> On April 4, 2016, 8:35 p.m., Jaimin Jetly wrote:
> > This patch doesn't address newly installed cluster case but does not 
> > address upgrade scenario i.e introducing new 
> > cluster-env/alerts_repeat_tolerance property to the existing cluster on 
> > upgrade with stack default value.
> > If there is not task tracking this work please create one.
> 
> Jaimin Jetly wrote:
>     Typo in the above comment.
>     This patch *do* address newly installed cluster case but does not address 
> upgrade scenario

Actually, BE should populate the initial value for newly installed or upgraded 
cluster. FE just handle the case that backend returns nothing, which should not 
really happen.


- Zhe (Joe)


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/45692/#review126935
-----------------------------------------------------------


On April 4, 2016, 5:58 p.m., Zhe (Joe) Wang wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/45692/
> -----------------------------------------------------------
> 
> (Updated April 4, 2016, 5:58 p.m.)
> 
> 
> Review request for Ambari, Alexandr Antonenko, Jaimin Jetly, Jonathan Hurley, 
> Oleg Nechiporenko, Richard Zang, Srimanth Gunturi, Xi Wang, and Yusaku Sako.
> 
> 
> Bugs: AMBARI-15603
>     https://issues.apache.org/jira/browse/AMBARI-15603
> 
> 
> Repository: ambari
> 
> 
> Description
> -------
> 
> The global repeat tolerance value for all alert definitions is set on the 
> cluster-env configuration. Unless an alert definition overrides this value, 
> it will be used for any definition in the system. By default, this value will 
> be set to 1, indicating that there is no tolerance and the alert state should 
> be taken at face value.
> 
> GET api/v1/clusters/<cluster>/configurations?type=cluster-env&tag=<tag>
> 
>       "Config": {
>         "cluster_name": "c1",
>         "stack_id": "HDP-2.4"
>       },
>       "properties": {
>         "command_retry_enabled": "true",
>         "command_retry_max_time_in_sec": "600",
>         ...
>         "alerts_repeat_tolerance" : "1"
>        ...
>       }
> 
> The web client should expose a way to update the {{cluster-env}} to set this 
> value. 
> 
> *UI Warning of Delayed Alerts*
> When changing the value of the global of definition-specific repeat 
> tolerance, a warning should be presented to the user to indicate that it will 
> now take longer for the alert notifications to be sent. This is because 
> notifications are delayed until the interval multiplied by the repeat 
> tolerance is reached. Consider the case where the check against an alert 
> happens every 5 minutes and the repeat tolerance is set to 5. It will be at 
> least 25 minutes before any outbound notifications are dispatched. This 
> warning can be on a per-alert definition basis as well as when setting the 
> global value.
> 
> 
> Diffs
> -----
> 
>   
> ambari-web/app/controllers/main/alerts/alert_definitions_actions_controller.js
>  dc9f78c 
>   ambari-web/app/messages.js 1b2a02f 
>   ambari-web/app/styles/application.less a6b79ce 
>   ambari-web/app/templates/common/modal_popups/prompt_popup.hbs 078cc65 
> 
> Diff: https://reviews.apache.org/r/45692/diff/
> 
> 
> Testing
> -------
> 
> Local ambari-web test passed.
> 25600 tests complete (23 seconds)
> 154 tests pending
> Manual testing done.
> 
> 
> Thanks,
> 
> Zhe (Joe) Wang
> 
>

Reply via email to