----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/28036/#review61429 -----------------------------------------------------------
Ship it! Ship It! - Dmitro Lisnichenko On Nov. 14, 2014, 11:49 a.m., Dmytro Sen wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/28036/ > ----------------------------------------------------------- > > (Updated Nov. 14, 2014, 11:49 a.m.) > > > Review request for Ambari, Andrew Onischuk and Dmitro Lisnichenko. > > > Bugs: AMBARI-8332 > https://issues.apache.org/jira/browse/AMBARI-8332 > > > Repository: ambari > > > Description > ------- > > Currently hive client retry settings are recommended to be: > hive.metastore.failure.retries=5 > hive.metastore.client.connect.retry.delay=5 > So we have overall time of 25 seconds to restart Hive. This doesn't seem > reasonable in a rolling upgrade scenario where we accomplish three things: > - Stop the Hive metastore > - Upgrade the schema (run the schemaTool upgrade) > - Start the Hive metastore > For now I think we should raise it to 2 minutes. We could do the following: > hive.metastore.failure.retries=24 > hive.metastore.connect.retries=24 > hive.metastore.client.connect.retry.delay=5s > > > Diffs > ----- > > > ambari-server/src/main/resources/stacks/HDP/2.2/services/HIVE/configuration/hive-site.xml > 12b4612 > > Diff: https://reviews.apache.org/r/28036/diff/ > > > Testing > ------- > > ---------------------------------------------------------------------- > Ran 242 tests in 3.192s > > OK > ---------------------------------------------------------------------- > Total run:684 > Total errors:0 > Total failures:0 > OK > > > Thanks, > > Dmytro Sen > >
