omniORB 2.8.0 bugs

The following bugs in omniORB 2.8.0 have been fixed. Update from Subversion to get the fixes.


Summary: System clock change causes the scavenger thread to consume lots of CPU cycles (bug number 13)
Date: Tue May 29 12:14:54 BST 2001
Reported by: Peter Rönnquist
Description: Scavenger now get the real time after each scan. This is to cope with the system clock set backward by a large amount. Seems to happen a lot to some system..

Summary: Hanging during boa->destroy() (bug number 12)
Date: Wed Nov 29 14:29:25 GMT 2000
Reported by: Tom Gerke
Description: I have a number of co-located CORBA servers which my application is using. I call boa->impl_shutdown() and then boa->destroy(). This boa->destroy() call hangs up for about 3 minutes before exiting. No exception iws thrown, and when it comes out, my application proceeds to exit normally. Remark: this fix is back ported from omniORB 3.0.

Summary: Race condition in object deletion (bug number 11)
Date: Fri Nov 3 10:28:33 GMT 2000
Reported by: Dmitry Dolinsky
Description: In a heavily concurrent server, deletion of a CORBA object may trigger a race condition that results in invalid object reference exception being thrown by the ORB.

Summary: Cookie was not passed to global exception handlers (bug number 10)
Date: Wed Jun 7 16:33:11 BST 2000
Reported by: Duncan Grisby

Summary: Fixed interoperability problem with Naming service IORs (bug number 9)
Date: Tue Mar 7 11:27:46 GMT 2000
Reported by: Duncan Grisby
Description: omniORB does not allow you to specify a naming service IOR which is derived from CosNaming::NamingContext in its configuration file.

Summary: create_union_tc failed if the discriminator type was an alias (bug number 8)
Date: Tue Feb 15 11:01:09 GMT 2000
Reported by: David Riddoch

Summary: Server and client call timeouts are now disabled by default (bug number 7)
Date: Fri Jan 7 14:47:08 GMT 2000
Reported by: David Riddoch

Summary: Name lookup error in inherited interfaces (bug number 6)
Date: Fri Dec 17 17:58:57 GMT 1999
Reported by: Francis Michel

Summary: Applied performance improvement for Any (bug number 5)
Date: Fri Dec 17 13:27:16 GMT 1999
Reported by: David Riddoch

Summary: Incorrect typecode for recursive sequence of objref (bug number 4)
Date: Tue Dec 7 18:08:27 GMT 1999
Reported by: David Evers

Summary: Bug in marshalling/unmarshalling of arrays (bug number 3)
Date: Thu Dec 2 19:01:41 GMT 1999
Reported by: David Riddoch

Summary: Assertion failure when the ORB cannot spawn a thread to serve a new connection. (bug number 2)
Date: Wed Oct 27 19:06:02 BST 1999
Reported by: Sai-Lai Lo

Summary: DynAny no longer do alias expand on the typecode (bug number 1)
Date: Tue Oct 26 20:34:57 BST 1999
Reported by: Sai-Lai Lo
Description: In other words, all aliases are preserved in the typecode.