Linux下AWS C++ SDK调用PutObject触发std::bad_alloc错误求助

问题背景 #

在Linux环境使用AWS C++ SDK共享库向S3传输文件时触发std::bad_alloc错误，尝试运行AWS官方S3列表示例代码后仍出现相同错误，错误发生在调用AwsDoc::S3::PutObject(bucket_name, object_name, clientConfig);时。

编译命令：

g++ testingAws.cpp  -I . -L  . -laws-cpp-sdk-transfer -laws-cpp-sdk-core -laws-cpp-sdk-s3

运行报错：

terminate called after throwing an instance of 'std::bad_alloc'
what():  std::bad_alloc

GDB堆栈回溯信息：

Backtrace: terminate called after throwing an instance of 'std::bad_alloc' what():
  std::bad_alloc Thread 1 "a.out" received signal SIGABRT, Aborted.
  __pthread_kill_implementation (no_tid=0, signo=6, threadid=140737314626112) at ./nptl/pthread_kill.c:44 44
  ./nptl/pthread_kill.c: No such file or directory.
  (gdb) bt 
  #0  __pthread_kill_implementation (no_tid=0, signo=6, threadid=140737314626112) at ./nptl/pthread_kill.c:44
  #1  __pthread_kill_internal (signo=6, threadid=140737314626112) at ./nptl/pthread_kill.c:78
  #2  __GI___pthread_kill (threadid=140737314626112, signo=signo@entry=6) at ./nptl/pthread_kill.c:89
  #3  0x00007ffff6b2d476 in __GI_raise (sig=sig@entry=6) at ../sysdeps/posix/raise.c:26
  #4  0x00007ffff6b137f3 in __GI_abort () at ./stdlib/abort.c:79
  #5  0x00007ffff6dd5bbe in ?? () from /lib/x86_64-linux-gnu/libstdc++.so.6
  #6  0x00007ffff6de124c in ?? () from /lib/x86_64-linux-gnu/libstdc++.so.6
  #7  0x00007ffff6de12b7 in std::terminate() () from /lib/x86_64-linux-gnu/libstdc++.so.6
  #8  0x00007ffff6de1518 in __cxa_throw () from /lib/x86_64-linux-gnu/libstdc++.so.6
  #9  0x00007ffff6dd57cc in ?? () from /lib/x86_64-linux-gnu/libstdc++.so.6
  #10 0x00007ffff6e7ee6f in std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::_M_assign(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) () from /lib/x86_64-linux-gnu/libstdc++.so.6
  #11 0x00007ffff6e7f22d in std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::operator=(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) () from /lib/x86_64-linux-gnu/libstdc++.so.6
  #12 0x000055555555b58d in Aws::S3::Model::PutObjectRequest::SetBucket(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) ()
  #13 0x000055555555a78b in AwsDoc::S3::PutObject(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, Aws::Client::ClientConfiguration const&) ()
  #14 0x0000000000000000 in ?? () 
  (gdb)

可能的遗漏配置及调试方案 #

1. 强制C++标准版本一致性 #

从堆栈回溯看，错误发生在std::string赋值操作，大概率是C++标准版本不匹配导致的ABI兼容问题：

• 确认编译AWS SDK时使用的C标准（如C11/C++17），编译代码时必须指定相同标准，例如：

  g++ testingAws.cpp -std=c++11 -I . -L . -laws-cpp-sdk-transfer -laws-cpp-sdk-core -laws-cpp-sdk-s3
  

• 确保SDK和你的代码使用相同的标准库（均为libstdc或libc），避免混合使用。

2. 修复共享库加载路径问题 #

编译时指定-L .仅告知编译期库位置，运行时系统可能找不到这些库，间接触发内存异常：

• 运行前设置环境变量指定库路径：

  export LD_LIBRARY_PATH=./:$LD_LIBRARY_PATH
  

• 或编译时添加-Wl,-rpath=./将库路径嵌入可执行文件：

  g++ testingAws.cpp -std=c++11 -I . -L . -Wl,-rpath=./ -laws-cpp-sdk-transfer -laws-cpp-sdk-core -laws-cpp-sdk-s3
  

3. 补全AWS客户端配置 #

确保clientConfig初始化完整，至少设置存储桶所在区域：

Aws::Client::ClientConfiguration clientConfig;
clientConfig.region = "us-east-1"; // 替换为你的S3存储桶区域

未配置区域会导致SDK内部初始化失败，进而触发内存分配错误。

4. 深入调试错误根源 #

• 开启SDK日志：在代码开头添加日志配置，查看SDK内部错误细节：

  Aws::SDKOptions apiOptions;
  Aws::InitAPI(apiOptions);
  Aws::Utils::Logging::ConsoleLogSystem::Initialize(Aws::Utils::Logging::LogLevel::Debug);
  

• GDB断点调试：在PutObject处打断点，检查传入的字符串参数是否有效（如空指针、已释放内存）：

  break AwsDoc::S3::PutObject
  run
  print bucket_name
  print object_name
  

• 检查系统内存：用free -h确认系统内存是否充足，内存耗尽也可能触发std::bad_alloc。

5. 重新编译AWS SDK确保完整性 #

重新编译SDK时，明确指定编译共享库：

cmake -DBUILD_SHARED_LIBS=ON ..
make -j$(nproc)

编译过程中检查是否有警告或错误，确保curl、openssl等依赖库已正确安装。

内容的提问来源于stack exchange，提问作者Abhishekh