----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/39693/#review104207 -----------------------------------------------------------
Ship it! Ship It! - Sid Wagle On Oct. 27, 2015, 7:42 p.m., Aravindan Vijayan wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/39693/ > ----------------------------------------------------------- > > (Updated Oct. 27, 2015, 7:42 p.m.) > > > Review request for Ambari, Dmytro Sen, Sumit Mohanty, Srimanth Gunturi, and > Sid Wagle. > > > Bugs: AMBARI-13585 > https://issues.apache.org/jira/browse/AMBARI-13585 > > > Repository: ambari > > > Description > ------- > > Problem > Flume Metrics shown in the flume summary page of Ambari have very small > values when correlated with the events being generated by the flume agents. > > Bug > AMS handles flume metrics in a special way since they are "rate" metrics. > While trying to calculate the "per second" metric value, the code calculates > the "per millisecond" value instead. > > Fix > Changed calculation logic to "per second". > > > Diffs > ----- > > > ambari-metrics/ambari-metrics-timelineservice/src/main/java/org/apache/hadoop/yarn/server/applicationhistoryservice/metrics/timeline/HBaseTimelineMetricStore.java > 21f193f > > Diff: https://reviews.apache.org/r/39693/diff/ > > > Testing > ------- > > Maunal Testing by creating flume agents. > > mvn clean test on ambari-metrics. > > > Thanks, > > Aravindan Vijayan > >
