[MAPREDUCE-4425] Speculation + Fetch failures can lead to a hung job

• Type: Bug
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 0.23.1
• Fix Version/s: 2.0.3-alpha, 0.23.5
• Component/s: mrv2
• Target Version/s: 2.0.3-alpha, 0.23.5
• Assignee: Jason Darrell Lowe
• Reporter: Siddharth Seth
• Created: 10/Jul/12 19:08
• Updated: 03/Sep/14 23:17
• Resolved: 12/Nov/12 19:03

Description

After a task goes to SUCCEEDED, FAILED/KILLED attempts are ignored.
1. attemp1 starts
2. speculative attempt starts
3. attempt 1 completes - Task moves to SUCCEEDED state
4. speculative attempt is KILLED
5. T_ATTEMPT_KILLED is ignored.
6. attemp1 1 fails with TOO_MANY_FETCH_FAILURES
The job will effectively hang, since a new task attempt isn't started.