This is related to the work in WT-3276.  If corruption is inserted in the middle of the final (most recent) log file, in the pre-zeroed portion of that file, the corruption is ignored, recovery completes and all records are visible. This is the current behavior without any changes for WT-3276. Recovery should in fact fail, and salvage should truncate the log at the point of corruption.

The work for WT-3276 contains a new test_txn19.py that has an NOTE: comment to highlight the trouble area.