Sun Link OSI

Hi!
Need help!!!
During building an application used SunLink OSI (integrated in Solaris 10) we have some troubles.
I need to perform outgoing connection over RFC 1006 using specific port number.
It’s essential because of our customer doesn’t open all ports due security reasons.
At the moment our application works with TLI interface according to “SunLink® OSI 8.1 TLI Programmer’s Reference”
See /opt/SUNWconn/osinet/example/rfc1006/rfcclient.c
/opt/SUNWconn/osinet/example/rfc1006/rfcserver.c
Our app does’t supply local address. TCP port for outgoing connection are being choosed automatically by driver (/dev/otk)
Addres format according to “SunLink® OSI 8.1 TLI Programmer’s Reference” is:
[tsaplen][tsap.............][nsaplen][nsap...................]
Tsap – string characters
Nsap – IP address 4 bytes
I’ve tried to add port information to the end of the NSAP (6bytes), but it cause function t_bind returns -1 and t_error is 1 (Incorrect address format))
I’ve also tried to pass 6 bytes of IP + port as TSAP but in vain.
How can I supply specific port for outgoning connection over RFC 1006?
Is it possible at all?
Thx

Hi fotis83 ,
Thanks for your reply.
when i test the connection, it shown as
I have connected successfully AD to SUN IDM.
but how can I check it,
whether if it is connected, then i can see all the users of AD in SUN IDM.
or
any thing else to do, for bring of user list from AD TO SUN IDM.
please suggest me, if you have any idea.
Thanks,
vardhan

Similar Messages

  • Passing "-t" option to link-editor via CC

    Hi!
    Here is the simple task.
    we have some source file which is compiling with simple command:
    CC ./some_simple_file.C -lsome_lib -lanother_libCompilation process succed but we have the warning message:
    ld: warning: symbol ‘array’ has differing sizes:
    (file /path/libsome_lib.a value=0x4; file /path/libanother_lib.o value=0x8);
    /path/libsome_lib.a definition takenWell, apart from this warning - all works fine and the link-editor do matching symbol correctly in this case.
    To do not confuse users/parsers/mgrs team - we decided to remove this warning. Ok, here is a quote from sun linker document about "how to remove such warning" in section "Symbols Processing"->"Complex resolution":
    +....In both of these cases, the diagnostic can be suppressed by using the link-editor's -t option....+
    Thats nice! Lets go with this new option:
    CC -t ./some_simple_file.C -lsome_lib -lanother_libWe finally removed warning. But...oops! weve got another one!
    CC: Warning: Option -t passed to ld, if ld is invoked, ignored otherwiseQuestion! How we can pass -t option to the link-editor via CC cmd call and without any warnings?

    There doesn't seem to be a way to pass -t to the linker without the warning. You could file a request for enhancement via your Oracle support channel, if you have a service contract. Or you could file one at [http://bugs.sun.com]. (I'm not sure if that site is still active.)
    The best solution is to fix the problem and eliminate the reason for the warning, by making the declarations consistent. You might suppress warnings that matter if you always use -t.

  • How to use Parallel Compilation -j using sun make?

    Hello,
    My c++ project have lots of cpp files and it takes around 5 - 5.5 hours to build the application which is definitely a lot of time. Little bit of googling landed me to the following sun link to reduce the build time.
    http://developers.sun.com/solaris/articles/parallel_make.html
    By using parallel make option of -j as it should reduce the build time considerably as i was using Solaris 9 on Sparc Sun-Fire-V440 which is having 4 CPU.
    But on invoking make with a -j option i.e "make all -j 4" I get the following error :
    make: Warning: Ignoring DistributedMake -j option
    I even tried to use gcc make instead of sun(/usr/ccs/bin/make) make but none of them is actually helping me to reduce my build time.
    Can some please guide me how should i go ahead withe usage of -j option on solaris.

    You should be aware that amount of parallelism that dmake can exploit depends fully on your Makefiles.
    Unnecessary dependencies can significantly reduce amount of parallelism.
    Recursive makefiles (make rules in turn calling makes) will hamper parallelism as well.
    Hidden dependencies (those not directly specified in rules) will lead to spontaneous build failures.
    It will take some discipline to write makefiles properly, but in a long term it will pay off.
    regards,
    __Fedor.

  • Sun ld comdat group support

    While this may be a little OT, given that it discusses gcc/binutils, the underlying topic is the solaris linker in /usr/ccs/bin/ld, and this seemed like the most likely forum for linker expertise.
    The following discussion concludes that the sun linker does not support COMDAT groups adequetly to build gcc 4 with the gnu assembler and the sun linker (as is recommmended by the gcc folks, and as is done by sun for /usr/sfw/bin/gcc in solaris 10)
    http://gcc.gnu.org/ml/gcc/2005-04/msg01332.html
    But the documentation for the linker seems to suggest otherwise:
    http://docs.sun.com/app/docs/doc/817-1983/6mhm6r4fj?a=view#chapter7-11598
    In the most recent binutils ld testsuite, there is a test for comdat support:
    -------- group1a.s ---------
    .section .text,"axG",%progbits,foo_group,comdat
    .weak foo
    foo:
    .word 0
    -------- group1b.s ---------
    .section .text,"axG",%progbits,foo_group,comdat
    .global foo
    .global bar
    foo:
    .word 0
    bar:
    Run these through the gnu assembler:
    /usr/local/binutils-2.16.1/bin/as --gdwarf2 group1a.s -o group1a.o
    /usr/local/binutils-2.16.1/bin/as --gdwarf2 group1b.s -o group1b.o
    And then link them, once with the sun linker, and once with the gnu linker:
    /usr/local/binutils-2.16.1/bin/ld group1a.o group1b.o -o linked.gnu
    /usr/ccs/bin/ld group1a.o group1b.o -o linked.sun
    Now, if we look at foo and bar in the generated images
    elfdump -s -N .symtab linked.gnu | egrep "foo|bar"
    [8] 0x08048074 0x00000000 NOTY WEAK D 0 .text foo
    [11] 0x00000000 0x00000000 NOTY GLOB D 0 UNDEF bar
    elfdump -s -N .symtab linked.sun | egrep "foo|bar"
    [15] 0x080501ef 0x00000000 NOTY GLOB D 0 .text bar
    [21] 0x080501ed 0x00000000 NOTY GLOB D 0 .text foo
    we can see that the gnu linker generated foo as a weak symbol, and left bar undefined, while the sun linker emitted foo as a global, and defined bar.
    ok. so now try without the comdat flags:
    ----- nogroup1a.s -----
    .section .text,"ax",%progbits
    .weak foo
    foo:
    .word 0
    ----- nogroup1b.s -----
    .section .text,"ax",%progbits
    .global foo
    .global bar
    foo:
    .word 0
    bar:
    and assemble and link as above. Now, if we look at the output from elfdump:
    elfdump -s -N .symtab nogroup.linked.sun | egrep "foo|bar"
    [15] 0x080501f2 0x00000000 NOTY GLOB D 0 .text bar
    [21] 0x080501f0 0x00000000 NOTY GLOB D 0 .text foo
    elfdump -s -N .symtab nogroup.linked.gnu | egrep "foo|bar"
    [8] 0x08048078 0x00000000 NOTY GLOB D 0 .text foo
    [11] 0x0804807a 0x00000000 NOTY GLOB D 0 .text bar
    Now, we get identical results for foo and bar with the two linkers. Even more interesting is that for the sun linker, the results did not change when we removed the comdat flags. It is as if the sun linker happily absorbs and then ignores the fact that the sections are marked as COMDAT. So what is going on here? Does the sun linker really support COMDAT? Have I misinterpreted something? Which linker is right, according to the ELF spec?
    Sorry for the long post, any insight would be appreciated.

    Ok. That is a reasonable, if not very helpful answer.
    In the meantime, here is a simpler example that shows the problem a little more clearly:
    cat foobar.s.section .text,"axG",%progbits,foo_group,comdat
    .global foo
    foo:
    .word 0
    .section .text,"ax",%progbits
    .global bar
    bar:
    .word 0
    .section .text,"ax",%progbits
    .global baz
    baz:
    .word 0
    Now, assemble this twice into two different object files:
    /usr/local/binutils-2.16.1/bin/as gdwarf2 foobar.s -o foobar2.o /usr/local//binutils-2.16.1/bin/as gdwarf2 foobar.s -o foobar1.o
    If we link these two object files together, we should get two errors about undefined symbols: one for bar, and one for baz. foo should be ok because its COMDAT.
    So:
    /usr/local/binutils-2.16.1/bin/ld -shared foobar1.o foobar2.o foobar2.o: In function `bar':
    : multiple definition of `bar'
    foobar1.o:: first defined here
    foobar2.o: In function `baz':
    : multiple definition of `baz'
    foobar1.o:: first defined here
    Which is what we expect. But /usr/ccs/bin/ld gives:
    /usr/ccs/bin/ld -G foobar1.o foobar2.o
    ld: fatal: symbol `foo' is multiply-defined:
    (file foobar1.o type=NOTY; file foobar2.o type=NOTY);
    ld: fatal: symbol `bar' is multiply-defined:
    (file foobar1.o type=NOTY; file foobar2.o type=NOTY);
    ld: fatal: symbol `baz' is multiply-defined:
    (file foobar1.o type=NOTY; file foobar2.o type=NOTY);
    ld: fatal: File processing errors. No output written to a.out
    which seems incorrect to me.

  • Linker problem using version 12.4 on Linux

    I am trying to use Solaris Studio 12.4 on Linux ( OpenSuse 13.2 ). I have installed the compiler and set up the PATH to its bin directory. When I compile/link a program I am receiving:
    "sun.link ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem
        "/opt/oracle/solarisstudio12.4/bin/CC" -g -m64 -KPIC   -o "../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem" "../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o"  -Bdynamic -lrt -Bstatic  -Bdynamic
    CC: Warning: failed to detect system linker version, falling back to custom linker usage"
    and this is followed by:a series of errors involving standard streams, standard output, and the like:
    "../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `__SLIP.INIT_A()':
    /opt/oracle/solarisstudio12.4/lib/compilers/CC-gcc/include/c++/4.8.2/iostream:74: undefined reference to `std::ios_base::Init::Init()'
    ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `__SLIP.FINAL__A':
    test_after_array_elem.cpp:(.text+0x28f): undefined reference to `std::ios_base::Init::~Init()'
    ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `boost::report_errors()':
    /home/suseldiener/vcs/modular-boost/libs/vmd/test//../../../boost/core/lightweight_test.hpp:133: undefined reference to `std::cerr'
    /home/suseldiener/vcs/modular-boost/libs/vmd/test//../../../boost/core/lightweight_test.hpp:139: undefined reference to `std::cerr'
    /home/suseldiener/vcs/modular-boost/libs/vmd/test//../../../boost/core/lightweight_test.hpp:139: undefined reference to `std::ostream::operator<<(int)'
    ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `boost::detail::error_impl(char const*, char const*, int, char const*)':
    /home/suseldiener/vcs/modular-boost/libs/vmd/test//../../../boost/core/lightweight_test.hpp:75: undefined reference to `std::cerr'
    /home/suseldiener/vcs/modular-boost/libs/vmd/test//../../../boost/core/lightweight_test.hpp:75: undefined reference to `std::ostream::operator<<(int)'
    ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `std::ostream & std::endl<char, std::char_traits<char> >(std::ostream &)':
    test_after_array_elem.cpp:(.text.std::ostream & std::endl<char, std::char_traits<char> >(std::ostream &)[std::ostream & std::endl<char, std::char_traits<char> >(std::ostream &)]+0x75): undefined reference to `std::ostream::put(char)'
    ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `std::ostream & std::operator<<<std::char_traits<char> >(std::ostream &, char const*)':
    /opt/oracle/solarisstudio12.4/lib/compilers/CC-gcc/include/c++/4.8.2/ostream:536: undefined reference to `std::ostream & std::__ostream_insert<char, std::char_traits<char> >(std::ostream &, char const*, long)'
    ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `std::ios::setstate(std::_Ios_Iostate)':
    test_after_array_elem.cpp:(.text.std::ios::setstate(std::_Ios_Iostate)[std::ios::setstate(std::_Ios_Iostate)]+0x7d): undefined reference to `std::ios::clear(std::_Ios_Iostate)'
    ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `std::ostream & std::flush<char, std::char_traits<char> >(std::ostream &)':
    test_after_array_elem.cpp:(.text.std::ostream & std::flush<char, std::char_traits<char> >(std::ostream &)[std::ostream & std::flush<char, std::char_traits<char> >(std::ostream &)]+0x37): undefined reference to `std::ostream::flush()'
    ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `std::ctype<char> const& std::__check_facet<std::ctype<char> >(std::ctype<char> const*)':
    /opt/oracle/solarisstudio12.4/lib/compilers/CC-gcc/include/c++/4.8.2/bits/basic_ios.h:49: undefined reference to `std::__throw_bad_cast()'
    ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem.o: In function `std::ctype<char>::widen(char) const':
    /opt/oracle/solarisstudio12.4/lib/compilers/CC-gcc/include/c++/4.8.2/bits/locale_facets.h:869: undefined reference to `std::ctype<char>::_M_widen_init() const'
    ...failed sun.link ../../../bin.v2/libs/vmd/test/test_after_array_elem.test/sun-cxx11-12.4/debug/address-model-64/test_after_array_elem..."
    This is all being done from Boost and Boost Build and the linker errors are occurring because of the run-time use of the lightweight_test component of Boost.
    There is no problem using gcc directly compiling in 64-bit mode but Solaris Studio C++ 12.4 is failing.
    Any ideas why it cannot find the system linker and use it, or why it is failing with the "custom linker", whatever that is, would be appreciated. If Boost Build needs some other compiler/linker option to work I can adjust that. Maybe I need to install some library for -std=c++11 mode or maybe -std=c++11 mode dopes has some sort of bug.

    > undefined reference to `std::ios_base::Init::~Init()'
    It looks like you link with different -library/-std set of options compared to those that you compiled with.
    > CC -g -m64 -o test *.o -Bdynamic -lrt -Bstatic  -Bdynamic
    This link line uses default settings, which is -library=Cstd for Studio 12.4.
    And since you use Boost most likely you compile with -library=stlport4 or with -std=c++03, which implies -library=stdcpp (gcc STL).
    Otherwise you would most likely get compilation failures (since -library=Cstd is a rather outdated STL).
    You should use pretty much the same set of options when you compile and when you link.
    Verify your build configuration (CXXFLAGS/LDFLAGS whatever).
    regards,
      Fedor.

  • Managing java files and packages

    Hi,
    I am just starting out with java and have reached the stage where I want to store my files in packages and the correct directory structure. I have read through the sun link below and have a question they seam to miss. They say that .java and .class files should be kept in separate places. ie sources and classes directories with the package directory names below. But I have found that unless you have the .class files you are referencing in the same directory as the .java files when you come to compile you will get an error so they have to be stored in the same directory???.. Does this make sence or have I gotten it all wrong.
    Thanks
    Joolz
    http://java.sun.com/docs/books/tutorial/java/interpack/managingfiles.html

    You should put path containing your classes in your classpath. There is a -classpath option in javac.

  • Please show me what to do per the security upgrades re firefox and ie7

    Hi, I didn't know where else to post this. I am just a home user. I have both Firefox 3.0 and Internet Explorer 7. I have windows premium home VISTA.
    I recieved the notice (portion below) and the instructions at sun links contained do not give directions which show explicitly what to do regarding each as pertains to BOTH firefox and ie7 browsers, at least in the first one which is all I read and then I registered here for help.
    It mentions firefox then shows nothing in what to do for firefox in the first one.
    So, can someone please give me step by step on each line item link; what to do to protect my system as regards this latest security notice?
    The links are below this text, and I just need a walk thru how to protect my sys as regards the below security notices; exactly what to do for both ie7 and firefox, and my windows is VISTA. Thanks so very much!
    I am just a home user with no idea about java as well new to firefox (i use both browsers) as well.
    Again, thank you..
    US-CERT encourages users to review the following Sun Alerts and applyany necessary updates:
    * Sun Alert 238628 - Security Vulnerabilities in the Java RuntimeEnvironment related to the processing of XML Data
    * Sun Alert 238666 - A Security Vulnerability with the processing offonts in the Java Runtime Environment may allow Elevation ofPrivileges
    * Sun Alert 238687 - Security Vulnerabilities in the Java RuntimeEnvironment Scripting Language Support
    * Sun Alert 238905 - Multiple Security Vulnerabilities in Java WebStart may allow Privileges to be Elevated* Sun Alert 238965 - Security Vulnerability in Java ManagementExtensions (JMX)
    * Sun Alert 238966 - Security Vulnerability in JDK/JRE Secure StaticVersioning
    * Sun Alert 238967 - Security Vulnerability in the Java RuntimeEnvironment Virtual Machine may allow an untrusted Application orApplet to Elevate Privileges
    * Sun Alert 238968 - Security Vulnerabilities in the Java RuntimeEnvironment may allow Same Origin
    Policy to be Bypassed US-CERT will provide additional information as it becomes available.
    Relevant Url(s):
    <http://sunsolve.sun.com/search/document.do?assetkey=1-66-238968-1> <http://sunsolve.sun.com/search/document.do?assetkey=1-66-238967-1> <http://sunsolve.sun.com/search/document.do?assetkey=1-66-238628-1> <http://sunsolve.sun.com/search/document.do?assetkey=1-66-238687-1> <http://sunsolve.sun.com/search/document.do?assetkey=1-66-238966-1> <http://sunsolve.sun.com/search/document.do?assetkey=1-66-238905-1> <http://sunsolve.sun.com/search/document.do?assetkey=1-66-238965-1> <http://sunsolve.sun.com/search/document.do?assetkey=1-66-238666-1>
    Edited by: rachael35 on Jul 10, 2008 10:10 AM

    JoachimSauer wrote:
    From a quick glance it seems like a list of vulnerabilities fixed in the latest Java releases.
    Just make sure that you've got the latest version of any Java Runtime Environment that you have.
    So if you've got Java 6, make sure it's Update 7.
    If you've got Java 5, make sure it's Update 16.
    You should get the updates automatically, if you don't want to wait for that, just go to [http://www.java.com/] and follow the instructions.
    If you've got any earlier versions, get rid of them ;-)Hi, thanks for replying,
    I am trying to understand what I need to do. In the first link in the vulnerabiliity list, the link by sun, when go in there, speaks of uninstalling which brings me here:
    http://java.com/en/download/help/uninstall_java.xml
    Please note that the uninstall at that page does not include VISTA.
    I have Vista
    Is there another link for uninstalling JAVA which works on Vista?
    Thanks!
    Edited by: rachael35 on Jul 10, 2008 11:20 AM

  • Ld.so.1 problem - fatal: relocation error - "referenced symbol not found"

    Hi, I've been trying to compile and run my application non Solaris using CC (detils below) but haven't got luck with running the application because of a linker error :
    ld.so.1: java: fatal: relocation error: file ./libtest.so: symbol __1cDstdMbasic_string4Ccn0ALchar_traits4Cc__n0AJallocator4Cc___J__nullref_: referenced symbol not found
    Here's what I'm trying to do :
    I'm compieing one library to include another shared library whcih uses the "string" class (STL). The compilation of both the libraries proceeds perfectly fine. However, when i preload this library with LD_PRELOAD and run a JAVA application (which will have calls to some function in the preloaded library), the get the above linker error.
    Here's what all I have tries to get thsi wroking : ( I understand that this error is due to a missing library that CC has not linked to and i've no idea what the library is!!!)
    1.> # LD_LIBRARY_PATH=/usr/lib:$LD_LIBRARY_PATH
    # export LD_LIBRARY_PATH - NO LUCK :-(
    2.> linked to /opt/SUNWspro/lib/libCrun.so (found this from another forum on the net), but NO LUCK :-(
    3.> export LD_DEBUG=bindings
    export LD_DEBUG_OUTPUT=/home/log.txt (so that all sysmbols refered are logged!!!)
    i did a grep on the log for the symbol but NO LUCK!!! :-(
    Please can you help me with this? What is the library that is to be linked too, or is there something else?
    Details About my system, compiler and compilation options :
    1.> uname -a output : SunOS sh1sol9 5.9 Generic_117171-12 sun4u sparc SUNW,Sun-Fire-280R Solaris
    2.> CC version : Sun WorkShop 6 update 2 C++ 5.3
    3.> Compilation flags i've used to build the library :
    FIRST LIBRARY : CC -G -xcode=pic32 -I../common/ -ldl ../common/logger.cpp helper_functions.cpp BCPEntries.cpp BCP_class.cpp -o libBCP.so
    SECOND LIBRARY : CC -G -xcode=pic32 -I../common/ test_BCP_construct.cpp -lBCP -L. -o libtest.so
    Please help me out of this,
    Any and every help (or even a hint to it) will be appreciated!!!
    Thanks in advance,
    Madhur Kumar Tanwani

    Refer to the C++ Users Guide that comes with the compiler, the chapter on Building Libraries, which explains these points in more detail.
    When you build a C++ shared library, you must explicitly list all the libraries that it depends on. Unlike building an executable, the CC -G command does not by default create dependencies on system libraries. Although not explained well in the Guide, you should be sure you get the shared (.so) versions of all libraries, not the static (.a) versions.
    All C++ programs require at least libCrun and libc. Programs that use the C++ Standard library (using anything in the std namespace, like std::string) also need to link to the appropriate version of the library.
    The default library is libCstd. If you compile using the optional STLport library, you need the libstlport instead. You cannot use both libCstd and libstlport in the same program.
    In your case, the build command would look something like this, assuming you are using the default libCstd:
    CC -G -o mylib.so ..... -library=Cstd,Crun -lc
    Be sure that any preceeding -Bstaic option is followed by -Bdynamic before listing these system libraries.
    To verify you have picked up the dynamic libraries, run the command
    ldd mylib.so
    to see a list of dependencies. If you don't see the following in the list
    /usr/lib/libCstd
    /usr/lib/libCrun
    /usr/lib/libc (or /lib/libc)
    something is wrong.
    A note about LD_LIBRARY_PATH: don't use it.
    Ideally, the environment variable should never be set. If you must set it to reach vendor-supplied libraries, never put the system directories /lib or /usr/lib on the path, and never put the compiler installation directory, like /opt/SUNWspro/... on the path. You can wind up forcing the runtime loader to seach directories in the wrong order and pick up inappropriate versions of libraries.
    For more on LD_LIBRARY_PATH, read this note in Rod Evans' blog. He's the Sun linker expert. The material here is available in the manuals, but his blog has a nice summary.
    http://blogs.sun.com/roller/page/rie/?anchor=tt_ld_library_path_tt

  • How latest version of fw 7.2.0 released before 7.1.7.f for T5240

    Hi Experts,
    There is some confusion related to firmware vesion of T5240(Maramba).
    7.2.0 -->Dec,23,2008
    7.1.7.f --> FEb,9,2009
    {font:Arial}{size:2px}{size}{font}{font:Arial}{size:2px}{size}{font}
    7.2.0 --> sunsolve.sun.com/search/document.do?assetkey=1-21-139444-01-1
    7.1.7.f --> sunsolve.sun.com/search/document.do?assetkey=1-21-136936-11-1
    As you can see the latest version of fw 7.2.0 was release before 7.1.7.f
    So can someone explain why this behaviour for fw release?
    Edited by: nitin.k on Apr 2, 2009 1:28 PM

    nitin.k wrote:
    Hi Experts,
    There is some confusion related to firmware vesion of T5240(Maramba).
    7.2.0 -->Dec,23,2008
    7.1.7.f --> FEb,9,2009
    {font:Arial}{size:2px}{size}{font}{font:Arial}{size:2px}{size}{font}
    7.2.0 --> sunsolve.sun.com/search/document.do?assetkey=1-21-139444-01-1
    7.1.7.f --> sunsolve.sun.com/search/document.do?assetkey=1-21-136936-11-1
    As you can see the latest version of fw 7.2.0 was release before 7.1.7.f
    So can someone explain why this behaviour for fw release?
    Edited by: nitin.k on Apr 2, 2009 1:28 PMWhen I was checking the links that I originally posted I discovered that the Sun link I used took me to the wrong FW patches. (as Roseanne Roseannadanna would say "never mind")
    It is interesting that they fix a much different set of problems.
    If it's important to you, I suggest contacting support.
    I guess I'll be thankful I have a T5220 ...
    have a good day,
    Glen
    Edited by: sysglen on Apr 2, 2009 9:04 AM

  • HTTP Adapter Settings (Receiver Channel)

    Hi Everyone,
    I need some help with setting up HTTP Receiver Channel.
    What I am trying to do is download some files from a URL. e.g www.xyz.com/download.asp.
    So I have set up a RFC destination to Remote server then we through a dummy message from the HTTP client to trigger the channel. It is all good and working. However the issue part is username/password has to be sent as application/x-www-form-urlencoded or multipart/form using POST method.
    So it has to go something like this username=abc&password=123
    And this is the part where I am stuck. I have read a bit about it and sap documentation suggests using prolog and epilog but I don't seem to find any example of it anywhere. If someone has done anything similar can they please help. Can't find any source relevant.
    regards

    Hi,
    Yes I understood your steps and the problem (actually after posting reply i had some doubts in my mind too, so went on a SDN forum hunt)
    After doing a bit of research work my finding are:
    1. IT is difficult to use http adapter with this content type as the credentials given in http adapter is not used when you have application/x-www-form-urlencoded. The thread which helped me  to understand was this
    Regarding Content type application/x-www-form-urlencoded in HTTP Adapter
    2. There in the thread stefan is proposing java proxy for doing this type of connectivity so next search tool for me was google. Doing a bit of search i found this SUN link http://developers.sun.com/mobility/midp/ttips/HTTPPost/
    3. So as a solution for you I would propose to use a receiver java proxy and write the code as mentioned in SUN.
    Hope this may be of some help to you
    Regards
    Suraj

  • Is it possible to obtain the src of pet store 2.0 without CVS?

    Is there any way I could obtain the petstore 2,0 as a zip file like it used to be in pet store 1.3.2?
    Please, let me know where I can get it?
    It is amazing, all the download button in Sun link to yet another place of blueprints.
    previous sites with download link says with source code but blueprint download only as *.class
    file inside of installer file.
    Also, http://blog.naver.com/bluejames77/80049199960 says No. 2
    2. &#45908;&#48660;&#53364;&#47533;&#54644;&#49436; &#50517;&#52629;&#51012; &#54396;&#45796;&#51020; index.html&#51012; &#48372;&#47732;
    Java EE 5 Application Server&#51012; &#49444;&#52824;&#54616;&#44144;&#45208; &#54841;&#51008; GlassFish Project&#47484; &#49444;&#52824;&#54616;&#46972;&#44256; &#45208;&#50728;&#45796;.
    Translate ----
    after unzip, there is index.html file which specifies to install Java EE 5 application server or glass fish project.
    But, no index.html when I download the same date installer. In fact, any one can go to the CVS site and see
    under www/ there is no *.java file???? only html file???
    Is there anyway somebody in blueprint who just create source zip file like it used to be in 1.3.2?
    with setup.xml and build.xml and link up there? It will say one day work for somebody.
    Also ERROR -
    http://webdev2.sun.com/petstore/faces/catalog.jsp?catid=Cats
    Also, shows error! at chrome I don't know at iexplorer.
    please try -

    You have to execute the installer jar file.
    This will create a folder with all the source codes inside.
    I wasn't aware of any need for cvs.

  • Gcc 3.3.2 problem:  libgcc_s.so.1 not found

    I am trying to compile a perl module (Math::GMP) on a Solaris 8 platform. We have gcc 3.3.2 installed, and everything appears to go well until I try the following:
    % ldd blib/arch/auto/Math/GMP/GMP.so
        libgmp.so.3 => /usr/local/lib/libgmp.so.3
        libc.so.1 => /usr/lib/libc.so.1
        libgcc_s.so.1 => (file not found)
        libdl.so.1 => /usr/lib/libdl.so.1
        /usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1
    I have confirmed that /usr/local/lib/libgcc_s.so.1 exists, so I don't understand what is wrong. Any clues?

    It isn't entierly clear if the libgcc_s.so.1 dependancy comes directly from "GMP.so" or indirectly from
    "libgmp.so.3".
    You can check that with the dump command, looking at the recorded "SONAME" dependencies:
    dump -Lv blib/arch/auto/Math/GMP/GMP.so
    and
    dump -Lv /usr/local/lib/libgmp.so.3
    While your looking at the output from "dump", have a look at the recorded runtime shared library path, too.
    (the "RUNPATH" or "RPATH" entries). Apparently one of "GMP.so" or "libgmp.so.3" depends on libgcc_s.so.1
    but does not have it's runtime shared library path set to include "/usr/local/lib".
    The fix is to link the bad shared library with the option "-R /usr/local/lib" (assuming your gcc is configured to
    use the Sun linker).

  • SunStudio 12 Fails to compile template function with template argument

    Hi
    Look at the code below:
    #include <iostream>
    template<class InIt, class OutIt>
    InIt copy_n(InIt first,  typename  std::iterator_traits<InIt>::difference_type  length, OutIt dest)
       for (;length--; ++dest, ++first)
          *dest = *first;
       return first;
    int main()
            char str1[]="Sumit";
            char str2[20];
            copy_n(str1,6,str2);
            std::cout << str2;
            return 0;
    }I have simplified this to pin point the problem, This code dosent compile with SunStudio12 C++ Compiler, (however this works fine on g++ and vc++). The problem is with second parameter
    typename std::iterator_traits<InIt>::difference_type  lengthinstead if we specify any simple type as below, it works fine
    int lengthSuns "http://developers.sun.com/sunstudio/documentation/ss12/whatsnew.html" says :
    Support for the following open-source libraries:
    BOOST http://www.boost.org
    The above code snippet is picked from "boost/boost_1_35_0/boost/interprocess/detail/algorithms.hpp" line number 65.
    Any Suggestions or work arounds??+
    The above Suns link also says:
    Complex expressions in template parameters.
    The compiler was previously unable to handle expressions in non-type template arguments in certain cases.
    I did not use any compilation flag, am I missing something?
    Edited by: sumitkumar on May 20, 2008 12:20 PM
    Edited by: sumitkumar on May 20, 2008 12:25 PM
    Edited by: sumitkumar on May 20, 2008 12:25 PM

    >
    Suns "http://developers.sun.com/sunstudio/documentation/ss12/whatsnew.html" says :
    Support for the following open-source libraries:
    BOOST http://www.boost.org
    Boost 1.34.1 should work OK. I haven't tried Boost 1.35 yet.
    The above code snippet is picked from "boost/boost_1_35_0/boost/interprocess/detail/algorithms.hpp" line number 65.
    Any Suggestions or work arounds??+
    The above Suns link also says:
    Complex expressions in template parameters.
    The compiler was previously unable to handle expressions in non-type template arguments in certain cases.
    I did not use any compilation flag, am I missing something?
    Make sure that you compile and link with -library=stlport4, otherwise you will be limited in the Boost libraries that you will be able to use.
    Paul

  • GGobi fails to compile, wrong function calls in graphviz.c

    Hi there,
    I'm having troubles to makepkg ggobi, where make fails due to the below pasted errors.
    I am on a 64-bit system (a MacBook I put Arch on), not sure if that is relevant.
    Does anyone have an idea what is wrong here?
    Thanks,
    Alex
    graphviz.c: In function 'dot_neato_layout_cb':
    graphviz.c:215:3: error: too few arguments to function 'aginit'
    aginit();
    ^
    In file included from /usr/include/graphviz/types.h:49:0,
    from /usr/include/graphviz/gvc.h:17,
    from graphviz.c:16:
    /usr/include/graphviz/cgraph.h:359:13: note: declared here
    extern void aginit(Agraph_t * g, int kind, char *rec_name, int rec_size,
    ^
    graphviz.c:220:3: error: incompatible type for argument 2 of 'agopen'
    graph = agopen("graph", kind);
    ^
    In file included from /usr/include/graphviz/types.h:49:0,
    from /usr/include/graphviz/gvc.h:17,
    from graphviz.c:16:
    /usr/include/graphviz/cgraph.h:266:18: note: expected 'Agdesc_t' but argument is of type 'gint'
    extern Agraph_t *agopen(char *name, Agdesc_t desc, Agdisc_t * disc);
    ^
    graphviz.c:220:3: error: too few arguments to function 'agopen'
    graph = agopen("graph", kind);
    ^
    In file included from /usr/include/graphviz/types.h:49:0,
    from /usr/include/graphviz/gvc.h:17,
    from graphviz.c:16:
    /usr/include/graphviz/cgraph.h:266:18: note: declared here
    extern Agraph_t *agopen(char *name, Agdesc_t desc, Agdisc_t * disc);
    ^
    graphviz.c:226:5: error: too few arguments to function 'agnode'
    agnode(graph, name);
    ^
    In file included from /usr/include/graphviz/types.h:49:0,
    from /usr/include/graphviz/gvc.h:17,
    from graphviz.c:16:
    /usr/include/graphviz/cgraph.h:280:18: note: declared here
    extern Agnode_t *agnode(Agraph_t * g, char *name, int createflag);
    ^
    graphviz.c:248:7: error: too few arguments to function 'agedge'
    edge = agedge(graph, tail, head);
    ^
    In file included from /usr/include/graphviz/types.h:49:0,
    from /usr/include/graphviz/gvc.h:17,
    from graphviz.c:16:
    /usr/include/graphviz/cgraph.h:291:18: note: declared here
    extern Agedge_t *agedge(Agraph_t * g, Agnode_t * t, Agnode_t * h,
    ^
    graphviz.c:285:36: error: 'Agnode_t' has no member named 'name'
    agsafeset (graph, "root", ctr->name, NULL);
    ^
    graphviz.c:315:13: warning: assignment makes pointer from integer without a cast [enabled by default]
    sym = agfindattr(graph,"dim");
    ^
    graphviz.c:317:15: warning: assignment makes pointer from integer without a cast [enabled by default]
    sym = agraphattr(graph,"dim","");
    ^
    graphviz.c:319:26: error: 'attrsym_t' has no member named 'index'
    agxset(graph, sym->index, buf);
    ^
    graphviz.c:354:11: error: 'Agnode_t' has no member named 'u'
    node->u.width = node->u.height = .001;
    ^
    graphviz.c:354:27: error: 'Agnode_t' has no member named 'u'
    node->u.width = node->u.height = .001;
    ^
    Makefile:393: recipe for target 'plugin_la-graphviz.lo' failed
    make[3]: *** [plugin_la-graphviz.lo] Error 1

    >
    Suns "http://developers.sun.com/sunstudio/documentation/ss12/whatsnew.html" says :
    Support for the following open-source libraries:
    BOOST http://www.boost.org
    Boost 1.34.1 should work OK. I haven't tried Boost 1.35 yet.
    The above code snippet is picked from "boost/boost_1_35_0/boost/interprocess/detail/algorithms.hpp" line number 65.
    Any Suggestions or work arounds??+
    The above Suns link also says:
    Complex expressions in template parameters.
    The compiler was previously unable to handle expressions in non-type template arguments in certain cases.
    I did not use any compilation flag, am I missing something?
    Make sure that you compile and link with -library=stlport4, otherwise you will be limited in the Boost libraries that you will be able to use.
    Paul

  • Is there pkg command like 'rpm --upgrade'

    Hi
    I am using the pkgadd to install the application pkg. I want to know if there is pkg command like 'rpm --upgrade' to upgrade a pkg from verion 1.0 (already installed) to 2.0 (new version) directly? ('pkg-get -U' looks just performing a pkgrm and pkgadd cycle)
    Regarding the installation on Soalris, I still have another three questions:
    1. Dose Solaris support rpm or some other pkg tools?
    2. What installation tools that other enterprise products used on the Solaris, like: Oracle or DB2?
    3. Dose Solaris has upgrade version for 5.10, like 5.10.1 or something?
    I am very appreciated for your response. Thank you very much!

    Hi,
    To answer some your questions, lets start with the rpm one: there is a pkg for it called rpm-4.2. I think it is already installed on the system and if it isn't you should be able to find it one the DVD. For the enterprise tools, I would strongly recommend you do a google search for them and you will get some results forthe different kinds of tools you can install starting here at Sun; check there pages. The next version doesn't upgrade like linux does; the versions go like this: 6/06, 11/06, 7/07 and so on...
    You might want to check also the sun links too, there's a lot information for Solaris.

Maybe you are looking for