ClearCase plug-in crash

I'm desparate to get the ClearCase plugin working on my machine (but having used ClearCase for a couple of years now, I'm getting used to being desparate!).

The plugin crashes in the initialise method (native).

I looked through the source on SourceForge, and figured a good basic test is to fire up the net.sourceforge.eclipseccase.jni.Clearcase.Main method, which should initialise the native dll. This is what I got:

Q:\tools\idea\idea664\jre\bin>java -cp ..\..\plugins\transparent.jar net.sourceforge.eclipseccase.jni.Clearcase

An unexpected exception has been detected in native code outside the VM.
Unexpected Signal : EXCEPTION_ACCESS_VIOLATION occurred at PC=0x1000F7E0
Function=[Unknown.]
Library=Q:\tools\idea\idea664\plugins\ccjni.dll

NOTE: We are unable to locate the function name symbol for the error
just occurred. Please refer to release documentation for possible
reason and solutions.


Current Java thread:
at net.sourceforge.eclipseccase.jni.Clearcase.initialize(Native Method)
at net.sourceforge.eclipseccase.jni.Clearcase.]]>(Clearcase.java:128)

Dynamic libraries:
0x00400000 - 0x00406000 Q:\tools\idea\idea664\jre\bin\java.exe
0x77F60000 - 0x77FBE000 C:\WINNT\System32\ntdll.dll
0x77DC0000 - 0x77DFF000 C:\WINNT\system32\ADVAPI32.dll
0x77F00000 - 0x77F5E000 C:\WINNT\system32\KERNEL32.dll
0x77E70000 - 0x77EC5000 C:\WINNT\system32\USER32.dll
0x77ED0000 - 0x77EFC000 C:\WINNT\system32\GDI32.dll
0x77E10000 - 0x77E67000 C:\WINNT\system32\RPCRT4.dll
0x78000000 - 0x78040000 C:\WINNT\system32\MSVCRT.dll
0x6D330000 - 0x6D45A000 Q:\tools\idea\idea664\jre\bin\client\jvm.dll
0x77FD0000 - 0x77FFA000 C:\WINNT\System32\WINMM.dll
0x6BC00000 - 0x6BC10000 C:\WINNT\System32\SMNT40.dll
0x6D1D0000 - 0x6D1D7000 Q:\tools\idea\idea664\jre\bin\hpi.dll
0x6D300000 - 0x6D30D000 Q:\tools\idea\idea664\jre\bin\verify.dll
0x6D210000 - 0x6D229000 Q:\tools\idea\idea664\jre\bin\java.dll
0x6D320000 - 0x6D32D000 Q:\tools\idea\idea664\jre\bin\zip.dll
0x10000000 - 0x10030000 Q:\tools\idea\idea664\plugins\ccjni.dll
0x77B20000 - 0x77BD7000 C:\WINNT\system32\ole32.dll
0x65340000 - 0x653DB000 C:\WINNT\system32\OLEAUT32.dll
0x0B2B0000 - 0x0B2D1000 C:\WINNT\System32\Hook95.dll
0x76AC0000 - 0x76ADD000 C:\WINNT\System32\imagehlp.dll
0x731B0000 - 0x731BA000 C:\WINNT\System32\PSAPI.DLL

Local Time = Fri Oct 25 14:44:40 2002
Elapsed Time = 0
#

  1. The exception above was detected in native code outside the VM

#

  1. Java VM: Java HotSpot(TM) Client VM (1.4.1_01-b01 mixed mode)

#

  1. An error report file has been saved as hs_err_pid344.log.

  2. Please refer to the file for further information.

#

I looked at the source for the ccjni.dll (which is the Eclipse plugin hosted at SourceForge too, right?), and the initialise method looks like it traps all exceptions, and passes them back with a message - but no messages match what I see. So, maybe the java/jni/COM framework is screwed up on my machine (NT4 sp6, IE6).

I'm way out of my depth here, and would appreciate a few pointers from anyone who knows. Can I setup a debugger to watch the java/jni/COM interfaces (sounds tough to me!). Any clues at all, because otherwise I have to manage the vcs manually using ClearCase...

thanks for any help,
Reg Skelton

4 comments
Comment actions Permalink

Reg,

Did you try to use the CommandLine implementation. Right now it is dead slow
but help is on the way. The owner of eclispeccase, Matthew Conway and I are
in the process of merging our java clearcase API in a separate sourceforge
project so we can reuse it across IDE. He has already created a pipelining
approach for the CommandLine that reduces significantly the latency on
cleartool command invocation (basically there is a long live cleartool
process that is fed the commands).
The other thing I am working on is a full Clearcase Automation Library port
to java. This will change the entry points and, I am not too hopeful, might
work around your problem.

Other avenue of investigation:
1) I have seen that netscape and twiki sometimes do not agree when
downloading files. Make sure you use IE or use the dll that you downloaded
from sourceforge.
2) Reinstall clearcase? I know it is a pain but I know for a fact (I even
put it in the documentation) that if you do not have a proper installation
(registry settings) it will crash.

Bottom line is if you do not find a solution quickly I will just work right
away on porting Matthew's pipe approach to transparent right away.

Let me know.

Jacques
"Reg Skelton" <itnadmin@jetbrains.com> wrote in message
news:6310453.1035554326344.JavaMail.jrun@is.intellij.net...

I'm desparate to get the ClearCase plugin working on my machine (but

having used ClearCase for a couple of years now, I'm getting used to being
desparate!).
>

The plugin crashes in the initialise method (native).

>

I looked through the source on SourceForge, and figured a good basic test

is to fire up the net.sourceforge.eclipseccase.jni.Clearcase.Main method,
which should initialise the native dll. This is what I got:
>

Q:\tools\idea\idea664\jre\bin>java -cp ..\..\plugins\transparent.jar

net.sourceforge.eclipseccase.jni.Clearcase
>

An unexpected exception has been detected in native code outside the VM.
Unexpected Signal : EXCEPTION_ACCESS_VIOLATION occurred at PC=0x1000F7E0
Function=[Unknown.]
Library=Q:\tools\idea\idea664\plugins\ccjni.dll

>

NOTE: We are unable to locate the function name symbol for the error
just occurred. Please refer to release documentation for possible
reason and solutions.

>
>

Current Java thread:
at net.sourceforge.eclipseccase.jni.Clearcase.initialize(Native

Method)

at

net.sourceforge.eclipseccase.jni.Clearcase.(Clearcase.java:128) >]]>

Dynamic libraries:
0x00400000 - 0x00406000 Q:\tools\idea\idea664\jre\bin\java.exe
0x77F60000 - 0x77FBE000 C:\WINNT\System32\ntdll.dll
0x77DC0000 - 0x77DFF000 C:\WINNT\system32\ADVAPI32.dll
0x77F00000 - 0x77F5E000 C:\WINNT\system32\KERNEL32.dll
0x77E70000 - 0x77EC5000 C:\WINNT\system32\USER32.dll
0x77ED0000 - 0x77EFC000 C:\WINNT\system32\GDI32.dll
0x77E10000 - 0x77E67000 C:\WINNT\system32\RPCRT4.dll
0x78000000 - 0x78040000 C:\WINNT\system32\MSVCRT.dll
0x6D330000 - 0x6D45A000

Q:\tools\idea\idea664\jre\bin\client\jvm.dll

0x77FD0000 - 0x77FFA000 C:\WINNT\System32\WINMM.dll
0x6BC00000 - 0x6BC10000 C:\WINNT\System32\SMNT40.dll
0x6D1D0000 - 0x6D1D7000 Q:\tools\idea\idea664\jre\bin\hpi.dll
0x6D300000 - 0x6D30D000 Q:\tools\idea\idea664\jre\bin\verify.dll
0x6D210000 - 0x6D229000 Q:\tools\idea\idea664\jre\bin\java.dll
0x6D320000 - 0x6D32D000 Q:\tools\idea\idea664\jre\bin\zip.dll
0x10000000 - 0x10030000 Q:\tools\idea\idea664\plugins\ccjni.dll
0x77B20000 - 0x77BD7000 C:\WINNT\system32\ole32.dll
0x65340000 - 0x653DB000 C:\WINNT\system32\OLEAUT32.dll
0x0B2B0000 - 0x0B2D1000 C:\WINNT\System32\Hook95.dll
0x76AC0000 - 0x76ADD000 C:\WINNT\System32\imagehlp.dll
0x731B0000 - 0x731BA000 C:\WINNT\System32\PSAPI.DLL

>

Local Time = Fri Oct 25 14:44:40 2002
Elapsed Time = 0
#

  1. The exception above was detected in native code outside the VM

#

  1. Java VM: Java HotSpot(TM) Client VM (1.4.1_01-b01 mixed mode)

#

  1. An error report file has been saved as hs_err_pid344.log.

  2. Please refer to the file for further information.

#

>

I looked at the source for the ccjni.dll (which is the Eclipse plugin

hosted at SourceForge too, right?), and the initialise method looks like it
traps all exceptions, and passes them back with a message - but no messages
match what I see. So, maybe the java/jni/COM framework is screwed up on my
machine (NT4 sp6, IE6).
>

I'm way out of my depth here, and would appreciate a few pointers from

anyone who knows. Can I setup a debugger to watch the java/jni/COM
interfaces (sounds tough to me!). Any clues at all, because otherwise I have
to manage the vcs manually using ClearCase...
>

thanks for any help,
Reg Skelton



0
Comment actions Permalink

Swift reply, Jacques! Thanks...

I can't make the command-line implementation use unreserved checkouts, no matter what I set in the project settings\VCS dialog - it always does Reserved checkouts, which would make me pretty unpopular! Maybe this too is a symptom of the underlying problem...

I've checked the ccjni.dll version, and it looks ok (192512 bytes)

I can't re-install ClearCase. We're not really supposed to do anything with our corporate desktops - running Idea is really pushing it!

The version of ClearCase we have to use is pretty old - v3.2.1. Thinking about it, I bet that's the problem. Maybe that version isn't COM'd...

The future avenues look very interesting. I don't know much about Automation and so on, but I think there's a product called JIntegra which can auto-generate java stubbs for COM objects (I think). You've probably already seen it.

Any further help on getting unreserved check-outs would be gratefully received.

thanks again,
Reg Skelton

0
Comment actions Permalink

Swift reply, Jacques! Thanks...

No problemo.

I can't make the command-line implementation use unreserved checkouts, no

matter what I set in the project settings\VCS dialog - it always does
Reserved checkouts, which would make me pretty unpopular! Maybe this too is
a symptom of the underlying problem...
I will look into it.

>

I've checked the ccjni.dll version, and it looks ok (192512 bytes)

>

I can't re-install ClearCase. We're not really supposed to do anything

with our corporate desktops - running Idea is really pushing it!
I wonder if all the corporation that use Clearcase have also their desktops
configuration locked? It would be an interesting study of which one of the 2
directives came first ;)

>

The version of ClearCase we have to use is pretty old - v3.2.1. Thinking

about it, I bet that's the problem. Maybe that version isn't COM'd...
Bet you found your problem. Can you send me your ccauto.dll in your
rational/clearcase/bin directory. I will look for any mismatch.

The future avenues look very interesting. I don't know much about

Automation and so on, but I think there's a product called JIntegra which
can auto-generate java stubbs for COM objects (I think). You've probably
already seen it.
No I did not find it. I am using jawin

Any further help on getting unreserved check-outs would be gratefully

received.
>

thanks again,
Reg Skelton

>


0
Comment actions Permalink

Thanks for your reply, Jacques.

I've attached the ccauto.dll, but a bad sign is that it comes from atria\bin\ - so I guess this was one of the first versions after Rational bought ClearCase from Atria, which was a very long time ago...

As for locked desktops, I'm pretty sure no-one here would use ClearCase if they had the choice of any other system.

btw, JIntegra lives at http://www.linar.com/, but jawin seems to do everything needed, and is open-source.

regards,
Reg Skelton



Attachment(s):
ccauto.dll
0

Please sign in to leave a comment.