AW: big problem with release version 1.0

From: Michael Schuldt (m..scensys.de)
Date: Fri Oct 03 2003 - 15:29:02 EDT

  • Next message: Andrus Adamchik: "Re: AW: big problem with release version 1.0"

    Hi Andrus,

    haven't tried to save the Datamap-file with the new modeler. Will try this
    first, so I can tell you about any Warnings or Errors.

    We checked further details of our application an recognized further errors.
    Because of the deadline I'd like to use the old Alpha-Version. There's only
    one known error to us. But I will keep on upgrading to the release version
    of cayenne asap.

    BTW: I told you in my first Mail about the error occurred in two threads
    after several time. Here it is again:

    Every thread causes this error after several hours:
    org.objectstyle.cayenne.CayenneRuntimeException: Global exception.
            at
    org.objectstyle.cayenne.access.util.SelectObserver.nextGlobalException(Selec
    tObserver.java:156)
            at
    org.objectstyle.cayenne.access.DataNode.performQueries(DataNode.java:332)
            at
    org.objectstyle.cayenne.access.DataDomain.performQueries(DataDomain.java:427
    )
            at
    org.objectstyle.cayenne.access.DataContext.performQueries(DataContext.java:1
    020)
            at
    org.objectstyle.cayenne.access.DataContext.performQuery(DataContext.java:111
    2)
            at
    org.objectstyle.cayenne.access.DataContext.performQuery(DataContext.java:949
    )
            at de.best4rest.boerse.DBCleanerThread.run(DBCleanerThread.java:125)
            at java.lang.Thread.run(Thread.java:536)
    Caused by: org.objectstyle.cayenne.CayenneRuntimeException: Query exception.
            at
    org.objectstyle.cayenne.access.util.SelectObserver.nextQueryException(Select
    Observer.java:147)
            at
    org.objectstyle.cayenne.access.DataNode.performQueries(DataNode.java:288)
            ... 6 more
    Caused by: java.lang.NullPointerException

    Do you have any suggestions? I found some hints in the mailing list about
    reconnect. I tried everything. I also can't understand, how there should be
    an idle connection. The threads are accessing the database through cayenne
    every minute. Is there another possibility for the errors?

    I'm using mysql 4.0 and tomcat 4.0.

    Thanks again

    Michael

    ascensys
    Software & Consulting
    Schuldt und Spajic GbR

    Individualsoftware . Sicherheitskonzepte . Internetlosungen . Schulungen
    individual software . security concepts . internet solutions . trainings

    Leidhecke 24
    D-51570 Windeck
    Germany

    fon: +49 (0) 2292 6801-47
    fax: +49 (0) 2292 6801-48

    internet: http://www.ascensys.de
    e-mail: inf..scensys.de

    Confidentiality Notice: This e-mail is privileged and confidential and for
    the use of the addressee only. Should you have received this e-mail in error
    please notify us by replying directly to the sender or by sending a message
    to the sender and delete the e-mail. Unauthorised use, disclosure or copying
    of the contents of this e-mail, or any similar action, is prohibited. Thank
    you.

    -----Ursprungliche Nachricht-----
    Von: Andrus Adamchik [mailto:andru..bjectstyle.org]
    Gesendet: Freitag, 3. Oktober 2003 21:02
    An: cayenne-use..bjectstyle.org
    Cc: m..scensys.de
    Betreff: Re: big problem with release version 1.0

    Ok, not so fast.... While I still see a theoretical possibility of such
    failure, I couldn't recreate a condition when this can happen....

    I guess if you could send me your datamap file privately, I can take
    another look. (BTW, are there any warnings when you save your datamap
    with the new version of the modeler?)

    If this doesn't help, I might simply implement the patch I planned and
    give it to you for testing...though I would still like to know the real
    reason of failure.

    Andrus

    On Friday, October 3, 2003, at 12:25 PM, Andrus Adamchik wrote:

    > Michael,
    >
    > Browsing through the code, now I can see this can possibly happen if
    > there
    > is a new (uncommitted) object being the target of to-one relationship
    > of
    > the object being fetched... This looks like your case.
    >
    > I will be patching 1.0 and make it available later today.
    >
    > If the patched version works for you, next week we will make it
    > official
    > via a 1.0.1 maintenance release.
    >
    > Andrus



    This archive was generated by hypermail 2.0.0 : Fri Oct 03 2003 - 15:29:04 EDT