OmniORB crash in OpenVMS

We have an application that uses CORBA for Alpha to Windows data
transfers. Our normal process is to start the OmniORB 3.0.0 naming
service on the Alpha server and everything works fine from there. That
is until the current installation. The Omninames program crashes with
an access violation that eventually results in a register dump. The
output is below. One difference between this system and our other
installs, is that while the other installs are Alpha DS10 systems
running 7.3-1 out of the box, this is an Alpha 800 that was upgraded
from 7.1-1H1 to 7.3-1 (including the latest patch sets). C and C++
were also upgraded. We have a version of the binaries built for OVMS
7.1 that we use as a distribution (via zip file). This is the
mechanism that has worked fine for several systems. In the event that
there were issues in the libraries or linkage, I started a rebuild of
the source (using Python-1_5_2 and MMK as the MMS substitution). This
caused the C++ compiler to crash while compiling IDLEXPR.CC. Not being
a CORBA expert or OmniORB developer, I don't have any idea were to
look. Anyone have a suggestion?
Thanks,
Edward Heller
TransCore, ITS

OMNINAMES CRASH:
$omninames -start
%CXXL-F-TERMINATE, terminate() or unexpected() called
%TRACE-F-TRACEBACK, symbolic stack dump follows
image module routine line rel PC
abs PC
OMNINAMES 0 00000000000FDBF8
000000000010DBF8
OMNINAMES 0 00000000000FFB94
000000000010FB94
----- above condition handler called with exception 05F7841C:
%CXXL-F-RETHROW, Exception rethrown at PC = 000BD440
----- end of exception message
0 FFFFFFFF800B7CBC
FFFFFFFF800B7CBC
OMNINAMES 0 00000000000FD088
000000000010D088
OMNINAMES 0 00000000000FD778
000000000010D778
OMNINAMES 0 00000000000AD440
00000000000BD440
OMNINAMES 0 00000000000DBF80
00000000000EBF80
OMNINAMES 0 00000000000DC160
00000000000EC160
OMNINAMES 0 00000000000BF0E0
00000000000CF0E0
OMNINAMES 0 0000000000070DE8
0000000000080DE8
OMNINAMES 0 0000000000070070
0000000000080070
OMNINAMES 0 00000000001055E8
00000000001155E8
PTHREAD$RTL 0 000000000003F0D0
000000007BD0F0D0
PTHREAD$RTL 0 000000000001C31C
000000007BCEC31C
0 FFFFFFFF802794D4
FFFFFFFF802794D4
omniORB: Assertion failed -- mutex destroyed whilst held.
This is a bug in omniORB. Please submit a report (with stack
trace if possible) to <omniorb@uk.research.att.com>.
%SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual
address=000000000000
0000, PC=000000000009A09C, PS=0000001B
%TRACE-F-TRACEBACK, symbolic stack dump follows
image module routine line rel PC
abs PC
OMNINAMES 0 000000000008A09C
000000000009A09C
OMNINAMES 0 00000000000D4618
00000000000E4618
0 FFFFFFFF8007E444
FFFFFFFF8007E444
PTHREAD$RTL 0 0000000000053DE8
000000007BD23DE8
PTHREAD$RTL 0 000000000003F0F4
000000007BD0F0F4
PTHREAD$RTL 0 000000000002CFD0
000000007BCFCFD0
0 0000000000000000
0000000000000000
PTHREAD$RTL ?
?
0 FFFFFFFF802794D4
FFFFFFFF802794D4

COMPILER CRASH:

$ CXX /Mms=file=.d/Except/Diag/Float=G_Float/NoDebug/Quiet
OMNIORB300ROOT:[000
000.SRC.TOOL.OMNIIDL.CXX]IDLEXPR.CC /object=IDLEXPR.OBJ
/define=("IDLMODULE_VERS
ION=""0x2301""","OMNIIDL_EXECUTABLE","YY_ALWAYS_IN TERACTIVE","USE_IDLERROR")-
/include=("./","/omni_include/","/python_include")-
/repository=([.CXX_REPOSITORY],[lib.OpenVMS_Alpha_7_3])-
-

%SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual
address=000000000000
0004, PC=000000000042D008, PS=0000001B
%TRACE-F-TRACEBACK, symbolic stack dump follows
image module routine line rel PC
abs PC
CXX$COMPILER ME_EXPR me_do_argument 139577 0000000000005C68
000000000042D008
CXX$COMPILER ME_EXPR me_generate_call
139263 0000000000004E20
000000000042C1C0
CXX$COMPILER ME_EXPR walk_operation_expression
141696 000000000000A844
0000000000431BE4
CXX$COMPILER ME_EXPR walk_expression
140533 00000000000086B0
000000000042FA50
CXX$COMPILER ME_STMT walk_if_statement
136843 000000000000080C
0000000000441ABC
CXX$COMPILER ME_STMT walk_statement 140688 00000000000043C0
0000000000445670
CXX$COMPILER ME_STMT walk_statement_list
136785 0000000000000664
0000000000441914
CXX$COMPILER ME_STMT walk_block_statement
139265 00000000000027EC
0000000000443A9C
CXX$COMPILER ME_STMT walk_statement 140726 00000000000049A8
0000000000445C58
CXX$COMPILER ME_FUNC walk_routine 144135 000000000000285C
0000000000434FCC
CXX$COMPILER ME walk_scope_routines 139830 00000000000011DC
000000000041956C
CXX$COMPILER ME me_driver 136270 00000000000007B4
0000000000418B44
CXX$COMPILER COMPILE dec_exx_back_end_one_o_file
170688 0000000000001AD8
00000000003F0DB8
CXX$COMPILER COMPILE dec_exx_back_end
169923 0000000000001314
00000000003F05F4
CXX$COMPILER CFE edg_main 55960 0000000000000214
000000000044F464
CXX$COMPILER COMPILE gem_xx_compile 171001 0000000000002214
00000000003F14F4
CXX$COMPILER GEM_CP_VMS GEM_CP_MAIN 2603 00000000000018CC
0000000000684CAC
CXX$COMPILER 0 0000000000984434
0000000000994434
0 FFFFFFFF802794D4
FFFFFFFF802794D4
%MMK-F-ERRUPD, error status %X1000000C occurred when updating target
IDLEXPR.OBJ
%MMK-F-ERRUPD, error status %X1C14803C occurred when updating target
EXPORT
%MMK-F-ERRUPD, error status %X1C14803C occurred when updating target
EXPORT
%MMK-F-ERRUPD, error status %X1C14803C occurred when updating target
EXPORT