Fix random UT failure on nuxeo-stream

2 probems:

• T&P stuck in org.nuxeo.lib.core.mqueues.tests.pattern.TestPatternQueuing#killMQManager probably in future.get()

  10:18:55 Tests run: 4, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 4.15 sec - in org.nuxeo.lib.core.mqueues.tests.TestLibChronicle
  10:18:55 Running org.nuxeo.lib.core.mqueues.tests.pattern.TestPatternQueuingChronicle
  10:18:55 10:18:55,354 [Nuxeo-ConsumerPool-00] WARN  [AbstractCallablePool] Start Nuxeo-Consumer Pool on 1 thread(s).
  10:18:55 10:18:55,378 [Nuxeo-ConsumerPool-00] WARN  [ConsumerPool] Consumers status: threads: 1, failure 0, messages committed: 1, elapsed: 0.01s, throughput: 200.00 msg/s
  10:18:55 10:18:55,382 [Nuxeo-ConsumerPool-00] WARN  [AbstractCallablePool] Start Nuxeo-Consumer Pool on 2 thread(s).
  10:18:55 10:18:55,393 [main] WARN  [TestPatternQueuing] Close the MQManager (errors expected)
  10:18:55 10:18:55,491 [Nuxeo-Consumer-00-1-batch now] ERROR [AbstractCallablePool] Exception catch in runner: The tailer has been closed.
  10:18:55 java.lang.IllegalStateException: The tailer has been closed.
  10:18:55 	at org.nuxeo.lib.core.mqueues.mqueues.chronicle.ChronicleMQTailer.read(ChronicleMQTailer.java:99)
  10:18:55 	at org.nuxeo.lib.core.mqueues.mqueues.chronicle.ChronicleMQTailer.read(ChronicleMQTailer.java:90)
  10:18:55 	at org.nuxeo.lib.core.mqueues.pattern.consumer.internals.ConsumerRunner.acceptBatch(ConsumerRunner.java:269)
  10:18:55 	at org.nuxeo.lib.core.mqueues.pattern.consumer.internals.ConsumerRunner.processBatch(ConsumerRunner.java:222)
  10:18:55 	at org.nuxeo.lib.core.mqueues.pattern.consumer.internals.ConsumerRunner.processBatchWithRetry(ConsumerRunner.java:185)
  10:18:55 	at org.nuxeo.lib.core.mqueues.pattern.consumer.internals.ConsumerRunner.consumerLoop(ConsumerRunner.java:169)
  10:18:55 	at org.nuxeo.lib.core.mqueues.pattern.consumer.internals.ConsumerRunner.call(ConsumerRunner.java:120)
  10:18:55 	at org.nuxeo.lib.core.mqueues.pattern.consumer.internals.ConsumerRunner.call(ConsumerRunner.java:57)
  10:18:55 	at org.nuxeo.lib.core.mqueues.pattern.consumer.internals.AbstractCallablePool.lambda$runPool$1(AbstractCallablePool.java:96)
  10:18:55 	at java.util.concurrent.CompletableFuture$AsyncSupply.run(CompletableFuture.java:1590)
  10:18:55 	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
  10:18:55 	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
  10:18:55 	at java.lang.Thread.run(Thread.java:745)
  10:18:55 10:18:55,492 [Nuxeo-ConsumerPool-00] ERROR [AbstractCallablePool] End of consumer in error: java.lang.IllegalStateException: The tailer has been closed.java.util.concurrent.CompletableFuture@2701bb56[Completed exceptionally]
  12:33:42 Slave went offline during the build
  12:33:42 ERROR: Connection was broken: java.io.IOException: Unexpected termination of the channel
  12:33:42 	at hudson.remoting.SynchronousCommandTransport$ReaderThread.run(SynchronousCommandTransport.java:50)
  12:33:42 Caused by: java.io.EOFException
  12:33:42 	at java.io.ObjectInputStream$PeekInputStream.readFully(ObjectInputStream.java:2624)
  12:33:42 	at java.io.ObjectInputStream$BlockDataInputStream.readShort(ObjectInputStream.java:3099)
  12:33:42 	at java.io.ObjectInputStream.readStreamHeader(ObjectInputStream.java:853)
  12:33:42 	at java.io.ObjectInputStream.<init>(ObjectInputStream.java:349)
  12:33:42 	at hudson.remoting.ObjectInputStreamEx.<init>(ObjectInputStreamEx.java:48)
  12:33:42 	at hudson.remoting.AbstractSynchronousByteArrayCommandTransport.read(AbstractSynchronousByteArrayCommandTransport.java:34)
  12:33:42 	at hudson.remoting.SynchronousCommandTransport$ReaderThread.run(SynchronousCommandTransport.java:48)
  12:33:42 
  12:33:42 Build step 'Invoke top-level Maven targets' marked build as failure
  12:33:42 Performing Post build task...
  12:33:42 Match found for : : True
  12:33:42 Logical operation result is TRUE
  12:33:42 Running script  : if [ -f .ci-metrics-mavenstart ]; then
  12:33:42   echo "Maven $(($(date +%s) - $(cat .ci-metrics-mavenstart)))" >> .ci-metrics
  12:33:42   rm .ci-metrics-mavenstart
  12:33:42 fi
  12:33:42 Exception when executing the batch command : no workspace from node DockerSwarmSlave{name=Docker-slave-20a22eb28e2b-qa-ovh01.nuxeo.com, 
  

  seen only once here: https://qa.nuxeo.org/jenkins/job/TestAndPush/job/ondemand-testandpush-yjulienne/35

• Test failure on TestMQComputationManagerChronicle.testComplexTopoManyRecordsOneThread

  15:01:17 Running org.nuxeo.lib.core.mqueues.tests.computation.TestMQComputationManagerChronicle
  15:02:06 15:02:06,053 [main] ERROR [TestComputationManager] count: 704704, expected: 64064, in 2.27s, throughput: 310032.56 records/s
  15:02:23 Tests run: 14, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 65.301 sec <<< FAILURE! - in org.nuxeo.lib.core.mqueues.tests.computation.TestMQComputationManagerChronicle
  15:02:23 testComplexTopoManyRecordsOneThread(org.nuxeo.lib.core.mqueues.tests.computation.TestMQComputationManagerChronicle)  Time elapsed: 3.87 sec  <<< FAILURE!
  15:02:23 java.lang.AssertionError: expected:<2006> but was:<1912>
  15:02:23 	at org.junit.Assert.fail(Assert.java:88)
  15:02:23 	at org.junit.Assert.failNotEquals(Assert.java:834)
  15:02:23 	at org.junit.Assert.assertEquals(Assert.java:645)
  15:02:23 	at org.junit.Assert.assertEquals(Assert.java:631)
  15:02:23 	at org.nuxeo.lib.core.mqueues.tests.computation.TestComputationManager.testComplexTopo(TestComputationManager.java:163)
  15:02:23 	at org.nuxeo.lib.core.mqueues.tests.computation.TestComputationManager.testComplexTopoManyRecordsOneThread(TestComputationManager.java:185)
  15:02:23 	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)

  https://qa.nuxeo.org/jenkins/job/TestAndPush/job/ondemand-testandpush-yjulienne/37

both the same day Sep 26 on ovh-1