To my surprise/dismay, it seems telling Heritrix to only keep the last checkpoint also means it deletes the previous checkpoint log files! This doesn't cause a problem when we're promptly syncing to HDFS, but is not desirable on other systems or in case we hit a bottleneck.
To my surprise/dismay, it seems telling Heritrix to only keep the last checkpoint also means it deletes the previous checkpoint log files! This doesn't cause a problem when we're promptly syncing to HDFS, but is not desirable on other systems or in case we hit a bottleneck.