----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/35251/#review87171 -----------------------------------------------------------
ambari-server/src/main/java/org/apache/ambari/server/controller/internal/BlueprintConfigurationProcessor.java <https://reviews.apache.org/r/35251/#comment139487> I think adding the references to Google Guice are not necessary here. Please see my comments above regarding this approach. Thanks. ambari-server/src/main/java/org/apache/ambari/server/controller/internal/BlueprintConfigurationProcessor.java <https://reviews.apache.org/r/35251/#comment139489> I believe the static updater registrations are sufficient for most cases. Since the stacks don't supply much of this information, and since we're so late in the release cycle for 2.1, we shouldn't be making a change like this at this time. ambari-server/src/test/java/org/apache/ambari/server/controller/internal/BlueprintConfigurationProcessorTest.java <https://reviews.apache.org/r/35251/#comment139490> There don't appear to be any unit tests associated with the new capacity updaters added here. - Robert Nettleton On June 9, 2015, 1:01 p.m., Andrew Onischuk wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/35251/ > ----------------------------------------------------------- > > (Updated June 9, 2015, 1:01 p.m.) > > > Review request for Ambari and Vitalyi Brodetskyi. > > > Bugs: AMBARI-11811 > https://issues.apache.org/jira/browse/AMBARI-11811 > > > Repository: ambari > > > Description > ------- > > This happens because some properties which show capacity don't have 'm' > appended when installing via blueprints (via UI installs it is done by UI) > this results in Oozie thinking that heapsize is in kb and failing with too low > heapsize. > > Also there are some other properties which don't have 'm' appended, but should > have, which results in missconfiguration issues (instead on megabytes > kilobytes are set) > > The proposal on how to fix it is to dynamically find the properties with > unit=MB and append 'm' to them, rather than hardcode that properties like it > is done now. > > > Diffs > ----- > > > ambari-server/src/main/java/org/apache/ambari/server/controller/internal/BlueprintConfigurationProcessor.java > ca7b2b1 > > ambari-server/src/main/java/org/apache/ambari/server/state/ValueAttributesInfo.java > 8054c54 > > ambari-server/src/main/java/org/apache/ambari/server/topology/ClusterConfigurationRequest.java > eb583fd > > ambari-server/src/main/java/org/apache/ambari/server/topology/TopologyManager.java > a75cccb > > ambari-server/src/main/resources/common-services/OOZIE/4.0.0.2.0/configuration/oozie-env.xml > 6439bc6 > > ambari-server/src/main/resources/common-services/OOZIE/5.0.0.2.3/configuration/oozie-env.xml > fe80bf5 > > ambari-server/src/main/resources/common-services/YARN/2.1.0.2.0/metainfo.xml > ca82d46 > > ambari-server/src/test/java/org/apache/ambari/server/api/query/render/ClusterBlueprintRendererTest.java > 96abb8c > > ambari-server/src/test/java/org/apache/ambari/server/controller/internal/BlueprintConfigurationProcessorTest.java > b049fd3 > > ambari-server/src/test/java/org/apache/ambari/server/topology/TopologyManagerTest.java > 64dfb28 > > Diff: https://reviews.apache.org/r/35251/diff/ > > > Testing > ------- > > mvn clean test > > > Thanks, > > Andrew Onischuk > >
