition is never true

if (previousSampleCount < noResultsReturned) {
                previousSampleCount = noResultsReturned;
                previousSampleTime = System.currentTimeMillis();
            } else {
                if (previousSampleTime < (System.currentTimeMillis() - jlbhOptions.timeout)) {
                    printStream.println("Sample timed out. Aborting test...");
                    abort();
                    break;
                }
            }

Meaning the previousSampleCount and previousSampleTime are never updated and unless the jlbhOptions.timeout value is larger than the total runtime of all remaining runs in the test it will eventually trigger a timeout when it should not.

To fix this previousSampleCount and previousSampleTime should be reset alongside the reset that already happens to noResultsReturned in the endOfRun method.

[JerryShea]

@ivoadf can you provide a failing unit test pls?