[omniORB] segmentation fault , context already bound but can't access it?

[bitostef@katamail.com]

Hi Bjorn
I must bother you once again . I have some news about that segfault.
It’s quite strange , it appears that the first time the server runs it creates my own context (“test”,”my_context”) ; when client calls the server everithing is accomplished correcly, 
once the server has send the message and  the client has received and printed it the server has a segfault . The next time I try to start the server it gives immediately the segfault . I ‘ve done a bit of debugging on the server code :
If the context name is changed (i.e. test2 , my_context2) happens as above 
The  exception  happens here:
try:
    testContext = rootContext.bind_new_context(name) #already bound
    print "New test context bound"
    
except CosNaming.NamingContext.AlreadyBound, ex: #happens something wrong
    print "Test context already exists"#this isn’t printed
    obj = rootContext.resolve(name )
    testContext = obj._narrow(CosNaming.NamingContext)
    if testContext is None:
        print "test.mycontext exists but is not a NamingContext"
        sys.exit(1)

The first time the server runs  it gives this messages:
…
receive codeset service context …
dispatching remote call ‘echostring’ to :root<0> (active)
creating new python state for thread id 4101 
       echostring called with message :hello from python
sendChunk..
throw giopstream:: CommFailure from giopStream.cc:812  
      (0,NO,COMM_FAILURE_UnMarshallArguments)
Segmentation Fault (here exit)

The next time is thrown a python user exception:
Cosnaming.NamingContext.AlreadyBound

Thank you for attention 
Regards 
Stefano Bitossi


__________________________________________


Fai i tuoi acquisti su www.kwshopping.it