On 10/7/18 7:16 PM, Bryan Pendleton wrote:
I'm not sure what to do about this, either.

Perhaps the simplest thing, for now, is to see if we can track which Jenkins slaves we are successful on, and which ones we fail on.

Maybe you can just make a tiny spreadsheet to track that?

And then we can see about changing our job definitions so that they only run on the "happy" machines, and avoid the "unhappy" ones.
It seems to me that sometimes the tests succeed on a given machine and sometimes they fail on the same machine. But systematically tracking this sounds like a good idea.

Thanks,
-Rick

It's not nice, but it keeps us productive without beating our heads against a wall...

bryan


On Sun, Oct 7, 2018 at 11:49 AM Rick Hillegas <[email protected] <mailto:[email protected]>> wrote:

    The Infrastructure group has no theories about our Jenkins test
    failures and they won't let us log onto the machines to debug the
    problem ourselves.



    -------- Forwarded Message --------
    Subject:    [jira] [Resolved] (INFRA-16759) Need instructions on how
    to log onto a Jenkins machine
    Date:       Sat, 6 Oct 2018 23:32:00 +0000 (UTC)
    From:       Chris Lambertus (JIRA) <[email protected]>
    <mailto:[email protected]>
    To:         [email protected] <mailto:[email protected]>



          
[https://issues.apache.org/jira/browse/INFRA-16759?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
  ]

    Chris Lambertus resolved INFRA-16759.
    -------------------------------------
         Resolution: Unresolved

    no feedback

    > Need instructions on how to log onto a Jenkins machine
    > ------------------------------------------------------
    >
    >                 Key: INFRA-16759
    >                 URL: https://issues.apache.org/jira/browse/INFRA-16759
    >             Project: Infrastructure
    >          Issue Type: Task
    >          Components: Jenkins
    >            Reporter: Rick Hillegas
    >            Priority: Major
    >
    > Checkins to the Derby codeline kick off continuous integration build/test 
cycles via Jenkins. The builds complete successfully but the tests raise errors 
frequently even though the same tests pass successfully on developers' personal 
machines. To debug this problem, we would like to be able to run the failing tests 
and other experiments on the Jenkins machines where the failures occur. Is it 
possible to ssh and sftp to one of these failing environments?
    > Thanks,
    > -Rick



    --
    This message was sent by Atlassian JIRA
    (v7.6.3#76005)


Reply via email to