Re: Re: Is there an ORBit2 port to SCO 5.0.5 ?
- From: "jean-philippe paillet" <jppaillet technodiva com>
- To: <orbit-list gnome org>
- Subject: Re: Re: Is there an ORBit2 port to SCO 5.0.5 ?
- Date: Fri, 17 Sep 2004 16:43:14 +0200
Hi,
I was not able to compile ORBit2 under my SCO server because the "configure"
failed :
>>
checking for glib-2.0 >= 2.0.0 gobject-2.0 >= 2.0.0 gthread-
2.0 >= 2.0.0 gmodule-2.0 >= 2.0.0... sh: gnome-config: not found
sh: gnome-config: not found
Package glib-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `glib-2.0.pc'
<<
Then, i tried to compile glib-2.4.6, with similar problems (libgettext
missing).
As it seems that it will take me a lot of time before the "configure" script
for ORBit2 is OK, i gave up.
If a binary package of ORBit2 for SCO exist, i would like to try it.
Regards,
____________________________________________________
Jean-Philippe.PAILLET
Chef de projet
127 Avenue Joseph Pierre Boitelet - BP 50
84 302 CAVAILLON CEDEX - FRANCE
Tel: +33 (0)4.90.76.08.08
Fax: +33 (0)4.90.78.15.84
E-mail: jppaillet technodiva com
Web : http://www.neopost.com/ls
This email and any files transmitted with it are confidential and intended
solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify the
system manager.
-----Message d'origine-----
De : orbit-list-bounces gnome org
[mailto:orbit-list-bounces gnome org]De la part de
orbit-list-request gnome org
Envoye : vendredi 17 septembre 2004 09:38
A : orbit-list gnome org
Objet : orbit-list Digest, Vol 5, Issue 3
Send orbit-list mailing list submissions to
orbit-list gnome org
To subscribe or unsubscribe via the World Wide Web, visit
http://mail.gnome.org/mailman/listinfo/orbit-list
or, via email, send a message with subject or body 'help' to
orbit-list-request gnome org
You can reach the person managing the list at
orbit-list-owner gnome org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of orbit-list digest..."
Today's Topics:
1. Re: Is there an ORBit2 port to SCO 5.0.5 ? (Magnus Bergman)
2. The minor of a CORBA_SystemException (Philip Van Hoof)
3. The minor of a CORBA_SystemException (Philip Van Hoof)
4. Passing arrays from ORBit2 to ORbit and from ORBit to ORBit2
(Philip Van Hoof)
5. Using existing/own network connections (Nikias)
6. Orbit in Win32 (Rodrigo Maldonado)
7. Using remote method-calls over a tcp/ip network (Philip Van Hoof)
8. Re: Using remote method-calls over a tcp/ip network
(Kuang-Chun Cheng)
9. Re: Using remote method-calls over a tcp/ip network
(Philip Van Hoof)
----------------------------------------------------------------------
Message: 1
Date: Fri, 3 Sep 2004 14:47:50 +0200
From: Magnus Bergman <magnus bergman observer net>
Subject: Re: Is there an ORBit2 port to SCO 5.0.5 ?
To: orbit-list gnome org
Message-ID: <20040903144750 56e5cd27 grommet observergroup com>
Content-Type: text/plain; charset=US-ASCII
On Thu, 2 Sep 2004 09:47:43 +0200
"jean-philippe paillet" <jppaillet technodiva com> wrote:
> Hello,
>
> I plan to implement a distributed application on Solaris,Linux and SCO
> servers thru ORBit2.
> Unfortunately, i can't switch from SCO to Linux because some specific
> communication cards and old applications are hosted on SCO servers.
> Is there any ORBit2 version for SCO ?
Are asking for some "official" binary package for SCO or are there
problems compiling ORBit2 for SCO?
------------------------------
Message: 2
Date: Thu, 09 Sep 2004 15:02:21 +0200
From: Philip Van Hoof <Philip VanHoof Cronos Be>
Subject: The minor of a CORBA_SystemException
To: orbit-list <orbit-list gnome org>
Message-ID: <1094734941 5433 20 camel localhost localdomain>
Content-Type: text/plain
Hi there,
While I was trying to figure out what the problem was with a specific
remote method-call that is always returning the CORBA Exception
COMM_FAILURE:1.0, I can across this page:
http://info.borland.com/techpubs/bes/v6/updates/except_cpp.html
It explains this about the exception:
[quote]
An existing connection may have closed due to failure at the other end
of the connection. A new connection request may have failed due to
resource limits on the client or server machine (the maximum number of
connections has been reached). When COMM_FAILURES occur due to system
exceptions, the system error number is set in the minor code of the
COMM_FAILURE. Check the minor code against the system-specific error
numbers (for example, in the include/sys/errno.h or msdev\include
\winerror.h files).
[/quote]
So I started trying to get that minor number out of the exception.
It looks like (after reading the ORBit2 sourcecode) the minor has been
set in an object of the type CORBA_SystemException in ev->_any._value
(in case ev is a pointer to my CORBA_Environment variable).
While reading the ORBit2 sourcecode, however, I came across the
following code:
void
CORBA_exception_set_system (CORBA_Environment *ev,
const CORBA_char *except_repos_id,
CORBA_completion_status completed)
{
CORBA_SystemException *se;
g_return_if_fail (ev != NULL);
se = CORBA_SystemException__alloc ();
/* I have never seen a case where 'minor' is actually necessary */
se->minor = 0 /* minor */;
se->completed = completed;
CORBA_exception_set (
ev, CORBA_SYSTEM_EXCEPTION, except_repos_id, se);
}
Reading that, it looks like the minor is always set to 0. Reading the
specification (I don't know whether or not the specification above is
correct) it should be something like se->minor = errno, no?
--
Philip Van Hoof, Software Developer @ Cronos
home: me at freax dot org
gnome: pvanhoof at gnome dot org
work: philip dot vanhoof at cronos dot be
junk: philip dot vanhoof at gmail dot com
http://www.freax.be, http://www.freax.eu.org
------------------------------
Message: 3
Date: Thu, 09 Sep 2004 15:18:59 +0200
From: Philip Van Hoof <spamfrommailing freax org>
Subject: The minor of a CORBA_SystemException
To: orbit-list <orbit-list gnome org>
Message-ID: <1094735939 5433 35 camel localhost localdomain>
Content-Type: text/plain
Hi there,
While I was trying to figure out what the problem was with a specific
remote method-call that is always returning the CORBA Exception
COMM_FAILURE:1.0, I can across this page:
http://info.borland.com/techpubs/bes/v6/updates/except_cpp.html
It explains this about the exception:
[quote]
An existing connection may have closed due to failure at the other end
of the connection. A new connection request may have failed due to
resource limits on the client or server machine (the maximum number of
connections has been reached). When COMM_FAILURES occur due to system
exceptions, the system error number is set in the minor code of the
COMM_FAILURE. Check the minor code against the system-specific error
numbers (for example, in the include/sys/errno.h or msdev\include
\winerror.h files).
[/quote]
So I started trying to get that minor number out of the exception.
It looks like (after reading the ORBit2 sourcecode) the minor has been
set in an object of the type CORBA_SystemException in ev->_any._value
(in case ev is a pointer to my CORBA_Environment variable).
While reading the ORBit2 sourcecode, however, I came across the
following code:
void
CORBA_exception_set_system (CORBA_Environment *ev,
const CORBA_char *except_repos_id,
CORBA_completion_status completed)
{
CORBA_SystemException *se;
g_return_if_fail (ev != NULL);
se = CORBA_SystemException__alloc ();
/* I have never seen a case where 'minor' is actually necessary
*/
se->minor = 0 /* minor */;
se->completed = completed;
CORBA_exception_set (
ev, CORBA_SYSTEM_EXCEPTION, except_repos_id, se);
}
Reading that, it looks like the minor is always set to 0. Reading the
specification (I don't know whether or not the specification above is
correct) it should be something like se->minor = errno, no?
--
Philip Van Hoof, Software Developer @ Cronos
home: me at freax dot org
gnome: pvanhoof at gnome dot org
work: philip dot vanhoof at cronos dot be
junk: philip dot vanhoof at gmail dot com
http://www.freax.be, http://www.freax.eu.org
------------------------------
Message: 4
Date: Fri, 10 Sep 2004 13:09:01 +0200
From: Philip Van Hoof <spamfrommailing freax org>
Subject: Passing arrays from ORBit2 to ORbit and from ORBit to ORBit2
To: ORBit list <orbit-list gnome org>
Message-ID: <1094814541 4896 12 camel localhost localdomain>
Content-Type: text/plain
Hi there,
I am trying to transfer an object that looks like this:
typedef struct SomeObject_Type SomeObject;
struct SomeObject_Type
{
CORBA_string name;
CORBA_short short1;
CORBA_short short2;
CORBA_long long1;
CORBA_short short3;
CORBA_long long2;
CORBA_long long3;
CORBA_unsigned_long arraylong1[6];
CORBA_unsigned_long arraylong2[6];
CORBA_unsigned_long arraylong3[6];
CORBA_unsigned_long arraylong4[6];
CORBA_long long4;
CORBA_long long5;
}
>From a ORBit1 (server) to a ORBit2 (client) by returning at a remote
function.
When we debug right before the "return" statement at the ORBit1
(server), all variables of the struct are set correctly.
However, at the ORBit2 (client) side, the fields long4 and long5 look
unitialized or have invalid values.
Is there an incompatibility with transferring arrays? I only have it
with objects that have arrays (like above).
--
Philip Van Hoof, Software Developer @ Cronos
home: me at freax dot org
gnome: pvanhoof at gnome dot org
work: philip dot vanhoof at cronos dot be
junk: philip dot vanhoof at gmail dot com
http://www.freax.be, http://www.freax.eu.org
------------------------------
Message: 5
Date: Mon, 13 Sep 2004 14:24:59 +0200
From: "Nikias" <nikias gmx li>
Subject: Using existing/own network connections
To: orbit-list gnome org
Message-ID: <20040913122446 0D2B33B1024 menubar gnome org>
Content-Type: text/plain; charset="iso-8859-1"
Hi all,
is it possible to use existing/own (self created) network connections
for ORBit2 ?
Greetings,
Nikias
------------------------------
Message: 6
Date: Thu, 16 Sep 2004 22:28:07 -0400
From: "Rodrigo Maldonado" <rodrigomaldonadoa vtr net>
Subject: Orbit in Win32
To: <orbit-list gnome org>
Message-ID: <002501c49c5d$f79bf020$b04978c8 rodrigo>
Content-Type: text/plain; charset="iso-8859-1"
Look .
I have a Middleware built in C in HP/UX and use SI/RPC how transport
mechanism. , the client app is in Visual Basic , the database is Oracle.8i
SI/RPC is a propietary protocol created for the company.
Is RPC and another functions , is in C.
The SI/RPC is obsolete and only works in HP/UX.
I need migrate the middleware to a Solaris and i need change the SI/RPC
protocol for another.
I think with my solution is ORBIT because is in C (the only ORB support C)
, and i dont need rewrite all code of the middleware. But i have a
question, with SI/RPC
i have a DLL (sirpc.dll), then i was call the functions of the DLL in
Visual Basic. and i can communicate with the middleware.
Is there any form to compile Orbit -win32 to DLL
and call functions to middleware when i finish the proccess of change code?
Thank you very much
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Rodrigo Maldonado.vcf
Type: text/x-vcard
Size: 214 bytes
Desc: not available
Url : /archives/attachments/20040916/c074487b/RodrigoMaldonado.vcf
------------------------------
Message: 7
Date: Fri, 17 Sep 2004 09:54:30 +0200
From: Philip Van Hoof <spamfrommailing freax org>
Subject: Using remote method-calls over a tcp/ip network
To: orbit-list gnome org
Cc: Peter Van Osta <pvosta maia-scientific com>
Message-ID: <1095407670 4382 16 camel localhost localdomain>
Content-Type: text/plain
Hi there ORBitters,
While I can use my remote-call perfectly if the ORBit-server is running
locally on the same machine as the client, and while this is working
like a charm, I can't get some specific methods, which return a
structure that for example contains a sequence, working over a remote
tcp/ip connection with client and server.
I am always getting a COMM_FAILURE exception. Do note, however, that
when I am running both on the same machine, it 'does' work perfectly
(really perfectly, yes). The server, in this case, is not segfaulting
nor crashing at all. It resumes normal operation indeed.
We are also having other problems with transferring structures that
contain an array between two ORB versions (being ORBit-2 for the client
and ORBit-1 for the server). The transfer of structures with arrays
between an ORBit-1 client and an ORBit-1 server also works perfectly.
But with a ORBit-2 client and an ORBit-1 server, all (stack) data after
the arrays is messed-up. It's causing the receiving server to crash
(segfault in eval.c at line 41. I can give you a backtrace from gdb but
it's not telling a lot except that it's crashing in the ORBit-1 shared
objectfiles).
We are preparing a Hello-World type of application to illustrate both
problems.
We are using the ORBit-1 and ORBit-2 softwares which are included with
Fedora Core 2 and/or with RedHat 7.2 and 8 and 9. It looks like we are
getting the same results on all versions (but we need to test some more
before drawing that conclusion).
--
Philip Van Hoof, Software Developer @ Cronos
home: me at freax dot org
gnome: pvanhoof at gnome dot org
work: philip dot vanhoof at cronos dot be
junk: philip dot vanhoof at gmail dot com
http://www.freax.be, http://www.freax.eu.org
------------------------------
Message: 8
Date: Fri, 17 Sep 2004 16:10:12 +0800
From: Kuang-Chun Cheng <kccheng openate com>
Subject: Re: Using remote method-calls over a tcp/ip network
To: spamfrommailing freax org
Cc: Peter Van Osta <pvosta maia-scientific com>, orbit-list gnome org
Message-ID: <414A9BE4 2090401 openate com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
For COMM_FAILURE problem. ORBit2 use hostname instead of IP
address to resolve server. Make sure you can connect to your server
via hostname.
I'm using ORBit2 myself and a lot of my customers have the same problem
when setup the ORBit2 client. I think this issue should be put in FAQ.
Regards,
Kuang-Chun Cheng
kccheng openate com
Philip Van Hoof wrote:
>Hi there ORBitters,
>
>While I can use my remote-call perfectly if the ORBit-server is running
>locally on the same machine as the client, and while this is working
>like a charm, I can't get some specific methods, which return a
>structure that for example contains a sequence, working over a remote
>tcp/ip connection with client and server.
>
>I am always getting a COMM_FAILURE exception. Do note, however, that
>when I am running both on the same machine, it 'does' work perfectly
>(really perfectly, yes). The server, in this case, is not segfaulting
>nor crashing at all. It resumes normal operation indeed.
>
>We are also having other problems with transferring structures that
>contain an array between two ORB versions (being ORBit-2 for the client
>and ORBit-1 for the server). The transfer of structures with arrays
>between an ORBit-1 client and an ORBit-1 server also works perfectly.
>But with a ORBit-2 client and an ORBit-1 server, all (stack) data after
>the arrays is messed-up. It's causing the receiving server to crash
>(segfault in eval.c at line 41. I can give you a backtrace from gdb but
>it's not telling a lot except that it's crashing in the ORBit-1 shared
>objectfiles).
>
>We are preparing a Hello-World type of application to illustrate both
>problems.
>
>We are using the ORBit-1 and ORBit-2 softwares which are included with
>Fedora Core 2 and/or with RedHat 7.2 and 8 and 9. It looks like we are
>getting the same results on all versions (but we need to test some more
>before drawing that conclusion).
>
>
>
>
------------------------------
Message: 9
Date: Fri, 17 Sep 2004 10:37:31 +0200
From: Philip Van Hoof <spamfrommailing freax org>
Subject: Re: Using remote method-calls over a tcp/ip network
To: kccheng openate com
Cc: Peter Van Osta <pvosta maia-scientific com>, orbit-list gnome org
Message-ID: <1095410251 4382 43 camel localhost localdomain>
Content-Type: text/plain
On Fri, 2004-09-17 at 16:10 +0800, Kuang-Chun Cheng wrote:
> For COMM_FAILURE problem. ORBit2 use hostname instead of IP
> address to resolve server. Make sure you can connect to your server
> via hostname.
>
> I'm using ORBit2 myself and a lot of my customers have the same problem
> when setup the ORBit2 client. I think this issue should be put in FAQ.
This sounds like an issue that we needed to solve for ORBit-1 to ORBit-1
tcp/ip connections. The question, however, remains in that case: howcome
some methods do work and some specific methods don't during the same
session?
Is ORBit-2 using different resolving-techniques for different data
structures?
However, both client and server can resolve the hostname and ip-
addresses correctly. To fix the issue above, we added this line at the
serverside:
192.168.0.23 linux-2
It makes the server resolve it's own hostname to his network ipadress in
stead of localhost.
At the server:
[pvanhoof linux-2 Linux_Srvr]$ cat /etc/hosts
192.168.0.23 linux-2
127.0.0.1 localhost.localdomain localhost
[pvanhoof linux-2 Linux_Srvr]$
At the client
[freax lort eazyx]$ cat /etc/hosts
127.0.0.1 localhost.localdomain lort localhost
# Maia
192.168.0.23 linux-2
[freax lort eazyx]$
Also note that the named for this network is also configured correctly
[freax lort eazyx]$ cat /etc/resolv.conf
; generated by /sbin/dhclient-script
search maia-scientific.com
nameserver 192.168.0.2
[freax lort eazyx]$ nslookup linux-2
Server: 192.168.0.2
Address: 192.168.0.2#53
Name: linux-2.maia-scientific.com
Address: 192.168.0.23
[freax lort eazyx]$
This is the IOR-string (I can't make much out of it myself):
IOR:010000001200000049444c3a457a78532f537276723a312e3000000002000000caaedfba
54000000010100002c0000002f746d702f6f726269742d7076616e686f6f662f6f72622d3230
3336343438393438333433333536303037000000000018000000000000001013b87e5139df3e
01000000ff2d996cf666a4340000000044000000010100001c0000006c696e75782d322e6d61
69612d736369656e74696669632e636f6d000797000018000000000000001013b87e5139df3e
01000000ff2d996cf666a434[
Thanks for the fast response
--
Philip Van Hoof, Software Developer @ Cronos
home: me at freax dot org
gnome: pvanhoof at gnome dot org
work: philip dot vanhoof at cronos dot be
junk: philip dot vanhoof at gmail dot com
http://www.freax.be, http://www.freax.eu.org
------------------------------
_______________________________________________
orbit-list mailing list
orbit-list gnome org
http://mail.gnome.org/mailman/listinfo/orbit-list
End of orbit-list Digest, Vol 5, Issue 3
****************************************
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]
