specify glibc on debugging core dump with gdb - gdb

I have a program that crashed on RHEL 6.5, also I have a coredump file.
I want to debug it on RHEL 6.4, but it has another version of glibc. I put glibc.so.6 from RHEL 6.5 to lib/ directory, but can't set gdb up to use it:
[nskdvlp#srv2-csd-mong01 snmp_int]$ LD_LIBRARY_PATH=/u01/app/maa/snmp_int/lib gdb bin/snmp_int
Reading symbols from /u01/app/maa/snmp_int/bin/snmp_int...done.
(gdb) info sharedlibrary
No shared libraries loaded at this time.
(gdb) set solib-search-path lib
warning: .dynamic section for "/lib64/libc.so.6" is not at the expected address (wrong library or version mismatch?)
Reading symbols from /lib64/libc.so.6...(no debugging symbols found)...done.
Loaded symbols for /lib64/libc.so.6
Reading symbols from /u01/app/maa/snmp_int/lib/libsnmp_int_snmp.so.6.0.5...done.
Loaded symbols for /u01/app/maa/snmp_int/lib/libsnmp_int_snmp.so.6.0.5
(gdb) core core.110358
[New Thread 110358]
[New Thread 110367]
warning: .dynamic section for "/lib64/libc.so.6" is not at the expected address (wrong library or version mismatch?)
warning: Could not load shared library symbols for (0x%0*Zx)
.
Do you need "set solib-search-path" or "set sysroot"?
Reading symbols from /lib64/libdl.so.2...(no debugging symbols found)...done.
Loaded symbols for /lib64/libdl.so.2
(gdb) bt
#0 0x0000003e93a32925 in ?? ()
#1 0x0000003e93a3408d in ?? ()
#2 0x0000000000000000 in ?? ()
Other .so libraries are loaded well from lib/

I put glibc.so.6 from RHEL 6.5 to lib/ directory, but can't set gdb up to use it:
There is no such thing as glibc.so.6. You mean libc.so.6.
This is nowhere near sufficient: you need all libraries and the dynamic loader from the system on which the core was produced. See this answer for what you need to do.

Related

Can't see symbols from Erlang NIF library in core file

I'm working on an Erlang wrapper over a 3rd party C library on Ubuntu Linux on x86, so I'm creating a NIF. Sometimes my code (I think) crashes, resulting in a core file. Unfortunately the stacktrace is not really helpful:
(gdb) bt
#0 0x00007fc22229968a in ?? ()
#1 0x0000000060e816d8 in ?? ()
#2 0x0000000007cd48b0 in ?? ()
#3 0x00007fc228031410 in ?? ()
#4 0x00007fc228040b80 in ?? ()
#5 0x00007fc228040c50 in ?? ()
#6 0x00007fc22223de0b in ?? ()
#7 0x0000000000000000 in ?? ()
even though I built my NIF .so file with debug info:
ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, BuildID[sha1]=b70dd1f2450f5c0e9980c8396aaad2e1cd29024c, with debug_info, not stripped
The beam also has debug info:
ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, BuildID[sha1]=e0a5dba6507b8c2b333faebc89fbc6ea2f7263b9, for GNU/Linux 3.2.0, with debug_info, not stripped
However, info sharedlibrary doesn't show neither the NIF nor the 3rd party lib:
(gdb) info sharedlibrary
From To Syms Read Shared Object Library
0x00007fc28942ed50 0x00007fc289432004 Yes /lib/x86_64-linux-gnu/libgtk3-nocsd.so.0
0x00007fc289429220 0x00007fc28942a179 Yes /lib/x86_64-linux-gnu/libdl.so.2
0x00007fc2892e83c0 0x00007fc28938ef18 Yes /lib/x86_64-linux-gnu/libm.so.6
0x00007fc2892b76a0 0x00007fc2892c517c Yes /lib/x86_64-linux-gnu/libtinfo.so.6
0x00007fc28928dae0 0x00007fc28929d4d5 Yes /lib/x86_64-linux-gnu/libpthread.so.0
0x00007fc2890b9630 0x00007fc28922e20d Yes /lib/x86_64-linux-gnu/libc.so.6
0x00007fc289657100 0x00007fc289679674 Yes (*) /lib64/ld-linux-x86-64.so.2
0x00007fc24459c040 0x00007fc2445ab8ad Yes /home/nar/otp/23.3.4.2/lib/crypto-4.9.0.2/priv/lib/crypto.so
0x00007fc2239e3000 0x00007fc223b7c800 Yes (*) /lib/x86_64-linux-gnu/libcrypto.so.1.1
0x00007fc2896500e0 0x00007fc28965028c Yes /home/nar/otp/23.3.4.2/lib/crypto-4.9.0.2/priv/lib/crypto_callback.so
0x00007fc289649380 0x00007fc28964bc1c Yes /home/nar/otp/23.3.4.2/lib/asn1-5.0.15/priv/lib/asn1rt_nif.so
0x00007fc289638720 0x00007fc28963bd70 Yes /lib/x86_64-linux-gnu/librt.so.1
I found this answer mentioning that "The Erlang VM doesn't load NIF libraries with global symbols exposed". Could this be the reason why I don't see the symbols? Is there a way to tell gdb to look up symbols from my .so file?
I built the Erlang VM with debug enabled (I used kerl to build and set KERL_BUILD_DEBUG_VM to true), then started the erlang with the -debug option. This way some asserts were seem to be enabled in the code, they crashed and that lead to me to the bugs in my code. Since then I don't have the crashes.

GDB reading symbols with "symbol-file" command on a core file

I am trying to analyze segfault on a core file on linux. I am not sure if the following behavior is correct, thus i deliberately caused a segfault using
#include <signal.h>
int main() {
raise(SIGSEGV);
}
the binary is build with debug info i.e.
file mainTestFile
mainTestFile: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/l, for GNU/Linux 3.2.0, with debug_info, not stripped
notice how it does say "with debug_info, not stripped" at the end
when i execute the binary i get a core file generated which is called core-mainTestFile.20474
(In order to generate the core file i hat to set my ulimit to unlimited i.e.
ulimit -c unlimited
)
if i run only the binary under GDB and do backtrace "bt" then i get the segfault and i get all names of the functions involved
printed nicely i.e. notice how the gdb says when starting "reading symbols from ./mainTestFile...done."
gdb ./mainTestFile
GNU gdb (Ubuntu 8.1-0ubuntu3) 8.1.0.20180409-git
....
reading symbols from ./mainTestFile...done.
(gdb) run
Starting program: /src/exe/mainTestFile
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
Program received signal SIGSEGV, Segmentation fault.
__GI_raise (sig=<optimized out>) at ../sysdeps/unix/sysv/linux/raise.c:51
51 ../sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt
#0 __GI_raise (sig=<optimized out>) at ../sysdeps/unix/sysv/linux/raise.c:51
#1 0x0000000000402dad in main (argc=1, argv=0x7fffffffda38) at /src/exe/main.cpp:53
(gdb)
however if i try to anaylise only the core file with gdb like that
gdb -c core-mainTestFile.20474
then i get only question marks
the when i execute "bt" then i do not see the names of the methods, instead i get question marks
(gdb) bt
#0 0x00007f34d8842e97 in ?? ()
#1 0x0000000000000000 in ?? ()
they only workaround i found is to supply the binary directly at the command line then it all gets printed nicely.
even if i try to tell GDB to use the symbols file and point that to the binary file which does have the symbols
i.e.
symbol-file /src/exe/mainTestFile
then GDB says
Reading symbols from /src/exe/mainTestFile...done
and when i execute bt i see the question marks again? Why is that. Is GDB not able to get the symbols out of the binary?
it only works if supply the binary directly on the command like like:
gdb /src/exe/mainTestFile -c core-mainTestFile.20474
my question is should the GDB be able to read symbols of the binary when directly supplying him the binary over the "symbol-file" command or not. Why is this working when supping him directly the binary over the command line, what is the difference?
should the GDB be able to read symbols of the binary when directly supplying him the binary over the "symbol-file" command or not.
In theory, using symbol-file and core-file commands in either order in GDB should be equivalent.
But there is a bug: symbol-file followed by core-file works, and the opposite order doesn't.
Since generally the end-user can always rearrange his commands into the order that works, this has never propagated to the top of any GDB developer's queue of things to fix.
Related bug (but not an exact duplicate).

Where to look when a daemonized version of a program crashes?

I have a C++ program that receives streaming music and plays it. I can run the program via the shell and run with it attached and it runs completely fine. I can stream audio to it and send things to it, but it will not crash. However, when I daemonize it and run the fork code, it will crash unexpectedly after a little bit of streaming. I try to use gdb to debug it, but it doesn't give much output.
./bin/sonar -d ; sleep 1 ; gdb ./bin/sonar $(cat sonar.pid )
GNU gdb (GDB) 7.4.1-debian
Copyright (C) 2012 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "arm-linux-gnueabihf".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /var/sonar/bin/sonar...done.
Attaching to program: /var/sonar/bin/sonar, process 4050
Reading symbols from /usr/lib/arm-linux-gnueabihf/libcofi_rpi.so...(no debugging symbols found)...done.
Loaded symbols for /usr/lib/arm-linux-gnueabihf/libcofi_rpi.so
Reading symbols from /var/sonar/bin/../lib/libboost_system.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_system.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_chrono.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_chrono.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_timer.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_timer.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_iostreams.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_iostreams.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_thread.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_thread.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_log_setup.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_log_setup.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_log.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_log.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_filesystem.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_filesystem.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_atomic.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_atomic.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_date_time.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_date_time.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libboost_regex.so.1.60.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libboost_regex.so.1.60.0
Reading symbols from /var/sonar/bin/../lib/libopus.so.0...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libopus.so.0
Reading symbols from /var/sonar/bin/../lib/libmysql.so.16...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libmysql.so.16
Reading symbols from /var/sonar/bin/../lib/libmysqlpp.so.3...(no debugging symbols found)...done.
Loaded symbols for /var/sonar/bin/../lib/libmysqlpp.so.3
Reading symbols from /usr/lib/arm-linux-gnueabihf/libcrypto.so.1.0.0...(no debugging symbols found)...done.
Loaded symbols for /usr/lib/arm-linux-gnueabihf/libcrypto.so.1.0.0
Reading symbols from /usr/lib/arm-linux-gnueabihf/libssl.so.1.0.0...(no debugging symbols found)...done.
Loaded symbols for /usr/lib/arm-linux-gnueabihf/libssl.so.1.0.0
Reading symbols from /usr/lib/arm-linux-gnueabihf/libasound.so.2...(no debugging symbols found)...done.
Loaded symbols for /usr/lib/arm-linux-gnueabihf/libasound.so.2
Reading symbols from /usr/lib/arm-linux-gnueabihf/libstdc++.so.6...(no debugging symbols found)...done.
Loaded symbols for /usr/lib/arm-linux-gnueabihf/libstdc++.so.6
Reading symbols from /lib/arm-linux-gnueabihf/libm.so.6...Reading symbols from /usr/lib/debug/lib/arm-linux-gnueabihf/libm-2.19.so...done.
done.
Loaded symbols for /lib/arm-linux-gnueabihf/libm.so.6
Reading symbols from /lib/arm-linux-gnueabihf/libgcc_s.so.1...(no debugging symbols found)...done.
Loaded symbols for /lib/arm-linux-gnueabihf/libgcc_s.so.1
Reading symbols from /lib/arm-linux-gnueabihf/libpthread.so.0...Reading symbols from /usr/lib/debug/lib/arm-linux-gnueabihf/libpthread-2.19.so...done.
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/arm-linux-gnueabihf/libthread_db.so.1".
[New Thread 0x749ff450 (LWP 4054)]
[New Thread 0x75358450 (LWP 4053)]
[New Thread 0x75b58450 (LWP 4052)]
done.
Loaded symbols for /lib/arm-linux-gnueabihf/libpthread.so.0
Reading symbols from /lib/arm-linux-gnueabihf/libc.so.6...Reading symbols from /usr/lib/debug/lib/arm-linux-gnueabihf/libc-2.19.so...done.
done.
Loaded symbols for /lib/arm-linux-gnueabihf/libc.so.6
Reading symbols from /lib/arm-linux-gnueabihf/librt.so.1...Reading symbols from /usr/lib/debug/lib/arm-linux-gnueabihf/librt-2.19.so...done.
done.
Loaded symbols for /lib/arm-linux-gnueabihf/librt.so.1
Reading symbols from /lib/arm-linux-gnueabihf/libbz2.so.1.0...(no debugging symbols found)...done.
Loaded symbols for /lib/arm-linux-gnueabihf/libbz2.so.1.0
Reading symbols from /lib/arm-linux-gnueabihf/libz.so.1...(no debugging symbols found)...done.
Loaded symbols for /lib/arm-linux-gnueabihf/libz.so.1
Reading symbols from /lib/arm-linux-gnueabihf/libdl.so.2...Reading symbols from /usr/lib/debug/lib/arm-linux-gnueabihf/libdl-2.19.so...done.
done.
Loaded symbols for /lib/arm-linux-gnueabihf/libdl.so.2
Reading symbols from /lib/ld-linux-armhf.so.3...(no debugging symbols found)...done.
Loaded symbols for /lib/ld-linux-armhf.so.3
Reading symbols from /lib/arm-linux-gnueabihf/libnss_files.so.2...Reading symbols from /usr/lib/debug/lib/arm-linux-gnueabihf/libnss_files-2.19.so...done.
done.
Loaded symbols for /lib/arm-linux-gnueabihf/libnss_files.so.2
0x763e1d14 in epoll_wait () at ../sysdeps/unix/syscall-template.S:81
81 ../sysdeps/unix/syscall-template.S: No such file or directory.
(gdb) c
Continuing.
Program received signal SIGABRT, Aborted.
[Switching to Thread 0x75358450 (LWP 4053)]
0x7633df70 in raise () from /lib/arm-linux-gnueabihf/libc.so.6
(gdb) bt
#0 0x7633df70 in raise () from /lib/arm-linux-gnueabihf/libc.so.6
#1 0x7633f324 in abort () from /lib/arm-linux-gnueabihf/libc.so.6
#2 0x7656eb5c in __gnu_cxx::__verbose_terminate_handler() () from /usr/lib/arm-linux-gnueabihf/libstdc++.so.6
#3 0x7656c9a0 in ?? () from /usr/lib/arm-linux-gnueabihf/libstdc++.so.6
#4 0x7656c9a0 in ?? () from /usr/lib/arm-linux-gnueabihf/libstdc++.so.6
Backtrace stopped: previous frame identical to this frame (corrupt stack?)
(gdb) c
Continuing.
[Thread 0x762ac000 (LWP 4050) exited]
[Thread 0x75b58450 (LWP 4052) exited]
[Thread 0x749ff450 (LWP 4054) exited]
Program terminated with signal SIGABRT, Aborted.
The program no longer exists.
(gdb)
The daemonizing code that I use:
// Daemonize
LOG_INFO( *logger ) << "Daemonizing.";
// Sig stuff
pid_t pid, sid;
// Start the forky pig
if( (pid = fork())<0 )
{
LOG_ERROR( *logger ) << "Error forking: " << strerror( errno );
return( EXIT_FAILURE );
}
// The parent can go home
if( pid>0 )
return( EXIT_SUCCESS );
// Make ourselves emancipate
if( (sid = setsid())<0 )
{
LOG_ERROR( *logger ) << "setsid error: " << strerror( errno );
return( EXIT_FAILURE );
}
// Ignore sigpipes
signal( SIGPIPE, SIG_IGN );
// Close the file descriptors to detach
close( STDIN_FILENO );
close( STDOUT_FILENO );
close( STDERR_FILENO );
If the program works when it's not daemonized and crashes when it is, there is obviously some difference in the way that it runs. So my question really is, where can I look for this particular point of failure?
Here is the same error on a 64-bit intel arch processor:
Program received signal SIGABRT, Aborted.
[Switching to Thread 0x7f80777b8700 (LWP 10314)]
0x00007f807baa7297 in __GI_raise (sig=sig#entry=6) at ../sysdeps/unix/sysv/linux/raise.c:55
55 ../sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt
#0 0x00007f807baa7297 in __GI_raise (sig=sig#entry=6) at ../sysdeps/unix/sysv/linux/raise.c:55
#1 0x00007f807baa862a in __GI_abort () at abort.c:89
#2 0x00007f807c5a700d in __gnu_cxx::__verbose_terminate_handler() () from /usr/lib/gcc/x86_64-pc-linux-gnu/4.9.3/libstdc++.so.6
#3 0x00007f807c5a4e96 in ?? () from /usr/lib/gcc/x86_64-pc-linux-gnu/4.9.3/libstdc++.so.6
#4 0x00007f807c5a4ee1 in std::terminate() () from /usr/lib/gcc/x86_64-pc-linux-gnu/4.9.3/libstdc++.so.6
#5 0x00007f807c600c21 in ?? () from /usr/lib/gcc/x86_64-pc-linux-gnu/4.9.3/libstdc++.so.6
#6 0x00007f807be16324 in start_thread (arg=0x7f80777b8700) at pthread_create.c:333
#7 0x00007f807bb5bf6d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:109
there is a difference in running a process in an interactive environment or as a daemon. When you run a process from the command line, then the process is inheriting all resources from the interactive environment like Paths, defined in environment PATH variables, or even the working directory, which is, in an interactive environment, the actual directory from where a process is started.
Without knowing your system or environment, one of the mistakes which are often done is trying to start background process, daemons, for example from a crontab, but to forget to give the information a process needs to work. When a process is running as a daemon, without knowing the circumstances of your work flow, for example, it i necessary to give the process the environment, to mention one, the working directory. I do it for example with "chdir("/");", so the forked child will know, on what working directory he is active.
Depending on your other resources, you should give the daemon all the information, it needs to fulfil its job, e.g. if it starts some process, for example the PATH to the process and so on.
regards
Turns out that it was a log file crash. I wasn't seeing the throw error because the error was with the logging system itself. Since all of the ties to stdout were cut, nothing was being displayed. It was exactly the issue in this post with Boost.Log:
Boost Log causes crash when trying first log statement (when not Administrator)

gdb no symbol table loaded for core file

There was a core dump produced at the customer end for my application and while looking at the backtrace I don't have the symbols loaded...
(gdb) where
#0 0x000000364c032885 in ?? ()
#1 0x000000364c034065 in ?? ()
#2 0x0000000000000000 in ?? ()
(gdb) bt full
#0 0x000000364c032885 in ?? ()
No symbol table info available.
#1 0x000000364c034065 in ?? ()
No symbol table info available.
#2 0x0000000000000000 in ?? ()
No symbol table info available.
One think I want to mention in here is that the application being used is build with -g option.
To me it seems that the required libraries are not being loaded. I tried to load the libraries manually using the "symbol-file", but this doesn't help.
What could be the possible issue?
No symbol table info available.
Chances are you invoked GDB incorrectly. Don't do this:
gdb core
gdb -c core
Do this instead:
gdb exename core
Also see this answer for what you'll likely have to do to get meaningful crash stack trace for a core from customer's machine.
I was facing a similar issue and later found out that I am missing -g option, Make sure you have compiled the binary with -g.
This happens when you run gdb with path to executable that does not correspond to the one that produced the core dump.
Make sure that you provide gdb with the correct path.
<put an example of correct code or commands here>

Debugging Firebreath using gdb in Linux

The issue is simple. I have my plugin created by Firebreath and would like to debug my *API.cpp file using gdb in Linux.
Yes, i already read this article, and i got this PID:
30227 0.0 0.4 115024 12300 pts/2 Sl+ 00:03 0:00 /usr/lib/chromium/chromium --type=plugin --plugin-path=/home/USER/Documents/mycode/0.1/npMyPlugin.so
When i opened with gdb i always get a similar output like this:
Reading symbols from /usr/lib/libpcrecpp.so.0...(no debugging symbols found)...done.
Loaded symbols for /usr/lib/libpcrecpp.so.0
Reading symbols from /usr/lib/libssh2.so.1...(no debugging symbols found)...done.
Loaded symbols for /usr/lib/libssh2.so.1
Reading symbols from /usr/lib/libssl.so.1.0.0...(no debugging symbols found)...done.
Loaded symbols for /usr/lib/libssl.so.1.0.0
Reading symbols from /usr/lib/libcrypto.so.1.0.0...(no debugging symbols found)...done.
Loaded symbols for /usr/lib/libcrypto.so.1.0.0
Probably not the right process. Any idea?
Thanks in advanced.
Update - Ignoring the message
Thanks guys for your comments. Well i ignored the message and did the following:
(gdb) r
The program being debugged has been started already.
Start it from the beginning? (y or n)
Program not restarted.
(gdb) b 328
No line 328 in the current file.
Make breakpoint pending on future shared library load? (y or n)
So basically gdb load my MyPlugin.cpp and not my MyPluginAPI.cpp. Any idea?
Setting a break point with just a line-number only works if you have exactly one source file.
Use e.g. b MyPluginAPI.cpp:328 or b MyPluginAPI::myMethod() instead.