Release Notes for Release on July 02, 2015
Minor changes and bug fixes are as listed below.
HADOOP
-
Introduce fs.s3n.parent.stat.disabled configuration variable to disable stat calls if needed while listing S3 directories
-
QBOL-4231: JTProxy mark each HTTP connection to close on response.
-
Upgrade jets3t to include exponential backoff in case of s3-xml-validation errors for listBuckets
HIVE-13
-
HIVE-838: Optimize Insert overwrite by skipping redundant directory checks
-
HAD-458: Optimizing summary computation for reducer estimation
-
HIVE-878, HIVE-918 Fix: intermittent NullPointerException during stats collection on completion of short MR jobs.
PRESTO
-
Presto has been upgraded to 0.105
-
PRES-202: Auto-scaling process can be configured to run in iterations adding/removing ‘x’ nodes at a time instead to prevent aggressive scaling
-
PRES-402: Number of elements in ‘IN’ clause that are allowed to be pushed down as is without conversion to range is now configurable
QDS
-
UI-1996: Error while creating a DB tap connection
-
Ignore all other filters if command ID is specified
-
UI-1925: Explore refresh button does not work if user does not have data store create perm
-
UI-1974: Selecting and closing the S3 location changes the folder icon
-
UI-2008: When there is no cluster type for a query, Analyze is not showing validation message and Run is enabled
-
Reduce NN and increase JT, RM default heap-size for r3.2xlarge instance type
-
HAD-453: For VPC configured with multiple domain names (separated by whitespace), use FQDN as value for slave.host.name property
-
UI-1907: Unable to list my bucket in certain cases
-
UI-1956: Improve saved query run instances query time
-
Show only last 7 days commands by default
-
UI-1897: updated message in saved queries runs instances dom
-
UI-1879: Hive bootstrap display messed up on the UI
SPARK
-
SPAR-370: Prevent exception during exception handling
ZEPPELIN
-
SPAR-357: If settings are not correct for an interpreter, an error message should be shown
Comments