After a replica set test has completed, the script will find the lines in the log file that were logged specifically for TLA+ trace checking. It will merge together these lines from every replica set member's log, to produce a single execution trace for the whole system. (It will assume it can rely on timestamp order because the mongods are running on the same machine.)

Output the execution trace in a format suitable for checking it against the traces permitted by our TLA+ specs. (This format is TBD, see discussion in the TLA+ Toolbox Github project.)