bugCERTI - Bugs: bug #23071, joinFederationExecution with empty...

 
 

bug #23071: joinFederationExecution with empty federate name hangs

Submitted by:  Petr Gotthard <gotthardp>
Submitted on:  Sat 26 Apr 2008 04:46:50 PM UTC  
 
Category: CERTISeverity: 3 - Normal
Item Group: Software errorStatus: Fixed
Privacy: PublicAssigned to: Eric NOULARD <erk>
Open/Closed: ClosedRelease: CVS
Reproducibility: Every TimeFixed Release: 3.3.0

Add a New Comment(Rich Markup)
   

You are not logged in

Please log in, so followups can be emailed to you.

 

Tue 29 Apr 2008 06:04:32 PM UTC, comment #4:

Hi Petr,

I switched it to Fixed.
Then I usually put it to closed as soons as I
know which [numbered] CERTI release
will contain the fix (Fixed Release).

This way it is easy to track which release contains which fixes
using a report form:
https://savannah.nongnu.org/bugs/?func=digest&group=certi

Thank your for your test.

Eric NOULARD <erk>
Project AdministratorIn charge of this item.
Tue 29 Apr 2008 04:56:09 PM UTC, comment #3:

I've rechecked this error.

With the latest CVS update the joinFederationexecution checks for an empty name and returns error. I guess this is correct so you can mark this error as "closed" with no further action.

Petr Gotthard <gotthardp>
Project Member
Sun 27 Apr 2008 07:09:50 PM UTC, comment #2:

It was working before the merge. The bug was found during regression tests of my federate.

Did you try removing the first joinAndLeave? In my test environment is also hangs up without it, so I think it doesn't come from the join-leave-join sequence.

The test application I sent you was meant as a modest contribution to your regression test suite, so please feel free to use it under any license you like.

Petr Gotthard <gotthardp>
Project Member
Sun 27 Apr 2008 12:03:41 PM UTC, comment #1:

OK Thank you for the testcase.
It seems that the trouble comes from the sequence
create
join
resign
create
join
resign

and not from the empty name,

I will track this issue, but I know that JYR did some
parallel work on create/join/resign/destroy sequence
while I was working on heterogeneity.

I may have broken this OR this wasn't working
even before the merge (not tested yet).

I intend to setup a kind of "TestSuite" in order
to track more easily regression issue and may be
HLA conformance too.

Do you mind if use your example in this testsuite and
put it under GPL?

Eric NOULARD <erk>
Project AdministratorIn charge of this item.
Sat 26 Apr 2008 04:46:50 PM UTC, original submission:

Hello,
thanks for the heterogeneous federation support. I found a first bug:

The latest CERTI changed its behaviour: when a federate with empty name ("") joins an execution, the joinFederationexecution never returns.
Attached is a test program to reproduce this bug.

Petr Gotthard <gotthardp>
Project Member

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach File(s):
   
   
Comment:
   

Attached Files
file #15531:  test01.cc added by gotthardp (1KiB - application/octet-stream - Test program to reproduce the bug.)

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by rousse (Updated the item)
  • -unavailable- added by erk (Posted a comment)
  • -unavailable- added by erk (May be related to create/join/resign/destroy)
  • -unavailable- added by gotthardp (Submitted the item)
  •  

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 9 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Wed 02 Jul 2008 08:59:03 AM UTCerkOpen/ClosedOpen=>Closed
      Fixed ReleaseCVS=>3.3.0
    Tue 29 Apr 2008 06:04:32 PM UTCerkStatusWorks For Me=>Fixed
      Fixed ReleaseUnknown=>CVS
    Tue 29 Apr 2008 01:22:38 PM UTCrousseStatusConfirmed=>Works For Me
    Sun 27 Apr 2008 12:03:41 PM UTCerkStatusNone=>Confirmed
      Assigned toNone=>erk
      Carbon-Copy-=>Added rousse
    Sat 26 Apr 2008 04:46:50 PM UTCgotthardpAttached File-=>Added test01.cc, #15531

    Back to the top


    Powered by Savane 3.1-cleanup1