Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 103601

Summary: Running TPTP Manual Test hangs, Manual test case runtime window unable to display
Product: z_Archived Reporter: Albert Manlangit <almanlan>
Component: TPTPAssignee: Samson Wai <samwai>
Status: CLOSED FIXED QA Contact:
Severity: blocker    
Priority: P3 CC: almanlan, amehrega, ewchan, jptoomey, samwai, srinivas.p.doddapaneni, suwanda, svaswani
Version: unspecified   
Target Milestone: ---   
Hardware: PC   
OS: Windows XP   
Whiteboard: closed460
Bug Depends on:    
Bug Blocks: 85841    
Attachments:
Description Flags
Manual tester gui
none
Full log from rac config
none
Debug hclaunch.dll output
none
Debug hclaunch.dll output
none
Core dump from Java VM
none
Screenshot using hclaunch.dll debug version (StartProcess0) none

Description Albert Manlangit CLA 2005-07-12 22:04:32 EDT
When you run the TPTP Manual Test, it hangs.  The "Component Test- Manual 
Testcase Remote Application" windows doesn't come up.  It appears a problem
occur during the execution run.  My Agent controller (RAC) is running fine.
Not sure what's causing this problem.  Everything is setup correctly on my
Windows XP machine.  Two of my colleagues had the same problem as I did.  
We are all using windows XP with service pack 1.

From the log file:

C:\Documents and Settings\alm\workspace\.metadata\.log

!SESSION 2005-07-12 13:28:12.468 
eclipse.buildId=I20050627-1435
java.version=1.4.2_03
java.vendor=Sun Microsystems Inc.
BootLoader constants: OS=win32, ARCH=x86, WS=win32, NL=en_US
Command-line arguments:  -os win32 -ws win32 -arch x86

!ENTRY org.eclipse.core.runtime 2 2 2005-07-12 13:36:14.187
!MESSAGE Job found still running after platform shutdown.  Jobs should be 
canceled by the plugin that scheduled them during shutdown: 
org.eclipse.hyades.execution.harness.TestExecutionHarness$2

!ENTRY org.eclipse.core.runtime 2 2 2005-07-12 13:36:14.187
!MESSAGE Job found still running after platform shutdown.  Jobs should be 
canceled by the plugin that scheduled them during shutdown: 
org.eclipse.debug.internal.ui.DebugUIPlugin$6

How to Reproduce:
Used component downloaded from www.eclipse.org/tptp/ build 
TPTP-4.0.0-200507120100

eclipse-SDK-3.1-win32
emf-sdo-SDK-2.1.0
xsd-SDK-2.1.0
tptp.runtime-TPTP-4.0.0-200507120100
tptpdc.win_ia32-TPTP-4.0.0-200507120100 - this the RAC

After extracting all the components to the appropriate directory:
set all the necessary path
launch the Agent Controller
start eclipse
create new project
go to new:other:test:TPTP Manual Test
go to the test cases tab and add test cases
on behavior tab, click on add to select invocation to run the test case
click Run, to run manual tester, you will see that it doesn't bring up the
manual test case runtime window
Comment 1 Albert Manlangit CLA 2005-07-12 22:28:59 EDT
Created attachment 24650 [details]
Manual tester gui

This is the screen display I save after executing the run for the TPTP
Manual Test. As you can see, it doesn't display the Manual TestCase window
because it never did display it after running the test.  I assume there was
an error during the run.
Comment 2 Albert Manlangit CLA 2005-07-12 22:37:37 EDT
I also got these information from the service log on 
C:\new.manual.tester\rac\config\servicelog.log, I just attached part of it.

<CommonBaseEvent creationTime="2005-07-12T20:28:07.859000Z" 
globalInstanceId="A7004CF2102D4060AC253145BC7B97EE" msg="Service starting" 
severity="10" version="1.0.1">
	<sourceComponentId component="Data Collection Engine" 
componentIdType="Application" location="9.52.76.68" locationType="IPV4" 
processId="248" subComponent="..\RAServerConfig\RAServerConfigBinding2.cpp" 
componentType="Eclipse_TPTP"/>
	<situation categoryName="ReportSituation">
		<situationType xmlns:xsi="http://www.w3.org/2001/XMLSchema-
instance" xsi:type="ReportSituation" reasoningScope="INTERNAL" 
reportCategory="LOG"/>
	</situation>
</CommonBaseEvent>
<CommonBaseEvent creationTime="2005-07-12T20:28:08.187000Z" 
globalInstanceId="A752EB58B1FD40FABB6D57A499A69D3D" msg="Start processing 
plugin: org.apache.jakarta_commons_logging_1.0.3" severity="10" 
version="1.0.1">
	<sourceComponentId component="Data Collection Engine" 
componentIdType="Application" location="9.52.76.68" locationType="IPV4" 
processId="248" subComponent="..\RAServerConfig\RAServerConfigBinding2.cpp" 
componentType="Eclipse_TPTP"/>
	<situation categoryName="ReportSituation">
		<situationType xmlns:xsi="http://www.w3.org/2001/XMLSchema-
instance" xsi:type="ReportSituation" reasoningScope="INTERNAL" 
reportCategory="LOG"/>
	</situation>
Comment 3 Joe Toomey CLA 2005-07-13 09:35:39 EDT
Can you please attach the entire log?  The entries you included here are not 
meaningful to this problem, but other entries in the log may be.

I hope this is some configuration problem, as I have not seen this reported by 
the other people testing.  Scott, can you please take a look?
Comment 4 Albert Manlangit CLA 2005-07-13 12:30:59 EDT
Created attachment 24677 [details]
Full log from rac config 

This is the full log from the C:\new.manual.tester\rac\config\servicelog.log.
There are only two logs that I know of, rac\config\servicelog.log and 
C:\Documents and Settings\alm\workspace\.metadata\.log.  The entire
.metadata\.log was already submitted on this defect.
Comment 5 Scott E. Schneider CLA 2005-07-14 11:44:34 EDT
I will attempt to reproduce on my machine -- more status will be known soon.
Comment 6 Joe Toomey CLA 2005-07-14 12:39:00 EDT
I'd like to see if this problem presents itself in older RAC builds.  Scott is 
going to investigate this.
Comment 7 Scott E. Schneider CLA 2005-07-14 13:43:19 EDT
Joe and I were able to verify that the regression is between RAC 6/21 and RAC
7/6 -- more info forthcoming.
Comment 8 Scott E. Schneider CLA 2005-07-14 14:14:43 EDT
This is also seen with any base TPTP test type -- on initial debug it appears to
crash on the sending of the launch message to RAC.  The runner process is never
in a state where it can be debugged (debug flags on this process have no effect)
-- it crashes immediately it seems.
Comment 9 Scott E. Schneider CLA 2005-07-14 16:37:17 EDT
Test execution no longer works (at least on the Windows platform) -- it appears
to crash with the startProcess0 method is called in the native hclaunch.dll
code.  I'm attaching some additional output from a debug version of hclaunch.dll
that Samson gave me on 7/14.

Re-assigning to RAC team, I double-checked everything on the test side that I
could think of and it appears to not be a test code issue.
Comment 10 Scott E. Schneider CLA 2005-07-14 16:38:18 EDT
Created attachment 24790 [details]
Debug hclaunch.dll output
Comment 11 Scott E. Schneider CLA 2005-07-14 16:38:46 EDT
Created attachment 24791 [details]
Debug hclaunch.dll output
Comment 12 Scott E. Schneider CLA 2005-07-14 16:40:00 EDT
Created attachment 24792 [details]
Core dump from Java VM
Comment 13 Scott E. Schneider CLA 2005-07-14 16:46:15 EDT
Created attachment 24794 [details]
Screenshot using hclaunch.dll debug version (StartProcess0)
Comment 14 Scott E. Schneider CLA 2005-07-18 14:53:54 EDT
*** Bug 85841 has been marked as a duplicate of this bug. ***
Comment 15 amehrega CLA 2005-07-18 15:28:37 EDT
*** Bug 104222 has been marked as a duplicate of this bug. ***
Comment 16 Hendra Suwanda CLA 2005-07-19 11:04:22 EDT
Transfering to Samson.
Fix checked in 07/18.
Samson, please list the files affected. Thanks.
Comment 17 Samson Wai CLA 2005-07-19 11:17:11 EDT
Files affected:
Under "platform/org.eclipse.tptp.platform.agentcontroller/src-native"
- ./collection/framework/HCLauncher/hclaunch.c
- ./collection/framework/HCLauncher/launcher_md.c
- ./collection/framework/RACommon/RAComm.c
Comment 18 Samson Wai CLA 2005-07-26 15:06:02 EDT
The reason for using ra_free() in hclaunch.dll is because the memory
block was allocated using ra_malloc(), which is residing under another moduke
hcclco.dll. Freeing a block of memory allocated in another module will
cause Windows to throw an exception occasionally. Any piece of memory
allocated using ra_malloc() should be freed using ra_free(). And any
local malloc() should use a local free(). The reason why Windows throws an
exception is still unknown, but that's the symptom we've seen.
Comment 19 Paul Slauenwhite CLA 2009-06-30 07:53:03 EDT
As of TPTP 4.6.0, TPTP is in maintenance mode and focusing on improving quality by resolving relevant enhancements/defects and increasing test coverage through test creation, automation, Build Verification Tests (BVTs), and expanded run-time execution. As part of the TPTP Bugzilla housecleaning process (see http://wiki.eclipse.org/Bugzilla_Housecleaning_Processes), this enhancement/defect is verified/closed by the Project Lead since this enhancement/defect has been resolved and unverified for more than 1 year and considered to be fixed. If this enhancement/defect is still unresolved and reproducible in the latest TPTP release (http://www.eclipse.org/tptp/home/downloads/), please re-open.