reef-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mariia Mykhailova <mamyk...@microsoft.com>
Subject RE: Pull request blocked by failing tests from REEF-1473
Date Fri, 30 Sep 2016 18:28:39 GMT
True, whenever we do a change that touches both Java and C# code, or the bridge, we have to
clean everything before build (that’s one of the reasons why I like AppVeyor – it always
starts from a clean build). I don’t think we have a habit of sending explicit warnings about
such changes to the dev list, we just kind of know to clean everything in case of weird test
failures.

-Mariia

From: Douglas Service [mailto:dsopsrc@gmail.com]
Sent: Thursday, September 29, 2016 5:10 PM
To: dev@reef.apache.org
Subject: Re: Pull request blocked by failing tests from REEF-1473

I found the issue. REEF-1611 moves the .proto files on the java side to a central location
and both the Java and C# builds must be cleaned before taking the change; otherwise, test
failures occur on the C# side. If this change is accepted, we should send email to the dev
list with a warning about cleaning both Java and C#.

Doug

On Thu, Sep 29, 2016 at 1:43 PM, Douglas Service <dsopsrc@gmail.com<mailto:dsopsrc@gmail.com>>
wrote:
Hello Mariia,

The failures were in visual studio. I would have sent the errors last night but I could not
find any way to export the results from visual studio. An png is attached that shows some
of the errors. I also included the output from git log.  Do you know how to export test results
from VisualStudio? I will try a build and test from the command line and if  that succeeds,
I will go ahead with the pull request.

Doug

commit 75f78ac8dfbcfa76bf4bde82948761d0fd1372db
Author: Mariia Mykhailova <mariia@apache.org<mailto:mariia@apache.org>>
Date:   Wed Sep 28 15:20:02 2016 -0700

    [REEF-1473] Add check for strStatus=null to TestBridgeClient

    This change adds check for strStatus being null in TestBridgeClient.
    It doesn't fix the root cause of failures, which can't be reproduced now,
    but slightly improves troubleshooting in case the failures repeat.

    JIRA:
      [REEF-1473](https://issues.apache.org/jira/browse/REEF-1473<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fissues.apache.org%2Fjira%2Fbrowse%2FREEF-1473&data=01%7C01%7Cmamykhai%40microsoft.com%7C4f7662ee3d3d4bdbc07a08d3e8c62623%7C72f988bf86f141af91ab2d7cd011db47%7C1&sdata=aDDyIchIaJrug7DBor4fJFuEaFibyCKO%2BMbtsY7iT90%3D&reserved=0>)

    Pull request:
      This closes #1139

commit 6ef63c19633554b3889c14aa203b8c628df4538e
Author: Mariia Mykhailova <mariia@apache.org<mailto:mariia@apache.org>>
Date:   Wed Sep 28 12:30:19 2016 -0700

    [REEF-1618] Add stack trace to exceptions handled in TaskHostBase

    This change adds full exception info to log printed in
    TaskHostBase.HandleException.

    JIRA:
      [REEF-1618](https://issues.apache.org/jira/browse/REEF-1618<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fissues.apache.org%2Fjira%2Fbrowse%2FREEF-1618&data=01%7C01%7Cmamykhai%40microsoft.com%7C4f7662ee3d3d4bdbc07a08d3e8c62623%7C72f988bf86f141af91ab2d7cd011db47%7C1&sdata=AWnTOiWOH2ISncS4Xi8HnZhDNQNIr9Lswvun4g39juk%3D&reserved=0>)

    Pull request:
      This closes #1138

commit a2278b4db74f6b60b41f1b4c1e848edaa3453de5
Author: Mariia Mykhailova <mariia@apache.org<mailto:mariia@apache.org>>
Date:   Tue Sep 27 17:36:13 2016 -0700

    [REEF-1623] Remove double-logging exception in Logger.Log(Exception)

    This change
     * removes logging of exception.Message and exception.StackTrace
       which are logged as part of exception anyways.
     * cleans up logic of when message is logged.

    JIRA:
      [REEF-1623](https://issues.apache.org/jira/browse/REEF-1623<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fissues.apache.org%2Fjira%2Fbrowse%2FREEF-1623&data=01%7C01%7Cmamykhai%40microsoft.com%7C4f7662ee3d3d4bdbc07a08d3e8c62623%7C72f988bf86f141af91ab2d7cd011db47%7C1&sdata=Ly%2BfxgLu%2BUrKxhjVih4smzHWl0%2FaCyCDAAvlwmFkfdE%3D&reserved=0>)

    Pull request:
      This closes #1137

commit 3ea89be97c20a440bba84c8c33338ee8e480875d
Author: Mariia Mykhailova <mariia@apache.org<mailto:mariia@apache.org>>
Date:   Mon Sep 26 11:53:29 2016 -0700

    [REEF-1451] Clean up IMRU Fault Tolerant scenario tests

    This change:
     * extends failures simulated with FailureType to include dispose.
     * adds tests for failure of mapper task on init and on dispose
       and failure of mapper evaluator on dispose.
     * cleans up checks and comments in other IMRU FT scenario tests.

    JIRA:
      [REEF-1451](https://issues.apache.org/jira/browse/REEF-1451<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fissues.apache.org%2Fjira%2Fbrowse%2FREEF-1451&data=01%7C01%7Cmamykhai%40microsoft.com%7C4f7662ee3d3d4bdbc07a08d3e8c62623%7C72f988bf86f141af91ab2d7cd011db47%7C1&sdata=imsPS6ZMhgbYMRfh6DluvKrMjINwYxKSQfNRnDtlzC8%3D&reserved=0>)

[Inline image 1]


On Thu, Sep 29, 2016 at 11:10 AM, Mariia Mykhailova <mamykhai@microsoft.com<mailto:mamykhai@microsoft.com>>
wrote:
Doug, you probably want to share the errors you're getting in the tests. I don't think REEF-1473
should cause any test failures, we've tested it locally, and it also passes AppVeyor build
(https://ci.appveyor.com/project/ApacheSoftwareFoundation/reef/build/1000-master<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fci.appveyor.com%2Fproject%2FApacheSoftwareFoundation%2Freef%2Fbuild%2F1000-master&data=01%7C01%7Cmamykhai%40microsoft.com%7C4f7662ee3d3d4bdbc07a08d3e8c62623%7C72f988bf86f141af91ab2d7cd011db47%7C1&sdata=ogT83uPmj37UAJ2Zu0WsUnggYDmowb7pfRcZA0XiNXk%3D&reserved=0>)

-Mariia

-----Original Message-----
From: Markus Weimer [mailto:markus@weimo.de<mailto:markus@weimo.de>]
Sent: Wednesday, September 28, 2016 10:28 PM
To: REEF Developers Mailinglist <dev@reef.apache.org<mailto:dev@reef.apache.org>>
Subject: Re: Pull request blocked by failing tests from REEF-1473

I have just tested both current `master` as well as your withdrawn PR #1140. Both pass tests
for me. Could it be that the issue is local on your machine?

Markus

On Wed, Sep 28, 2016 at 9:21 PM, Markus Weimer <markus@weimo.de<mailto:markus@weimo.de>>
wrote:
> That is odd, I tested that PR before merging it. What's the error
> message of the tests that failed?
>
> Markus
>
> On Wed, Sep 28, 2016 at 7:34 PM, Douglas Service <dsopsrc@gmail.com<mailto:dsopsrc@gmail.com>>
wrote:
>> I have a pull request ready to go for REEF-1611 but cannot push it do
>> a failing test from REEF-1473. I verified the problem on a clean
>> branch with a pull on upstream master.  A number of tests in
>> Org.Apache.REEF.Tests.Functional.Group are failing and it appears to
>> be due to changes in TestBridgeClient.cs.
>>
>> Doug


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message