[
https://issues.apache.org/jira/browse/DERBY-6307?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13748315#comment-13748315
]
Myrna van Lunteren commented on DERBY-6307:
-------------------------------------------
Thanks Bryan, for spelling it out in detail...
The javadump method allows a programmer with access to ibm jvms to request a
javacore dump file at any practical spot.
Yes, adding such a fixture to that test passes, I tried with ibm 16, and jdk
prerelease 1.8.
However, I have the following concerns:
- with IBM jvms (I believe > 1.5) the javadump method generates a file,
something like javacore.20130822.123456.1234.0001.txt. I'm not entirely sure of
all the pieces, but the first is the date, and the last is a sequence number.
This file should get deleted after the test is done, for it might seem like
something bad happened
- with IBM jvms, it also prints out a message to the console, if my test dir
is: c:\myrna\d6164tst it is:
"JVMDUMP034I User requested Java dump using
'C:\myrna\javacore.20130822.144351.7688.0001.txt' through
com.ibm.jvm.Dump.JavaDump
JVMDUMP010I Java dump written to
C:\myrna\d6164tst\javacore.20130822.144351.7688.0001.txt"
I think this is two lines.
I know some automatic checking scripts look for strings like this to decide on
failure, so these lines should get caught and perhaps even be checked.
I've been trying to mimic what's done in the SysinfoCPCheckPathTest fixture
testClassPathChecker but so far it's not working. I thought perhaps checking on
System.out was the wrong thing, and I should be redirecting System.err instead,
but so far, it still prints out to the console, and my stream is null.
I'll give it another go tomorrow, if I cannot figure it out, I'll post my
attempt, perhaps someone else can see where I go wrong.
> Improve the code coverage of org.apache.derby.iapi.services.info.JVMInfo
> ------------------------------------------------------------------------
>
> Key: DERBY-6307
> URL: https://issues.apache.org/jira/browse/DERBY-6307
> Project: Derby
> Issue Type: Sub-task
> Components: Test
> Reporter: ahsan shamsudeen
> Assignee: ahsan shamsudeen
> Priority: Minor
> Attachments: DERBY-6307.patch, DERBY-6307Temp.patch
>
>
> According to the coverage reports,
> org.apache.derby.iapi.services.info.JVMInfo is poorly exercised by the Derby
> regression tests.
> This sub-task is to improve that test coverage.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
