bugCERTI - Bugs: bug #23135, RTIG/RTIA generates no coredump...

 
 

bug #23135: RTIG/RTIA generates no coredump after unhandled exception

Submitted by:  Petr Gotthard <gotthardp>
Submitted on:  Fri 02 May 2008 07:31:09 AM UTC  
 
Category: CERTISeverity: 1 - Wish
Item Group: Software errorStatus: Fixed
Privacy: PublicAssigned to: Eric NOULARD <erk>
Open/Closed: ClosedRelease: CVS
Reproducibility: NoneFixed Release: 3.4.2

Add a New Comment(Rich Markup)
   

You are not logged in

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

 

Sat 03 May 2008 06:34:32 PM UTC, comment #4:

RTIG fix done, checked-in CVS.
RTIA will come soon.

Eric NOULARD <erk>
Project AdministratorIn charge of this item.
Sat 03 May 2008 04:24:05 PM UTC, comment #3:

Thank you Eric,
that's exactly what I was thinking of.

Petr

Petr Gotthard <gotthardp>
Project Member
Sat 03 May 2008 10:44:32 AM UTC, comment #2:

Hi Petr,

I'm basically OK with your point of view.
I've just attached a patch concerning RTIG.
Is my solution OK for you?

If yes I'll check this in.
If not would you propose another patch?

Concerning RTIA I'll check but I think
we should try/catch around the constructor call
since it may fails if RTIG is not reachable
and we should report a meaning full error for this.

during execute I think you are right
we should either try/catch inside execute or die.

(file #15594)

Eric NOULARD <erk>
Project AdministratorIn charge of this item.
Sat 03 May 2008 09:50:36 AM UTC, comment #1:

If you abort() or re-throw the exception in the top-most try-catch the core dump contains trace to this point, instead of pointing to the bug.
To trace the bug it's necessary to completely remove the top-most try-catch. This might not harm anything, I think exceptions should be used only to catch expected situations.

Petr Gotthard <gotthardp>
Project Member
Fri 02 May 2008 07:31:09 AM UTC, original submission:

Just a suggestion:
The topmost catch in RTIG:main and RTIA:main finishes with exit(EXIT_FAILURE). In case of an unexpected exception the RTIA/RTIG thus exists without generating a core dump.
Since a coredump might be useful for debugging purposes, I suggest to use abort() instead of exit(), or to re-throw the exception so the coredump can be generated.

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 #15594:  RTIG-bug23135-erk1.patch added by erk (1KiB - text/x-diff)

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by erk (Updated the item)
  • -unavailable- added by gotthardp (Submitted the item)
  •  

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

     

     

    Follow 7 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Fri 06 Dec 2013 12:27:33 PM UTCerkStatusReady For Test=>Fixed
      Open/ClosedOpen=>Closed
    Fri 27 Sep 2013 08:20:30 AM UTCerkStatusIn Progress=>Ready For Test
      Fixed ReleaseUnknown=>3.4.2
    Sat 03 May 2008 10:44:32 AM UTCerkStatusNone=>In Progress
      Assigned toNone=>erk
    Sat 03 May 2008 10:44:31 AM UTCerkAttached File-=>Added RTIG-bug23135-erk1.patch, #15594

    Back to the top


    Powered by Savane 3.1-cleanup1