Re: Unbalbe to gdb core file generted from customer machine on local host
Hi there, Karl provided me some general information. Could gdb team please provide me some information? Thanks for your help. Best Regards, James Chen IBM - Forwarded by James C Chen/Ontario/IBM on 11/17/2014 11:09 AM - From: Karl Berry via RT i...@fsf.org To: James C Chen/Ontario/IBM@IBMCA Date: 11/16/2014 10:21 AM Subject:[gnu.org #960633] Unbalbe to gdb core file generted from customer machine Hi - thanks for writing, but sorry, but this is the general information address for the FSF and GNU. We aren't experts in gdb. I suggest writing to the public gdb help list, gdb@gnu.org. (FWIW, as far as I know it is necessary to use the machine the dump was made on, or an exact clone, since by its nature it will depend on many tiny details of the environment. But maybe there is some trick or alternative I don't know about.) Best wishes, Karl --- Have we been helpful to you today? Would you like to help the FSF continue to spread the word about software freedom? You too can become a member! Learn more at: http://donate.fsf.org
Re: Unbalbe to gdb core file generted from customer machine on local host
On Sunday, November 16 2014, James C. Chen wrote: Hi there, Thanks for providing gdb support. Hello, James. First of all, this mailing list is not used by the GDB project anymore. You should write to g...@sourceware.org, which is the official mailing list for the project. I created an executable binary (x86-64) on my local Linux server and let it run on customer Linux server. Later on, my binary generated a core file. I tried to gdb the core file on my local Linux server, but I can not get correct stack trace. It is not clear whether you compiled your binary with debug information or not. Have you used the -g flag on GCC? Also, if your binary is linked against shared libraries, you have to make sure that the debug information for those libraries is also installed. And you have to make sure that the versions of the libraries installed on your machine and on your client's machine are the same. (gdb) bt full #0 0x003a5d07288e in ?? () No symbol table info available. #1 0x1b36d548 in ?? () No symbol table info available. #2 0x1b2fab70 in ?? () No symbol table info available. #3 0x7fff60c30a9c in ?? () No symbol table info available. #4 0x1b36d548 in ?? () No symbol table info available. #5 0x1aafc860 in ?? () No symbol table info available. #6 0x2230 in ?? () from /lib64/libm.so.6 No symbol table info available. #7 0x545194cc0221 in ?? () No symbol table info available. #8 0x in ?? () No symbol table info available. These errors happen when GDB does not have access to the debug information of your binary/libraries. But when I gdb the core file on customer server where the core file was created, I am able to get correct stack trace. Question, do I have to gdb the core file on the customer server to get correct stack trace? Any way to gdb the core file on my local server? If you are using the same versions of the same programs/libraries, then you will be able to inspect the corefile in your machine without problems. -- Sergio GPG key ID: 0x65FC5E36 Please send encrypted e-mail if possible http://sergiodj.net/
Re: Unbalbe to gdb core file generted from customer machine on local host
Thank you very much Sergio. You have answered my question. I was suspecting that the difference in version of library is the cause. Best Regards, James Chen IBM From: Sergio Durigan Junior sergi...@sergiodj.net To: James C Chen/Ontario/IBM@IBMCA Cc: gdb@gnu.org Date: 11/17/2014 03:28 PM Subject:Re: Unbalbe to gdb core file generted from customer machine on local host On Sunday, November 16 2014, James C. Chen wrote: Hi there, Thanks for providing gdb support. Hello, James. First of all, this mailing list is not used by the GDB project anymore. You should write to g...@sourceware.org, which is the official mailing list for the project. I created an executable binary (x86-64) on my local Linux server and let it run on customer Linux server. Later on, my binary generated a core file. I tried to gdb the core file on my local Linux server, but I can not get correct stack trace. It is not clear whether you compiled your binary with debug information or not. Have you used the -g flag on GCC? Also, if your binary is linked against shared libraries, you have to make sure that the debug information for those libraries is also installed. And you have to make sure that the versions of the libraries installed on your machine and on your client's machine are the same. (gdb) bt full #0 0x003a5d07288e in ?? () No symbol table info available. #1 0x1b36d548 in ?? () No symbol table info available. #2 0x1b2fab70 in ?? () No symbol table info available. #3 0x7fff60c30a9c in ?? () No symbol table info available. #4 0x1b36d548 in ?? () No symbol table info available. #5 0x1aafc860 in ?? () No symbol table info available. #6 0x2230 in ?? () from /lib64/libm.so.6 No symbol table info available. #7 0x545194cc0221 in ?? () No symbol table info available. #8 0x in ?? () No symbol table info available. These errors happen when GDB does not have access to the debug information of your binary/libraries. But when I gdb the core file on customer server where the core file was created, I am able to get correct stack trace. Question, do I have to gdb the core file on the customer server to get correct stack trace? Any way to gdb the core file on my local server? If you are using the same versions of the same programs/libraries, then you will be able to inspect the corefile in your machine without problems. -- Sergio GPG key ID: 0x65FC5E36 Please send encrypted e-mail if possible http://sergiodj.net/
