Export - CSV (All fields)
Export - CSV (Current fields)
MAPREDUCE-4061
RM only has 1 AM launcher thread
Secret keys set in Credentials are not seen by tasksMAPREDUCE-4039
Sort Avoidance
Unable to view task logs on history server with mapreduce.job.acl-view-job=*
AM can crash if task attempt reports bogus progress value
history server container log web UI sometimes combines stderr/stdout/syslog contents together
AM container logs URL is broken for completed apps when log aggregation is enabled
zookeeper artifact is missing from the hadoop-dist assemblyMAPREDUCE-3989
cap space usage of default log4j rolling policy (mr specific changes)
handleTaskAttemptCompletion() called twice for same task attempt from KillWaitAttemptKilledTransition & AttemptKilledTransition
TestEmptyJob fails with FileNotFound
LogAggregationService leaks log aggregator objects
TestRMContainerAllocator failing
Default value not set for Configuration parameter mapreduce.job.local.dir
ResourceManager does not have JVM metrics
Map/ReduceSlotMillis computation incorrect
web proxy doesn't forward request to AM with configured hostname/IP
Clean up passing HEAPSIZE to yarn and mapred commands.MAPREDUCE-3944
JobHistory web services are slower then the UI and can easly overload the JH
MR tasks failing due to changing timestamps on Resources to download
The AM page for a Reducer that has not been launched causes an NPE
output of mapred -showacl is not clear
Fix the potential problem compiling 32 bit binaries on a x86_64 host.
Revise yarn default port number selection
proc_historyserver no longer in command line arguments for HistoryServer
RM application webpage is unresponsive after 2000 jobs
user not allowed to submit jobs even though queue -showacls shows it allows
[NPE] Job history produced with mapreduce.cluster.acls.enabled false can not be viewed with mapreduce.cluster.acls.enabled true
no admin override to view jobs on mr app master and job history server
lazy load JobHistory Task and TaskAttempt details
capacity scheduler - maxActiveApplicationsPerUser calculation can be wrong
pig job through oozie hangs
Change to nodemanager build now requires 32-bit libraries
PWD should be first in the classpath of MR tasks
Null user on filtered jobhistory job page
MAPREDUCE-3877
Add a test to formalise the current state transitions of the yarn lifecycle
bin/yarn prints the command line unnecessarily
Fix cluster setup docs for correct SNN HTTPS parameters
Nodemanager can appear to hang on shutdown due to lingering DeletionService threads
TestLinuxResourceCalculatorPlugin is failing on trunk and 0.23 branch.
test TestLinuxResourceCalculatorPlugin failing
Change TokenCache's reading of the binary token file
MapReduce job submission time has increased in 0.23 when compared to 0.20.206
capacity scheduler web ui bar graphs for used capacity wrong- MAPREDUCE-3800
TestHarFileSystem testRelativeArchives - MAPREDUCE-3799
TestServiceLevelAuthorization testServiceLevelAuthorization failing - MAPREDUCE-3798
TestJobCleanup testCustomCleanup is failing
job -list displays only the jobs submitted by a particular user
Broken pipe on streaming job can lead to truncated output for a successful job
Add the equivalent of JobStatus to end of JobHistory file