-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/35351/#review87575
-----------------------------------------------------------



ambari-server/src/main/resources/common-services/STORM/0.9.1.2.1/package/scripts/params_linux.py
<https://reviews.apache.org/r/35351/#comment139962>

    I want to clarify that this value should be 
    Math.floor(n/2) + 1. to give concrete examples:
    
    if n = 1, the value should be 1.
    if n = 2, the value should be 2.
    if n = 3, the value should still be 3.
    if n = 4, the value should be 3.
    if n = 5, the value should be 3.
    
    I am no java script expert but i think without the floor n/2 + 1 will 
produce a float i.e. for n = 3 the result will be 2.5 which will brake storm. 
we need to ensure this property is an integer.


- Parth Brahmbhatt


On June 11, 2015, 12:15 p.m., Dmitro Lisnichenko wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/35351/
> -----------------------------------------------------------
> 
> (Updated June 11, 2015, 12:15 p.m.)
> 
> 
> Review request for Ambari, Jonathan Hurley, Parth Brahmbhatt, Sriharsha 
> Chintalapani, and Vitalyi Brodetskyi.
> 
> 
> Bugs: AMBARI-11435
>     https://issues.apache.org/jira/browse/AMBARI-11435
> 
> 
> Repository: ambari
> 
> 
> Description
> -------
> 
> Parth Brahmbhatt mentioned that in case of HA topology.min.replication.count 
> should be set to > 1. Ideally it should be floor(n/2 +1).
> 
> 
> Diffs
> -----
> 
>   
> ambari-server/src/main/resources/common-services/STORM/0.9.1.2.1/package/scripts/params_linux.py
>  68f2d0e 
>   
> ambari-server/src/main/resources/common-services/STORM/0.9.1.2.1/package/scripts/params_windows.py
>  4853762 
> 
> Diff: https://reviews.apache.org/r/35351/diff/
> 
> 
> Testing
> -------
> 
> ----------------------------------------------------------------------
> Ran 266 tests in 8.558s
> 
> OK
> ----------------------------------------------------------------------
> Total run:785
> Total errors:0
> Total failures:0
> OK
> 
> 
> Thanks,
> 
> Dmitro Lisnichenko
> 
>

Reply via email to