> On May 29, 2015, 7:51 p.m., Vaibhav Gumashta wrote: > > itests/hive-unit/src/test/java/org/apache/hive/jdbc/TestJdbcWithMiniHS2.java, > > line 126 > > <https://reviews.apache.org/r/34727/diff/3/?file=974222#file974222line126> > > > > Test looks good. Can you also add a Statement#cancel in the mix? > > Chaoyu Tang wrote: > Vaibhav, thanks for review. There is already a case testCancel provided > in HIVE-6472. Do you want to incorporate the same (similar one) into the test > of this patch? or could you elaborate more?
Thanks for pointing to that test; I think that should be sufficient to cover the cancel use case. - Vaibhav ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/34727/#review85782 ----------------------------------------------------------- On May 29, 2015, 11:44 a.m., Chaoyu Tang wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/34727/ > ----------------------------------------------------------- > > (Updated May 29, 2015, 11:44 a.m.) > > > Review request for hive, Szehon Ho, Thejas Nair, and Xuefu Zhang. > > > Bugs: HIVE-10835 > https://issues.apache.org/jira/browse/HIVE-10835 > > > Repository: hive-git > > > Description > ------- > > There exist race conditions between DatabaseMetaData, Statement and ResultSet > when they make RPC calls to HS2 using same Thrift transport, which happens > within same connection. > The patch is to have a connection level lock to serialize the RPC calls > within a single connection. > > > Diffs > ----- > > > itests/hive-unit/src/test/java/org/apache/hive/jdbc/TestJdbcWithMiniHS2.java > 7210480 > jdbc/src/java/org/apache/hive/jdbc/HiveConnection.java 1b2891b > jdbc/src/java/org/apache/hive/jdbc/HiveQueryResultSet.java e93795a > jdbc/src/java/org/apache/hive/jdbc/HiveStatement.java 6b3d05c > > Diff: https://reviews.apache.org/r/34727/diff/ > > > Testing > ------- > > Some multi-thread tests (80 threads for 200 iterations) > > > Thanks, > > Chaoyu Tang > >
