Investigate testNodeCatchUpAfterCompaction

[Zelldon]

Description

We have currently an test which is called testNodeCatchUpAfterCompaction in the raft tests, which doesn't work anymore as expected.

With refactoring of Atomix #4466 we migrated this test to rely on the actual log, but it seems to bring a problem or shows a bug it is currently not clear.

  @Test
  @Ignore
  public void testNodeCatchUpAfterCompaction() throws Exception {
    // given
    raftRule.shutdownServer("1");
    raftRule.awaitNewLeader();
    raftRule.appendEntries(100);
    raftRule.tryToCompactLogsOnServersExcept("1", 100).join();

    // when
    final var future = raftRule.startServer("1");

    // then
    future.join();
  }

If we run this test we see the following exception on rejoining of Member one.

06:39:40.505 [] ERROR io.atomix.raft.impl.RaftContext - RaftServer{1} - An uncaught exception occurred, transition to inactive role
java.lang.IndexOutOfBoundsException: Cannot truncate committed index: 0
	at io.atomix.storage.journal.SegmentedJournalWriter.truncate(SegmentedJournalWriter.java:107) ~[classes/:?]
	at io.atomix.storage.journal.SegmentedJournalWriter.reset(SegmentedJournalWriter.java:99) ~[classes/:?]
	at io.atomix.storage.journal.DelegatingJournalWriter.reset(DelegatingJournalWriter.java:61) ~[classes/:?]
	at io.atomix.raft.roles.PassiveRole.appendEntries(PassiveRole.java:523) ~[classes/:?]
	at io.atomix.raft.roles.PassiveRole.handleAppend(PassiveRole.java:387) ~[classes/:?]
	at io.atomix.raft.roles.ActiveRole.onAppend(ActiveRole.java:51) ~[classes/:?]
	at io.atomix.raft.roles.FollowerRole.onAppend(FollowerRole.java:197) ~[classes/:?]
	at io.atomix.raft.impl.RaftContext.lambda$registerHandlers$12(RaftContext.java:222) ~[classes/:?]
	at io.atomix.raft.impl.RaftContext.lambda$runOnContext$19(RaftContext.java:233) ~[classes/:?]
	at io.atomix.utils.concurrent.SingleThreadContext$WrappedRunnable.run(SingleThreadContext.java:188) [classes/:?]
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515) [?:?]
	at java.util.concurrent.FutureTask.run(FutureTask.java:264) [?:?]
	at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:304) [?:?]
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128) [?:?]
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628) [?:?]
	at java.lang.Thread.run(Thread.java:834) [?:?]

We have to investigate how problematic this really is and what it means.