[DRILL-3937] We are not pruning when we have a metadata cache and auto partitioned data in some cases

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.3.0
• Component/s: Metadata
• Assignee: Aman Sinha
• Reporter: Rahul Kumar Challapalli
• Created: 15/Oct/15 01:59
• Updated: 10/Dec/15 00:25
• Resolved: 30/Oct/15 02:02

Description

git.commit.id.abbrev=2736412

The below plan indicates that we are not pruning

explain plan for select count(*) from dfs.`/drill/comscore/orders2` where o_clerk='Clerk#000079443';
+------+------+
| text | json |
+------+------+
| 00-00    Screen
00-01      Project(EXPR$0=[$0])
00-02        StreamAgg(group=[{}], EXPR$0=[COUNT()])
00-03          Project($f0=[0])
00-04            SelectionVectorRemover
00-05              Filter(condition=[=($0, 'Clerk#000079443')])
00-06                Scan(groupscan=[ParquetGroupScan [entries=[ReadEntryWithPath [path=maprfs:///drill/comscore/orders2/1_0_9999.parquet], ReadEntryWithPath [path=maprfs:///drill/comscore/orders2/1_0_9998.parquet]], selectionRoot=/drill/comscore/orders2, numFiles=2, usedMetadataFile=true, columns=[`o_clerk`]]])

Error from the logs

2015-10-15 01:24:28,467 [29e0ffb4-1c91-f40a-8bf0-5e3665dcf107:foreman] WARN  o.a.d.e.p.l.partition.PruneScanRule - Exception while trying to prune partition.
java.lang.ClassCastException: java.util.LinkedHashMap cannot be cast to parquet.io.api.Binary
        at org.apache.drill.exec.store.parquet.ParquetGroupScan.populatePruningVector(ParquetGroupScan.java:414) ~[drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.drill.exec.planner.ParquetPartitionDescriptor.populatePartitionVectors(ParquetPartitionDescriptor.java:96) ~[drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.drill.exec.planner.logical.partition.PruneScanRule.doOnMatch(PruneScanRule.java:212) ~[drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.drill.exec.planner.logical.partition.ParquetPruneScanRule$2.onMatch(ParquetPruneScanRule.java:87) [drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.calcite.plan.volcano.VolcanoRuleCall.onMatch(VolcanoRuleCall.java:228) [calcite-core-1.4.0-drill-r6.jar:1.4.0-drill-r6]
        at org.apache.calcite.plan.volcano.VolcanoPlanner.findBestExp(VolcanoPlanner.java:808) [calcite-core-1.4.0-drill-r6.jar:1.4.0-drill-r6]
        at org.apache.calcite.tools.Programs$RuleSetProgram.run(Programs.java:303) [calcite-core-1.4.0-drill-r6.jar:1.4.0-drill-r6]
        at org.apache.calcite.prepare.PlannerImpl.transform(PlannerImpl.java:303) [calcite-core-1.4.0-drill-r6.jar:1.4.0-drill-r6]
        at org.apache.drill.exec.planner.sql.handlers.DefaultSqlHandler.logicalPlanningVolcanoAndLopt(DefaultSqlHandler.java:545) [drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.drill.exec.planner.sql.handlers.DefaultSqlHandler.convertToDrel(DefaultSqlHandler.java:213) [drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.drill.exec.planner.sql.handlers.DefaultSqlHandler.convertToDrel(DefaultSqlHandler.java:248) [drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.drill.exec.planner.sql.handlers.ExplainHandler.getPlan(ExplainHandler.java:61) [drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.drill.exec.planner.sql.DrillSqlWorker.getPlan(DrillSqlWorker.java:178) [drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.drill.exec.work.foreman.Foreman.runSQL(Foreman.java:905) [drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at org.apache.drill.exec.work.foreman.Foreman.run(Foreman.java:244) [drill-java-exec-1.2.0-SNAPSHOT.jar:1.2.0-SNAPSHOT]
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) [na:1.7.0_71]
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [na:1.7.0_71]
        at java.lang.Thread.run(Thread.java:745) [na:1.7.0_71]

The partition column type in this case is binary which could be causing the issue.

Partition pruning seems to be working when we have Metadata Caching + Auto Partitioned Files with integer partition column