[MAPREDUCE-5760] Reduce task percentage is going beyond 100% for a job

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 2.3.0
• Assignee: Manoj Kumar Danthala
• Reporter: Nishan Shetty
• Created: 20/Feb/14 03:30
• Updated: 31/Oct/14 05:11

Description

1.Run job with some 600 maps and 1 reduce task

Observe that for that Job reduce percentage is going beyond 100%
Finally job is succeded.

Client Log
--------------
2014-02-06 14:10:13,647 INFO org.apache.hadoop.mapreduce.Job: map 100% reduce 128%
2014-02-06 14:10:19,687 INFO org.apache.hadoop.mapreduce.Job: map 100% reduce 135%
2014-02-06 14:10:22,706 INFO org.apache.hadoop.mapreduce.Job: map 100% reduce 143%
2014-02-06 14:10:25,724 INFO org.apache.hadoop.mapreduce.Job: map 100% reduce 149%
2014-02-06 14:10:28,493 INFO org.apache.hadoop.mapreduce.Job: map 100% reduce 153%
2014-02-06 14:10:31,512 INFO org.apache.hadoop.mapreduce.Job: map 100% reduce 157%
2014-02-06 14:10:34,553 INFO org.apache.hadoop.mapreduce.Job: map 100% reduce 163%
2014-02-06 14:10:37,572 INFO org.apache.hadoop.mapreduce.Job: map 100% reduce 171%

AM Log
--------------
2014-02-06 14:13:27,590 INFO [IPC Server handler 26 on 65120] org.apache.hadoop.mapred.TaskAttemptListenerImpl: Progress of TaskAttempt attempt_1391663543178_0003_r_000000_0 is : 1.7558539
2014-02-06 14:13:30,566 INFO [IPC Server handler 25 on 65120] org.apache.hadoop.mapred.TaskAttemptListenerImpl: Status update from attempt_1391663543178_0003_r_000000_0
2014-02-06 14:13:30,566 INFO [IPC Server handler 25 on 65120] org.apache.hadoop.mapred.TaskAttemptListenerImpl: Progress of TaskAttempt attempt_1391663543178_0003_r_000000_0 is : 1.7558539

Issue Links

• is related to: MAPREDUCE-5958 Wrong reduce task progress if map output is compressed