----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/626/#review671 -----------------------------------------------------------
trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/Utilities.java <https://reviews.apache.org/r/626/#comment1339> The loop wrongly assumes that calling Thread.currentThread().interrupt() will throw ThreadInterrupted while it only sets interrupted flag on the thread. The result is infinite loop with executionDone never set to true. The loop is actually not needed as executor.shutdown() does the same thing - waits for all queued jobs to complete. - georgedj On 2011-05-10 21:55:56, georgedj wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/626/ > ----------------------------------------------------------- > > (Updated 2011-05-10 21:55:56) > > > Review request for hive. > > > Summary > ------- > > Fixed by adding INT signal hander that interrupts the CLI thread. The CLI > thread gets InterruptedException and it stops the current command. > > > This addresses bug HIVE-243. > https://issues.apache.org/jira/browse/HIVE-243 > > > Diffs > ----- > > trunk/cli/src/java/org/apache/hadoop/hive/cli/CliDriver.java 1100312 > > trunk/common/src/java/org/apache/hadoop/hive/common/HiveInterruptCallback.java > PRE-CREATION > trunk/common/src/java/org/apache/hadoop/hive/common/HiveInterruptUtils.java > PRE-CREATION > trunk/metastore/src/java/org/apache/hadoop/hive/metastore/ObjectStore.java > 1100312 > trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/HadoopJobExecHelper.java > 1100312 > trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/Utilities.java 1100312 > > Diff: https://reviews.apache.org/r/626/diff > > > Testing > ------- > > Manual test Ctrl+C on bunch of command (show tables, select, etc) > Tested interrupting the command on large data set in different stages of the > processing (JDO, MapReduce, building the input summary). > example queries: > explain select count(*) from xyz where ds = '2011-02-01'; > select count(*) from xyz where ds = '2011-02-01'; > > > Thanks, > > georgedj > >
