This is the documentation for CDH 4.6.0.
Documentation for other versions is available at Cloudera Documentation.

Known Issues Fixed in CDH4.5.0

This section lists important problems fixed in CDH4.5.0. For more information about upstream fixes applied in CDH4.5.0, see What's New in CDH4.5.0.

Apache HDFS

transitionToActive blocks if the SBN is doing checkpoint image transfer

-transitionToActive should cause an immediate transition, but it does not if the Standby node is doing a checkpoint image transfer.

Bug: HDFS-4816

Resolution: Fixed in CDH 4.5.0

—Secondary NameNode fails to checkpoint if error occurs downloading edits on first checkpoint

If an error occurs between downloading and loading the FSImage on the first checkpoint attempt, the Secondary NameNode will never load the FSImage. Then on subsequent checkpoint attempts it will not load the on-disk FSImage and thus will never checkpoint successfully.

Bug: HDFS-5159

Resolution: Fixed in CDH 4.5.0

Apache MapReduce

— JobTracker memory leak in CDH4.4

The JobTracker has a memory leak caused by subtleties in the way UserGroupInformation interacts with the file-system cache. The number of cached file-system objects can grow without bound. This issue has appeared in previous releases as well, albeit for slightly different reasons.


Severity: Medium

Resolution: Fixed in CDH 4.5.0

Workaround: Set keep.failed.task.files to true, which will sidestep the memory leak but require job staging directories to be cleaned out manually.