-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/44594/#review122815
-----------------------------------------------------------


Ship it!




Ship It!

- Aravindan Vijayan


On March 9, 2016, 9:44 p.m., Sid Wagle wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/44594/
> -----------------------------------------------------------
> 
> (Updated March 9, 2016, 9:44 p.m.)
> 
> 
> Review request for Ambari, Aravindan Vijayan, Dmytro Sen, and Sumit Mohanty.
> 
> 
> Bugs: AMBARI-15361
>     https://issues.apache.org/jira/browse/AMBARI-15361
> 
> 
> Repository: ambari
> 
> 
> Description
> -------
> 
> Two Alter table calls one of TTL and the other for setting the Region 
> policies could result in a possible race condition on HBase side. 
> 
> Here is some details:
> The assignment operation on the region in question 
> ({{2e13a2d264c50e27c32cd4d3cad6ebca}}) further indicates possible network 
> partition - the region is CLOSED and OFFLINED successfully, and trying to 
> OPEN again, but it found conflict information from zk node (the zk node had a 
> stale information and still thought that it is still in the CLOSING state, 
> which is not true based on the log) - the assignment succeeds, but updating 
> the assignment information to zk failed due to incorrect version in the zk 
> node.
> 
> 
> Diffs
> -----
> 
>   
> ambari-metrics/ambari-metrics-timelineservice/src/main/java/org/apache/hadoop/yarn/server/applicationhistoryservice/metrics/timeline/HBaseTimelineMetricStore.java
>  f460292 
>   
> ambari-metrics/ambari-metrics-timelineservice/src/main/java/org/apache/hadoop/yarn/server/applicationhistoryservice/metrics/timeline/PhoenixHBaseAccessor.java
>  4afb722 
> 
> Diff: https://reviews.apache.org/r/44594/diff/
> 
> 
> Testing
> -------
> 
> Unit tests passed. No funtional change.
> 
> 
> Thanks,
> 
> Sid Wagle
> 
>

Reply via email to