shardul-cr7 commented on a change in pull request #3119: [DOC]Updated the documentation for a varchar scenario and compressor supported.
URL:
https://github.com/apache/carbondata/pull/3119#discussion_r265043105
##########
File path: docs/faq.md
##########
@@ -493,3 +494,28 @@ Note : Refrain from using "mvn clean package" without specifying the profile.
Drop that particular datamap using Drop Table command using table name as
parentTableName_datamapName so as to clear the stale folders.
+
+ ## Failed to execute select query on column having varchar datatype.
+
+ **Symptom**
+
+ Execution fails with following exception :
+
+ ```
+ java.lang.RuntimeException: java.lang.RuntimeException: Cannot reserve additional contiguous bytes in the vectorized reader (requested 1839007356 bytes).
+ at org.apache.carbondata.core.datastore.chunk.impl.DimensionRawColumnChunk.convertToDimColDataChunkAndFillVector(DimensionRawColumnChunk.java:140)
+ at org.apache.carbondata.core.scan.scanner.LazyPageLoader.loadPage(LazyPageLoader.java:74)
+ at org.apache.spark.sql.CarbonVectorProxy$ColumnVectorProxyWithLazyLoad.checkPageLoaded(CarbonVectorProxy.java:511)
+ ```
+
+ **Possible Cause**
+
+ JVM cannot reserve the additional contigous bytes in the the vectorized reader.
+
+ **Procedure**
+
+ For carbondata, as a workaround you can try decreasing the number of executor cores or increasing the executor memory.If this does not work you can disable the following property in carbon.properties.
Review comment:
No general guideline as we don't know exactly for what executor memory size this scenario won't occur .Depends on the user how much extra executor memory he can afford to allocate for the job.Will also depends on the size of data.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[hidden email]
With regards,
Apache Git Services
Locked
