Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
0.17.2, 0.18.1
-
None
-
None
-
Reviewed
Description
When dfs is getting full (80+% of reserved space), the rate of write failures increases, such that more map-reduce tasks can fail. By not cleaning up the temporary output directory of tasks the situation worsens over the lifetime of a job, increasing the probability of the whole job failing.
Attachments
Attachments
Issue Links
- is related to
-
HADOOP-4759 HADOOP-4654 to be fixed for branches >= 0.19
- Closed