[omniORB] Abort in omniORB2.8.0

[Hinkofer, Fa. pds, E230, D]

Hello,

we are using omniORB2.8.0 on platform: Sparc/Solaris2.7 (SunOS KFPAH015 =
5.7
Generic_106541-12 sun4us sparc FJSV,GPUS), and Forte C++ Enterp Ed 6
compiler (CC: Sun WorkShop 6 2000/08/30 C++ 5.1 Patch 109490-01).

On June 06, 2001 I sent the following email:
...
>Subject: Informations on Installing, building, and using omniORB2.8.0 =
on
>Sparc, Solaris2.7, Forte C++ Enterp Ed 6 platform
>
>
>Hello,
>we have to use omniORB2.8.0 on platform: Sparc/Solaris2.7, and Forte =
C++
>Enterp Ed 6 compiler.
>Some Questions:
>1. Are there a port on this platform ?
>2. Are there restrictions ?
>3. Is it possible to perform program linkage with the delivered =
binaries
>(libraries) ore is it necessary to compile and link the =
omniORB-Sources ?
>
>Thank you very much for your efforts
>
> G=FCnter Hinkofer

On June 06, 2001 Sai-Lai Lo immediatly sent me back the following =
email:
...
>Subject: [omniORB] Re: Informations on Installing, building, and using
>omniORB2.8.0 on Sparc, Solaris2.7, Forte C++ Enterp Ed 6 platform
>
>G=FCnter,
>
>Some months ago we found got the workaround to make omniORB compiled =
with
>Forte 6 compiler to stop crashing when multiple threads raise =
exceptions at
>the same time. Your email prompted me to check if the fix was =
backported to
>omniORB 2.8.0. It hasn't but I've now checkin the fix. You should get =
the
>update-to-date source of 2.8.0 from here:
>
>ftp://ftp.uk.research.att.com/pub/omniORB/omniORB_28_snapshots/omniORB-=
late
s
>t.tar.gz
>
>Just choose the sun4_sosV_5.7 platform and recompile everything.
>
>You must not use the binaries you downloaded from our site for Solaris =
2.5.
>The compilers are not compatible.
>
>You should also update your Forte 6 with all the latest patches. Trust =
me,
>that would save you a lot of pain.
>
>Sai-Lai

We are using the recommended snapshot of omniORB2.8.0. But now we have =
a
problem with our application and omniORB2.8.0. Our application requests =
and
receives tiff-images via omniORB2.8.0. After processing a tiff-image =
our
application immediatly sends back the results of processing via
omniORB2.8.0. In a range of  100.000 to 600.000 processed images we get =
an
abort in omniORB2.8.0. Look to the following stack-dump (created by =
dbx):
(/opt/SUNWspro/bin/../WS6/bin/sparcv9/dbx) where -h
current thread: t@4
  [1] __sigprocmask(0x0, 0xfea0b468, 0x0, 0x0, 0x0, 0x0), at 0xfeef9b1c
  [2] _resetsig(0xfeefc468, 0x0, 0x0, 0xfea0bd78, 0xfef0e000, 0x0), at
0xfeeee54c
  [3] _sigon(0xfea0bd78, 0xfef159a8, 0x6, 0xfea0b53c, 0xfea0bd78,
0xff3307b0), at 0xfeeedc3c
  [4] _thrp_kill(0x0, 0x4, 0x6, 0xfef0e000, 0x4, 0xfeebc440), at =
0xfeef0db0
  [5] raise(0x6, 0x0, 0x0, 0xffffffff, 0x16ad28, 0xfef54b68), at =
0xfee4a028
  [6] abort(0xfeeb8000, 0xfef6a844, 0xfef6a054, 0x2, 0x1bf54, =
0xfef54b68),
at 0xfee35594
=3D>[7] omni_abort() (optimized), at 0xff2b1374 (line ~248) in =
"corbaOrb.cc"
  [8] __Cimpl::ex_terminate(0x0, 0x0, 0xfea0bd18, 0x0, 0xfea0bd18,
0x12c2dc), at 0xfef57240
  [9] __Crun::ex_throw(0x0, 0x0, 0xfef6a880, 0xfef6a054, 0x0, =
0xfef6a8c8),
at 0xfef54eac
  [10] CORBA::Object::_duplicate(obj =3D ???) (optimized), at =
0xff2af304 (line
~127) in "corbaObject.cc"
  [11] SarImm::Imm_Helper::duplicate(p =3D 0x1a4aac), line 689 in
"oif_sar_immSK.cc"
=20
[12]_CORBA_ObjRef_Var<SarImm::Imm,SarImm::Imm_Helper>::_CORBA_ObjRef_Var=
(thi
s =3D 0xfea0b9f4, p =3D CLASS), line 359 in "templatedecls.h"
  [13] Sar_i::getMyImm(), line 81 in "sar_i.h"
  [14] FsmIc::a_put_decode_result(p_this =3D 0x159554, p_pb =3D =
0xfea0bbac),
line 12535 in "fsm_ic.cpp"
  [15] Fsm::control(this =3D 0x159554, p_descr =3D 0x1595c0, p_pb =3D =
0xfea0bbac),
line 276 in "fsm.cpp"
  [16] FsmDescr::scheduler(this =3D 0x159594, p_pb =3D 0xfea0bc2c), =
line 559 in
"fsm.cpp"
  [17] SoiServer::processing(this =3D 0xfea0bca4), line 187 in =
"soi.cpp"
  [18] Soi::server(p_pb =3D 0x16ad28), line 139 in "soi.cpp"
  [19] omni_thread_wrapper(ptr =3D ???) (optimized), at 0xff2625e8 =
(line ~406)
in "posix.cc"
(/opt/SUNWspro/bin/../WS6/bin/sparcv9/dbx)=20

Questions:
Is there still a problem in the recommended snapshot of omniORB2.8.0 ? =
Is
there an newer snapshot of omniORB2.8.0 than we have ? Are we using the
right patch of C++ Compiler (look above) ? Do you think there is a =
problem
in the application ? Or do you have an other idea what's the problem ?

Best regards,

G=FCnter

=20

___________________________________________________________________
Deutsche Post AG         Fon:   +49 211 163 1485
SNL IT-Systeme BRIEF     Mobil:=20
AE Produktion Brief 1    Fax:   +49 211 163 1783
                        =20
Briefanschrift:          Postfach 10 12 65, 40003 D=FCsseldorf
Hausanschrift:           Konrad-Adenauer-Platz 1, 40210 D=FCsseldorf