Issue Details (XML | Word | Printable)

Key: DERBY-303
Type: Bug Bug
Status: Closed Closed
Resolution: Fixed
Priority: Major Major
Assignee: David Van Couvering
Reporter: Øystein Grøvlen
Votes: 0
Watchers: 0
Operations

If you were logged in you would be able to see more operations.
Derby

derbynet/testconnection.java fails

Created: 20/May/05 06:52 PM   Updated: 21/May/05 08:53 AM
Return to search
Component/s: Test
Affects Version/s: 10.1.1.0
Fix Version/s: None

Time Tracking:
Not Specified

File Attachments:
  Size
File Licensed for inclusion in ASF works DERBY-303.diff 2005-05-21 06:23 AM David Van Couvering 0.6 kB
Environment: Seem to occur on several platforms

Resolution Date: 21/May/05 07:45 AM


 Description  « Hide
When I run the testsuite, derbynet/testconnection.java fails.
testconnection.tmp contains the following:

Testing testconnection
org.apache.derby.drda.NetworkServerControl ping
Could not connect to Derby Network Server on host localhost, port 1527.
java.lang.IllegalThreadStateException: process hasn't exited
        at java.lang.UNIXProcess.exitValue(UNIXProcess.java:116)
        at org.apache.derbyTesting.functionTests.tests.derbynet.testconnection.execCmdDumpResults(Unknown Source)
        at org.apache.derbyTesting.functionTests.tests.derbynet.testconnection.execCmdDumpResults(Unknown Source)
        at org.apache.derbyTesting.functionTests.tests.derbynet.testconnection.main(Unknown Source)

 All   Comments   Work Log   Change History   Subversion Commits      Sort Order: Ascending order - Click to sort in descending order
Øystein Grøvlen added a comment - 20/May/05 06:55 PM
This seems to be a regression from DERBY-214.
The call to exitValue that fails was added in DERBY-214.

It may seem like the preceeding calls to ProcessStreamResult.Wait() does not guarantee that the process has terminated.

David Van Couvering added a comment - 21/May/05 01:59 AM
This definitely worked in my sandbox when I tested it; I will work on this ASAP.

David Van Couvering added a comment - 21/May/05 06:23 AM
Looking at the very helpful test results from Ole, it appears this bug only occurs on Solaris 10x86.
I can't reproduce on any platforms I have available. However, a code inspection shows that
I am missing a call to Process.waitFor() The attached patch adds this call; Oystein, can you
see if this fixes it for you on your Solaris 10 x86 box?

Andrew McIntyre added a comment - 21/May/05 07:00 AM
This is also failing on a Linux-x86 box that I have. I will also test this patch and provide feedback.

Andrew McIntyre added a comment - 21/May/05 07:45 AM
Resolved issue on my Linux box, committed patch with revision 171164. Will close issue after waiting for Ole's results.

Andrew McIntyre added a comment - 21/May/05 08:53 AM
Confirmed as fixed by Øystein.