-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/46200/#review128891
-----------------------------------------------------------


Ship it!




Ship It!

- Andrew Onischuk


On April 14, 2016, 1:13 p.m., Dmytro Sen wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/46200/
> -----------------------------------------------------------
> 
> (Updated April 14, 2016, 1:13 p.m.)
> 
> 
> Review request for Ambari, Andrew Onischuk, Sumit Mohanty, and Vitalyi 
> Brodetskyi.
> 
> 
> Bugs: AMBARI-15506
>     https://issues.apache.org/jira/browse/AMBARI-15506
> 
> 
> Repository: ambari
> 
> 
> Description
> -------
> 
> Currently Ambari does not support the definition of multiple Nameservices. It 
> is always assumed hdfs_site['dfs.nameservices'] is just a string defining one 
> nameservice.
> Multiple nameservices can be configured for exmaple to support seamless 
> distcp between two HA clusters. The nameservices are defined as a comma 
> separated list in hdfs_site['dfs.nameservices'].
> This patch introduces the method get_nameservice(hdfs_site), which splits the 
> value into in an array and identifies the nameservice for the current cluster 
> with what is set in hdfs_site['dfs.namenode.shared.edits.dir']. By default 
> the first nameservice is returned or empty.
> To verify the current namservice it would be preferred to use fs.defaultFS in 
> core-site, but getting this config into namenode_ha_utils.py seems more 
> involved.
> 
> 
> Diffs
> -----
> 
>   ambari-agent/src/test/python/resource_management/TestNamenodeHaUtils.py 
> PRE-CREATION 
>   
> ambari-common/src/main/python/resource_management/libraries/functions/namenode_ha_utils.py
>  b6f3bee 
> 
> Diff: https://reviews.apache.org/r/46200/diff/
> 
> 
> Testing
> -------
> 
> ----------------------------------------------------------------------
> Ran 426 tests in 17.864s
> 
> OK
> 
> 
> Thanks,
> 
> Dmytro Sen
> 
>

Reply via email to