[ 
https://issues.apache.org/jira/browse/DERBY-5850?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13805449#comment-13805449
 ] 

Mike Matrigali commented on DERBY-5850:
---------------------------------------

looking closer the stacks and the test it looks like we are in rollforward 
recovery, and it may be the case
that a StandardException is expected in this test case.    The stack clearly 
show we are in redo recovery.
And the PageBasicOperation.findpage() has code to catch StandardException's 
thrown by calls to
BaseContainerHandle.getAnyPage() and the following code in that routine looks 
like it probably handles
the exception if we had not gotten an ArrayIndexOutOfBoundsException() while 
building what I now
think was an "expected" exception in this test case.

{noformat}
        // maybe we are in rollforward recovery and this is an init page operati
on,
        // give subclass a chance to create the page
        if (page == null && getPageException != null)
        {
            //if are rolloforward recovery reload the page using load tran metho
ds
            //that initialize the page. because in rollforward recovery, we
            //might be actually recreating the page container did not exist
            //in the backup when we started the rollforward recovery.

            if (rtran.inRollForwardRecovery())
            {
                if (SanityManager.DEBUG)
                    if(SanityManager.DEBUG_ON("LoadTran"))
                        SanityManager.DEBUG_PRINT(
                            "Trace", "got null page " + pageId +
                            " and getPageException, attempt last ditch effort");

                page = getPageForRedoRecovery(xact);

                if (SanityManager.DEBUG)
                    if(SanityManager.DEBUG_ON("LoadTran"))
                        SanityManager.DEBUG_PRINT(
                            "Trace"," getPageForRedoRecovery, got page=" +
                            (page != null));
            }
        }
{noformat}

So likely only bug here is the JVM bug being tracked by DERBY-6393.



> store/rollForwardRecovery fails with java.lang.ArrayIndexOutOfBoundsException
> -----------------------------------------------------------------------------
>
>                 Key: DERBY-5850
>                 URL: https://issues.apache.org/jira/browse/DERBY-5850
>             Project: Derby
>          Issue Type: Bug
>          Components: Test
>    Affects Versions: 10.10.1.1, 10.11.0.0
>         Environment: ava Version:    1.7.0
> Java Vendor:     IBM Corporation
> Java home:       /local1/cloudtst/dev/src/ibm17/jre
> OS name:         Linux
> OS architecture: x86
> OS version:      2.6.32-220.7.1.el6.i686
> java.specification.name: Java Platform API Specification
> java.specification.version: 1.7
> java.runtime.version: pxi3270sr1-20120330_01 (SR1)
> java.fullversion: JRE 1.7.0 IBM J9 2.6 Linux x86-32 20120322_106209 (JIT 
> enabled, AOT enabled)
> J9VM - R26_Java726_SR1_20120322_1720_B106209
> JIT  - r11_20120322_22976
> GC   - R26_Java726_SR1_20120322_1720_B106209
> J9CL - 20120322_106209
>            Reporter: Kathey Marsden
>         Attachments: derby.log
>
>
> On trunk IBM 1.7 run on July 5 rolForwardRecovery failed with the following 
> diff ...
> Failure Details:
> ********* Diff file derbyall/storeall/storemore/rollForwardRecovery.diff
> *** Start: rollForwardRecovery jdk1.7.0 storeall:storemore 2012-07-06 
> 00:33:18 ***
> 730a731,732
> > ERROR XJ040: Failed to start database 'wombat' with class loader XXXX,  see 
> > the next exception for details.
> > ERROR XJ001: Java exception: ': java.lang.ArrayIndexOutOfBoundsException'.
> 734 del
> < A                                                                           
>                                                     |B                        
>                                                                               
>                          |C                                                   
>                                                                             
> |D                                                                            
>                                                    |E                         
>                                                                               
>                         
> 735 del
> < 
> --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> 736 del
> < 1                                                                           
>                                                    &|2                        
>                                                                               
>                         &|3                                                   
>                                                                            
> &|4                                                                           
>                                                    &|5                        
>                                                                               
>                         &
> 737 del
> < 10                                                                          
>                                                    &|20                       
>                                                                               
>                         &|30                                                  
>                                                                            
> &|40                                                                          
>                                                    &|50                       
>                                                                               
>                         &
> 738 del
> ....
> Unfortunately things didn't get pushed out to Apache. I will follow up with 
> the stack trace



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Reply via email to