gdb工具介绍
程序运行过程中可能会异常终止或崩溃，系统会把程序挂掉时的内存状态记录下来，写入core文件（coredump），可以通过gdb工具来分析core文件。GDB(GNU Debugger)是Linux下的一款C/C++程序调试工具，通过在命令行中执行相应的命令实现程序的调试。
GDB主要有以下功能：
- 设置断点
- 单步调试
- 查看变量的值
- 动态改变程序的执行环境
- 分析崩溃程序产生的core文件
安装gdb
# yum install gdb gcc -y
# gdb --version
GNU gdb (GDB) Red Hat Enterprise Linux 7.6.1-120.el7
测试程序
# vi test.c
#include<stdio.h>
int main(){
int arr[4] = {1,2,3,4};
int i=0;
for(i=0;i<4;i++){
printf("%d\n",arr[i]);
}
return 0;
}
###编译的时候带上-g,才能用gdb
# gcc -g test.c
# ./a.out
1
2
3
4
gdb用法
# gdb a.out
GNU gdb (GDB) Red Hat Enterprise Linux 7.6.1-120.el7
Copyright (C) 2013 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 "x86_64-redhat-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /root/a.out...done.
(gdb)
### r 命令，让程序跑起来
(gdb) r
Starting program: /root/a.out
1
2
3
4
[Inferior 1 (process 1656) exited normally]
Missing separate debuginfos, use: debuginfo-install glibc-2.17-326.el7_9.3.x86_64
##quit退出
(gdb) quit
gdb最常用的10个命令：
break [file:]functiop
Set a breakpoint at function (in file).
### b break 打断点
函数的地方（函数名字）
在第几行打断点
run [arglist]
Start your program (with arglist, if specified).
### run r 运行程序
bt Backtrace: display the program stack.
print expr
Display the value of an expression.
###print 打印变量
arr[0]
&arr[0]
c Continue running your program (after stopping, e.g. at a breakpoint).
next ###执行一条程序，不进入函数内部
Execute next program line (after stopping); step over any function calls in the line.
edit [file:]function
look at the program line where it is presently stopped.
list [file:]function ###列出源代码的一部分(10行)
type the text of the program in the vicinity of where it is presently stopped.
step ### step s进去某一个具体的函数调试
Execute next program line (after stopping); step into any function calls in the line.
help [name]
Show information about GDB command name, or general information about using GDB.
quit ###退出gdb环境
Exit from GDB.
打断点两种方式：
（1）函数的地方
（2）在第几行打断点
(gdb) b main -->在这个main函数打断点
Breakpoint 1 at 0x400535: file test.c, line 4.
## list查看源代码
(gdb) list
1 #include<stdio.h>
2
3 int main(){
4 int arr[4] = {1,2,3,4};
5 int i=0;
6 for(i=0;i<4;i++){
7 printf("%d\n",arr[i]);
8 }
9 return 0;
10 }
## 在上面的代码第7行打断点
(gdb) b 7
Breakpoint 2 at 0x400561: file test.c, line 7.
## 查看已设置的断点
(gdb) info b
Num Type Disp Enb Address What
1 breakpoint keep y 0x0000000000400535 in main at test.c:4 -->在第4行打了断点
2 breakpoint keep y 0x0000000000400561 in main at test.c:7 -->在第7行打了断点
运行程序：
(gdb) r
Starting program: /root/a.out
Breakpoint 1, main () at test.c:4
4 int arr[4] = {1,2,3,4};
Missing separate debuginfos, use: debuginfo-install glibc-2.17-326.el7_9.3.x86_64
(gdb) n -->往下走一行
5 int i=0;
(gdb) p arr[0] -->打印变量的值
$1 = 1
(gdb) p &arr[0] -->打印变量的arr[0]的地址
$3 = (int *) 0x7fffffffe4b0
(gdb) p arr[1]
$2 = 2
(gdb) p &arr[1]
$4 = (int *) 0x7fffffffe4b4 -->打印变量的arr[1]的地址，发现比arr[0]多了4个字节
(gdb) n
6 for(i=0;i<4;i++){
(gdb) n
Breakpoint 2, main () at test.c:7
7 printf("%d\n",arr[i]);
–步进
# cp test.c test1.c
# vi test1.c
#include<stdio.h>
void hello(){
printf("hello echo~ \n");
}
int main(){
int arr[4] = {1,2,3,4};
int i=0;
for(i=0;i<4;i++){
printf("%d\n",arr[i]);
}
hello();
return 0;
}
# gcc -g test1.c
# gdb ./a.out
(gdb) list
1 #include<stdio.h>
2
3 void hello(){
4 printf("hello echo~ \n");
5 }
6 int main(){
7 int arr[4] = {1,2,3,4};
8 int i=0;
9 for(i=0;i<4;i++){
10 printf("%d\n",arr[i]);
(gdb) list
11 }
12 hello();
13 return 0;
14 }
##一次显示不完整，可以继续list
##在第12行打一个断点
(gdb) b 12
Breakpoint 1 at 0x4005e5: file test1.c, line 12.
##打完断点，让程序跑起来
(gdb) r
Starting program: /root/./a.out
1
2
3
4
Breakpoint 1, main () at test1.c:12
12 hello();
Missing separate debuginfos, use: debuginfo-install glibc-2.17-326.el7_9.3.x86_64
##用s命令，进入函数调用
(gdb) s
hello () at test1.c:4
4 printf("hello echo~ \n");
(gdb) n
hello echo~
5 }
(gdb) n
main () at test1.c:13
13 return 0;
(gdb) n
14 }
(gdb) n
0x00007ffff7a2f555 in __libc_start_main () from /lib64/libc.so.6
gdb的骚操作
（1）通过shell命令调用系统命令
(gdb) shell ls
test1.c test.c a.out
(gdb) shell cat test.c
#include<stdio.h>
int main(){
int arr[4] = {1,2,3,4};
int i=0;
for(i=0;i<4;i++){
printf("%d\n",arr[i]);
}
return 0;
}
（2）日志记录功能
(gdb) set logging on
Copying output to gdb.txt.
(gdb) list
1 #include<stdio.h>
2
3 void hello(){
4 printf("hello echo~ \n");
5 }
6 int main(){
7 int arr[4] = {1,2,3,4};
8 int i=0;
9 for(i=0;i<4;i++){
10 printf("%d\n",arr[i]);
(gdb) list
11 }
12 hello();
13 return 0;
14 }
(gdb) b 12
Breakpoint 1 at 0x4005e5: file test1.c, line 12.
(gdb) r
Starting program: /root/a.out
1
2
3
4
Breakpoint 1, main () at test1.c:12
12 hello();
Missing separate debuginfos, use: debuginfo-install glibc-2.17-326.el7_9.3.x86_64
(gdb) s
hello () at test1.c:4
4 printf("hello echo~ \n");
(gdb) n
hello echo~
5 }
(gdb) n
main () at test1.c:13
13 return 0;
(gdb) n
14 }
(gdb) n
0x00007ffff7a2f555 in __libc_start_main () from /lib64/libc.so.6
(gdb) quit
# cat gdb.txt
（3）watchpoint：观察变量是否发现变化
info watchpoints：查看watchpoint
(gdb) list
1 #include<stdio.h>
2
3 void hello(){
4 printf("hello echo~ \n");
5 }
6 int main(){
7 int arr[4] = {1,2,3,4};
8 int i=0;
9 for(i=0;i<4;i++){
10 printf("%d\n",arr[i]);
(gdb) b 8
Breakpoint 1 at 0x4005b1: file test1.c, line 8.
(gdb) r
Starting program: /root/a.out
Breakpoint 1, main () at test1.c:8
8 int i=0;
Missing separate debuginfos, use: debuginfo-install glibc-2.17-326.el7_9.3.x86_64
(gdb) print &i
$2 = (int *) 0x7fffffffe4cc
(gdb) watch *0x7fffffffe4cc
Hardware watchpoint 2: *0x7fffffffe4cc -->设置了一个观察点
(gdb) info watchpoints
Num Type Disp Enb Address What
2 hw watchpoint keep y *0x7fffffffe4cc
(gdb) n
9 for(i=0;i<4;i++){
(gdb) n
10 printf("%d\n",arr[i]);
(gdb) n
1
9 for(i=0;i<4;i++){
(gdb) n
Hardware watchpoint 2: *0x7fffffffe4cc -->发现变量发生了变化
Old value = 0
New value = 1
0x00000000004005df in main () at test1.c:9
9 for(i=0;i<4;i++){
gdb分析MySQL core文件
core 文件是进程在崩溃时的内存快照，可以用于故障排查和调试。
分析core文件的方法：gdb 二进制文件 core文件
系统打开coredump
####如果core文件没有生成，那么需要查看ulimit限制
###没有生成core文件，要通过ulimit命令进行设置
# ulimit -a
core file size (blocks, -c) 0 -->设置为0不会生成
data seg size (kbytes, -d) unlimited
scheduling priority (-e) 0
file size (blocks, -f) unlimited
pending signals (-i) 31117
max locked memory (kbytes, -l) 64
max memory size (kbytes, -m) unlimited
open files (-n) 1024
pipe size (512 bytes, -p) 8
POSIX message queues (bytes, -q) 819200
real-time priority (-r) 0
stack size (kbytes, -s) 8192
cpu time (seconds, -t) unlimited
max user processes (-u) 31117
virtual memory (kbytes, -v) unlimited
file locks (-x) unlimited
####开启core文件
# ulimit -c unlimited
# ulimit -a
core file size (blocks, -c) unlimited
####配置core路径
echo "1" > /proc/sys/kernel/core_uses_pid
echo 2 > /proc/sys/fs/suid_dumpable
mkdir /data/core && chmod 777 /data/core && echo "/data/core/%e.core.%p" > /proc/sys/kernel/core_pattern
###为了下次系统重启时可以生效，写入配置文件
echo "mysql - core unlimited" >> /etc/security/limits.conf
在/etc/sysctl.conf中，增加配置：
kernel.core_uses_pid = 1
fs.suid_dumpable = 2
kernel.core_pattern=/data/core/%e.core.%p
当core产生时，会在/data/core生成名字为： 程序名+core+PID号
模拟MySQL进程异常
#mysql配置core，配置完重启生效
vi /etc/my.cnf
[mysqld]
core_file
#模拟异常
[root@mysql80:/data/core]# kill -SEGV `pidof mysqld`
[root@mysql80:/data/core]# ls -lh
total 1.2G
-rw------- 1 mysql mysql 2.5G Mar 12 23:57 mysqld.core.2996
解析core文件
gdb加载core文件，使用bt命令查看堆栈回溯，以诊断崩溃原因：
[root@mysql80:/data/core]# gdb /usr/local/mysql/bin/mysqld mysqld.core.2996
GNU gdb (GDB) Red Hat Enterprise Linux 7.6.1-120.el7
Copyright (C) 2013 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 "x86_64-redhat-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /usr/local/mysql-8.0.27-linux-glibc2.12-x86_64/bin/mysqld...Dwarf Error: wrong version in compilation unit header (is 0, should be 2, 3, or 4) [in module /usr/local/mysql-8.0.27-linux-glibc2.12-x86_64/bin/mysqld]
(no debugging symbols found)...done.
[New LWP 1721]
[New LWP 1727]
[New LWP 1726]
[New LWP 1728]
[New LWP 1725]
[New LWP 1729]
[New LWP 1730]
[New LWP 1731]
[New LWP 1732]
[New LWP 1733]
[New LWP 1740]
[New LWP 1734]
[New LWP 1735]
[New LWP 1798]
[New LWP 1736]
[New LWP 1797]
[New LWP 1737]
[New LWP 1796]
[New LWP 1738]
[New LWP 1739]
[New LWP 1795]
[New LWP 1794]
[New LWP 1741]
[New LWP 1742]
[New LWP 1793]
[New LWP 1792]
[New LWP 1763]
[New LWP 1791]
[New LWP 1764]
[New LWP 1790]
[New LWP 1769]
[New LWP 1784]
[New LWP 1770]
[New LWP 1783]
[New LWP 1771]
[New LWP 1779]
[New LWP 1773]
[New LWP 1778]
[New LWP 1777]
[New LWP 1774]
[New LWP 1766]
[New LWP 1775]
[New LWP 1765]
[New LWP 1776]
[New LWP 1762]
[New LWP 1785]
[New LWP 1760]
[New LWP 1759]
[New LWP 1786]
[New LWP 1787]
[New LWP 1788]
[New LWP 1799]
[New LWP 1754]
[New LWP 1753]
[New LWP 1756]
[New LWP 1744]
[New LWP 1746]
[New LWP 1747]
[New LWP 1750]
[New LWP 1751]
[New LWP 1752]
[New LWP 1755]
[New LWP 1757]
[New LWP 1758]
[New LWP 1748]
[New LWP 1745]
[New LWP 1749]
[New LWP 1743]
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
Dwarf Error: wrong version in compilation unit header (is 0, should be 2, 3, or 4) [in module /usr/local/mysql-8.0.27-linux-glibc2.12-x86_64/lib/plugin/component_reference_cache.so]
Core was generated by `/usr/local/mysql/bin/mysqld --defaults-file=/data/3306/my.cnf --user=mysql'.
Program terminated with signal 11, Segmentation fault.
#0 0x00007f0c89a42aa1 in pthread_kill () from /lib64/libpthread.so.0
Missing separate debuginfos, use: debuginfo-install glibc-2.17-326.el7_9.3.x86_64 libaio-0.3.109-13.el7.x86_64 libgcc-4.8.5-44.el7.x86_64 libstdc++-4.8.5-44.el7.x86_64 numactl-libs-2.0.12-5.el7.x86_64
### 多个命令都可以显示程序的调用轨迹： bt, where, info stack, backtrace
(gdb) bt
#0 0x00007f0c89a42aa1 in pthread_kill () from /lib64/libpthread.so.0
#1 0x000000000102909d in handle_fatal_signal ()
#2 <signal handler called>
#3 0x00007f0c87d4dddd in poll () from /lib64/libc.so.6
#4 0x000000000101dd38 in Mysqld_socket_listener::listen_for_connection_event() ()
#5 0x0000000000df44a9 in mysqld_main(int, char**) ()
#6 0x00007f0c87c7c555 in __libc_start_main () from /lib64/libc.so.6
#7 0x0000000000dd68a5 in _start ()
或
(gdb) where
#0 0x00007f0c89a42aa1 in pthread_kill () from /lib64/libpthread.so.0
#1 0x000000000102909d in handle_fatal_signal ()
#2 <signal handler called>
#3 0x00007f0c87d4dddd in poll () from /lib64/libc.so.6
#4 0x000000000101dd38 in Mysqld_socket_listener::listen_for_connection_event() ()
#5 0x0000000000df44a9 in mysqld_main(int, char**) ()
#6 0x00007f0c87c7c555 in __libc_start_main () from /lib64/libc.so.6
#7 0x0000000000dd68a5 in _start ()
从下往上查看这些信息，我们可以分析出导致崩溃的函数和具体位置，进而进行更深层次的排查和调试。
有时候，需要查看所有线程，才能排查问题，info threads查看所有线程
(gdb) info threads
Id Target Id Frame
68 Thread 0x7f82fb39b700 (LWP 3003) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
67 Thread 0x7f82f27fc700 (LWP 3007) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
66 Thread 0x7f82f1ffb700 (LWP 3008) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
65 Thread 0x7f82f17fa700 (LWP 3009) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
64 Thread 0x7f82aa276700 (LWP 3012) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
63 Thread 0x7f82a9a75700 (LWP 3013) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
62 Thread 0x7f82aaa77700 (LWP 3011) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
61 Thread 0x7f82a9274700 (LWP 3014) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
60 Thread 0x7f82a8a73700 (LWP 3015) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
59 Thread 0x7f82a8272700 (LWP 3016) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
58 Thread 0x7f82a7a71700 (LWP 3017) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
57 Thread 0x7f82a7270700 (LWP 3018) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
56 Thread 0x7f82a6a6f700 (LWP 3019) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
55 Thread 0x7f82a526c700 (LWP 3022) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
54 Thread 0x7f82a426a700 (LWP 3024) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
53 Thread 0x7f82a3a69700 (LWP 3025) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
52 Thread 0x7f82a2a67700 (LWP 3027) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
51 Thread 0x7f82a2266700 (LWP 3028) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
50 Thread 0x7f82a1a65700 (LWP 3029) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
49 Thread 0x7f82a1264700 (LWP 3030) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
48 Thread 0x7f829f260700 (LWP 3034) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
47 Thread 0x7f829ea5f700 (LWP 3035) 0x00007f832cc1e644 in __io_getevents_0_4 () from /lib64/libaio.so.1
gdb用来抽丝剥茧某些疑难case的时候非常有用，甚至gdb可以在紧急情况下救你一命，例如，当MySQL数据库连接打满又没有后台线程可以连接到MySQL的时候，你可以通过gdb来修改MySQL的连接数：
gdb -p $(pidof mysqld) -ex “set max_connections=2000” -batch
总结
通过以上步骤，我们可以有效地分析 MySQL 产生的 Core 文件，找出程序崩溃的根本原因。理解这些步骤对于数据库管理员和开发人员是非常重要的，有助于迅速定位问题。
最后修改时间：2025-03-13 09:36:27
「喜欢这篇文章，您的关注和赞赏是给作者最好的鼓励」
关注作者
【版权声明】本文为墨天轮用户原创内容，转载时必须标注文章的来源（墨天轮），文章链接，文章作者等基本信息，否则作者和墨天轮有权追究责任。如果您发现墨天轮中有涉嫌抄袭或者侵权的内容，欢迎发送邮件至：contact@modb.pro进行举报，并提供相关证据，一经查实，墨天轮将立刻删除相关内容。
