Mailing-List: contact dev-help@ambari.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@ambari.apache.org
Content-Type: multipart/alternative;
 boundary="===============6249971159747577214=="
MIME-Version: 1.0
Subject: Re: Review Request 38492: RU - Skipping failed task caused remaining
 pending tasks to be ABORTED
From: "Nate Cole" <ncole@hortonworks.com>
To: "Alejandro Fernandez" <afernandez@hortonworks.com>,
 "Nate Cole" <ncole@hortonworks.com>
Cc: "Ambari" <dev@ambari.apache.org>,
 "Jonathan Hurley" <jhurley@hortonworks.com>
Date: Fri, 18 Sep 2015 17:54:19 -0000
Message-ID: <20150918175419.9550.83853@reviews.apache.org>
Auto-Submitted: auto-generated
Sender: "Nate Cole" <noreply@reviews.apache.org>
References: <20150918161727.3773.25610@reviews.apache.org>
In-Reply-To: <20150918161727.3773.25610@reviews.apache.org>
Reply-To: "Nate Cole" <ncole@hortonworks.com>

--===============6249971159747577214==
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/38492/#review99571
-----------------------------------------------------------


ambari-server/src/main/java/org/apache/ambari/server/actionmanager/ActionScheduler.java (lines 1023 - 1025)
<https://reviews.apache.org/r/38492/#comment156497>

    Nice!


- Nate Cole


On Sept. 18, 2015, 12:17 p.m., Jonathan Hurley wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/38492/
> -----------------------------------------------------------
> 
> (Updated Sept. 18, 2015, 12:17 p.m.)
> 
> 
> Review request for Ambari, Alejandro Fernandez and Nate Cole.
> 
> 
> Bugs: AMBARI-13145
>     https://issues.apache.org/jira/browse/AMBARI-13145
> 
> 
> Repository: ambari
> 
> 
> Description
> -------
> 
> Aborting a failed task during an upgrade causes the entire upgrade request to become ABORTED. The ActionScheduler has logic which will abort an entire request if the command's success factor was not met. This logic also needs to take into account skippable stages which will be marked as COMPLETED even with failed tasks.
> 
> 
> Diffs
> -----
> 
>   ambari-server/src/main/java/org/apache/ambari/server/actionmanager/ActionScheduler.java 7d93638 
>   ambari-server/src/test/java/org/apache/ambari/server/actionmanager/TestActionScheduler.java 31356bb 
> 
> Diff: https://reviews.apache.org/r/38492/diff/
> 
> 
> Testing
> -------
> 
> Instrumented my environment so that every single stage failed in either one of the following two ways:
> - A timeout of the task by Ambari Server (placing into HOLDING_TIMEDOUT and then eventually TIMEDOUT
> - A timeout of the python executor (placing into HOLDING_FAILED) and then eventually FAILED.
> 
> mvn clean test
> 
> Tests run: 3186, Failures: 0, Errors: 0, Skipped: 25
> 
> [INFO] ------------------------------------------------------------------------
> [INFO] BUILD SUCCESS
> [INFO] ------------------------------------------------------------------------
> [INFO] Total time: 23:04 min
> [INFO] Finished at: 2015-09-18T11:06:17-04:00
> [INFO] Final Memory: 50M/1347M
> [INFO] ------------------------------------------------------------------------
> 
> 
> Thanks,
> 
> Jonathan Hurley
> 
>


--===============6249971159747577214==--