貌似出现了Bug

[CodeKirin-dragon]

程序在正常运行过程中没有问题，如果长时间运行会出现占用大量CPU的情况，初步怀疑是PHP代码出现死循环，排查下来无果，请问要怎么排查？

看文档尝试使用 GDB 调试
因为不会使用，下面是堆栈的打印信息。还请各位大佬帮忙看一下。其他的进程呀打印了下面的输出都是类似的。

(gdb) bt
#0  0x00007f477ab0d91e in __libc_recv (fd=272, buf=0x7f47142cc358, len=1, flags=0)
    at ../sysdeps/unix/sysv/linux/recv.c:28
#1  0x00007f476ff1ca49 in recv (__flags=0, __n=1, __buf=0x7f47142cc358, __fd=<optimized out>)
    at /usr/include/x86_64-linux-gnu/bits/socket2.h:44
#2  swoole::network::Socket::recv (this=0x55fa85b73ef0, __buf=0x7f47142cc358, __n=1, 
    __flags=__flags@entry=0) at /root/swoole/swoole-src-4.8.13/src/network/socket.cc:709
#3  0x00007f476ff022e3 in swoole::coroutine::Socket::recv_all (this=0x55fa851644a0, __buf=<optimized out>, 
    __buf@entry=0x7f47142cc358, __n=<optimized out>)
    at /root/swoole/swoole-src-4.8.13/src/coroutine/socket.cc:1026
#4  0x00007f476fee12b2 in swoole_socket_coro_recv (type=SOCKET_RECV_ALL, return_value=0x7f46d4e09da0, 
    execute_data=0x7f46d4e09e10) at /root/swoole/swoole-src-4.8.13/ext-src/swoole_socket_coro.cc:1299
#5  zim_swoole_socket_coro_recvAll (execute_data=0x7f46d4e09e10, return_value=0x7f46d4e09da0)
    at /root/swoole/swoole-src-4.8.13/ext-src/swoole_socket_coro.cc:1328
#6  0x000055fa8022cac0 in execute_ex ()
#7  0x00007f476fe68536 in swoole::PHPCoroutine::main_func (arg=0x7fff79974dd0)
    at /root/swoole/swoole-src-4.8.13/ext-src/swoole_coroutine.cc:801
#8  0x00007f476fef6ebd in std::function<void (void*)>::operator()(void*) const (__args#0=<optimized out>, 
    this=0x55fa850eafa0) at /usr/include/c++/7/bits/std_function.h:706
---Type <return> to continue, or q <return> to quit---

5. What version of Swoole are you using (show your php --ri swoole)?

root@S58-181:~# php --ri swoole

swoole

Swoole => enabled
Author => Swoole Team <team@swoole.com>
Version => 4.8.13
Built => Mar 27 2023 23:02:04
coroutine => enabled with boost asm context
epoll => enabled
eventfd => enabled
signalfd => enabled
cpu_affinity => enabled
spinlock => enabled
rwlock => enabled
openssl => OpenSSL 1.1.1  11 Sep 2018
dtls => enabled
curl-native => enabled
pcre => enabled
zlib => 1.2.11
mutex_timedlock => enabled
pthread_barrier => enabled
futex => enabled
async_redis => enabled

Directive => Local Value => Master Value
swoole.enable_coroutine => On => On
swoole.enable_library => On => On
swoole.enable_preemptive_scheduler => Off => Off
swoole.display_errors => On => On
swoole.use_shortname => Off => Off
swoole.unixsock_buffer_size => 8388608 => 8388608

7. What is your machine environment used (show your uname -a & php -v & gcc -v) ?

root@S58-181:~# uname -a
Linux S58-181 4.15.0-20-generic #21-Ubuntu SMP Tue Apr 24 06:16:15 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
root@S58-181:~# php -v
PHP 8.0.11 (cli) (built: Sep 25 2021 08:23:49) ( NTS )
Copyright (c) The PHP Group
Zend Engine v4.0.11, Copyright (c) Zend Technologies
root@S58-181:~# gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/7/lto-wrapper
OFFLOAD_TARGET_NAMES=nvptx-none
OFFLOAD_TARGET_DEFAULT=1
Target: x86_64-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Ubuntu 7.5.0-3ubuntu1~18.04' --with-bugurl=file:///usr/share/doc/gcc-7/README.Bugs --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --prefix=/usr --with-gcc-major-version-only --program-suffix=-7 --program-prefix=x86_64-linux-gnu- --enable-shared --enable-linker-build-id --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --libdir=/usr/lib --enable-nls --enable-bootstrap --enable-clocale=gnu --enable-libstdcxx-debug --enable-libstdcxx-time=yes --with-default-libstdcxx-abi=new --enable-gnu-unique-object --disable-vtable-verify --enable-libmpx --enable-plugin --enable-default-pie --with-system-zlib --with-target-system-zlib --enable-objc-gc=auto --enable-multiarch --disable-werror --with-arch-32=i686 --with-abi=m64 --with-multilib-list=m32,m64,mx32 --enable-multilib --with-tune=generic --enable-offload-targets=nvptx-none --without-cuda-driver --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu
Thread model: posix
gcc version 7.5.0 (Ubuntu 7.5.0-3ubuntu1~18.04) 
root@S58-181:~# 

[NathanFreeman]

strace -p 进程id 看看进程当前在做什么

[CodeKirin-dragon]

strace -p 进程id 看看进程当前在做什么

openat(AT_FDCWD, "/proc/1077/stat", O_RDONLY) = 412
read(412, "1077 (redis-server) R 1 1077 107"..., 4096) = 363
close(412)                              = 0
wait4(-1, 0x7ffc10b51654, WNOHANG, NULL) = 0
write(111, "*2\r\n$6\r\n103974\r\n*0\r\n", 20) = 20
write(136, "$11\r\nd:48657820;\r\n", 18) = 18
write(393, "*2\r\n$5\r\n29060\r\n*0\r\n", 19) = 19
epoll_wait(5, [{EPOLLIN, {u32=239, u64=239}}, {EPOLLIN, {u32=173, u64=173}}], 10128, 100) = 2
read(239, "*6\r\n$4\r\nSCAN\r\n$6\r\n151011\r\n$5\r\nCO"..., 16384) = 124
read(173, "*6\r\n$4\r\nSCAN\r\n$6\r\n103974\r\n$5\r\nCO"..., 16384) = 124
write(173, "*2\r\n$5\r\n67233\r\n*0\r\n", 19) = 19
write(239, "*2\r\n$6\r\n195735\r\n*0\r\n", 20) = 20
epoll_wait(5, [{EPOLLIN, {u32=243, u64=243}}, {EPOLLIN, {u32=382, u64=382}}, {EPOLLIN, {u32=395, u64=395}}], 10128, 77) = 3
read(243, "*6\r\n$4\r\nSCAN\r\n$5\r\n29060\r\n$5\r\nCOU"..., 16384) = 123
read(382, "*6\r\n$4\r\nSCAN\r\n$5\r\n67233\r\n$5\r\nCOU"..., 16384) = 123
read(395, "*6\r\n$4\r\nSCAN\r\n$6\r\n195735\r\n$5\r\nCO"..., 16384) = 124
write(395, "*2\r\n$1\r\n0\r\n*0\r\n", 15) = 15
write(382, "*2\r\n$6\r\n207461\r\n*0\r\n", 20) = 20
write(243, "*2\r\n$6\r\n233666\r\n*0\r\n", 20) = 20
epoll_wait(5, [{EPOLLIN, {u32=199, u64=199}}], 10128, 47) = 1
read(199, "*2\r\n$3\r\nDEL\r\n$47\r\ncache.file.inf"..., 16384) = 67
write(199, ":0\r\n", 4)                 = 4
epoll_wait(5, [{EPOLLIN, {u32=190, u64=190}}, {EPOLLIN, {u32=384, u64=384}}], 10128, 47) = 2
read(190, "*6\r\n$4\r\nSCAN\r\n$6\r\n207461\r\n$5\r\nCO"..., 16384) = 124
read(384, "*6\r\n$4\r\nSCAN\r\n$6\r\n233666\r\n$5\r\nCO"..., 16384) = 124
write(384, "*2\r\n$6\r\n103974\r\n*0\r\n", 20) = 20
write(190, "*2\r\n$6\r\n151011\r\n*0\r\n", 20) = 20
epoll_wait(5, [{EPOLLIN, {u32=204, u64=204}}, {EPOLLIN, {u32=240, u64=240}}], 10128, 31) = 2
read(204, "*2\r\n$6\r\nEXISTS\r\n$56\r\nswoole.prox"..., 16384) = 79
read(240, "*6\r\n$4\r\nSCAN\r\n$6\r\n103974\r\n$5\r\nCO"..., 16384) = 124
write(240, "*2\r\n$5\r\n67233\r\n*0\r\n", 19) = 19
write(204, ":0\r\n", 4)                 = 4
epoll_wait(5, [{EPOLLIN, {u32=355, u64=355}}, {EPOLLIN, {u32=393, u64=393}}], 10128, 23) = 2
read(355, "*6\r\n$4\r\nSCAN\r\n$6\r\n151011\r\n$5\r\nCO"..., 16384) = 124
read(393, "*6\r\n$4\r\nSCAN\r\n$5\r\n67233\r\n$5\r\nCOU"..., 16384) = 123
getpid()                                = 1077
openat(AT_FDCWD, "/proc/1077/stat", O_RDONLY) = 412
read(412, "1077 (redis-server) R 1 1077 107"..., 4096) = 362
close(412)                              = 0
wait4(-1, 0x7ffc10b51654, WNOHANG, NULL) = 0
write(393, "*2\r\n$6\r\n207461\r\n*0\r\n", 20) = 20
write(355, "*2\r\n$6\r\n195735\r\n*0\r\n", 20) = 20
epoll_wait(5, [{EPOLLIN, {u32=27, u64=27}}, {EPOLLIN, {u32=419, u64=419}}, {EPOLLIN, {u32=175, u64=175}}], 10128, 100) = 3
read(27, "*6\r\n$4\r\nSCAN\r\n$1\r\n0\r\n$5\r\nCOUNT\r\n"..., 16384) = 119
read(419, "*2\r\n$3\r\nGET\r\n$36\r\ncache.system.n"..., 16384) = 56
read(175, "*2\r\n$3\r\nGET\r\n$31\r\ncache.system.n"..., 16384) = 51
write(175, "$11\r\nd:48657820;\r\n", 18) = 18
write(419, "$23\r\ns:15:\"181497319767611\";\r\n", 30) = 30
write(27, "*2\r\n$5\r\n29060\r\n*0\r\n", 19) = 19
epoll_wait(5, [{EPOLLIN, {u32=243, u64=243}}, {EPOLLIN, {u32=383, u64=383}}], 10128, 89) = 2
read(243, "*6\r\n$4\r\nSCAN\r\n$6\r\n207461\r\n$5\r\nCO"..., 16384) = 124
read(383, "*6\r\n$4\r\nSCAN\r\n$6\r\n195735\r\n$5\r\nCO"..., 16384) = 124
write(383, "*2\r\n$1\r\n0\r\n*0\r\n", 15) = 15
write(243, "*2\r\n$6\r\n151011\r\n*0\r\n", 20) = 20
epoll_wait(5, [{EPOLLIN, {u32=156, u64=156}}, {EPOLLIN, {u32=884, u64=884}}], 10128, 70) = 2
read(156, "*6\r\n$4\r\nSCAN\r\n$5\r\n29060\r\n$5\r\nCOU"..., 16384) = 123
read(884, "*4\r\n$5\r\nSETEX\r\n$31\r\ncache.system"..., 16384) = 83
write(884, "+OK\r\n", 5)                = 5
write(156, "*2\r\n$6\r\n233666\r\n*0\r\n", 20) = 20
epoll_wait(5, [{EPOLLIN, {u32=111, u64=111}}, {EPOLLIN, {u32=384, u64=384}}, {EPOLLIN, {u32=410, u64=410}}], 10128, 59) = 3
read(111, "*2\r\n$3\r\nDEL\r\n$47\r\ncache.file.inf"..., 16384) = 67
read(384, "*6\r\n$4\r\nSCAN\r\n$6\r\n151011\r\n$5\r\nCO"..., 16384) = 124
read(410, "*4\r\n$5\r\nSETEX\r\n$36\r\ncache.system"..., 16384) = 100
write(410, "+OK\r\n", 5)                = 5
write(384, "*2\r\n$6\r\n195735\r\n*0\r\n", 20) = 20
write(111, ":0\r\n", 4)                 = 4
epoll_wait(5, [{EPOLLIN, {u32=349, u64=349}}], 10128, 47) = 1
read(349, "*6\r\n$4\r\nSCAN\r\n$6\r\n233666\r\n$5\r\nCO"..., 16384) = 124
^Cstrace: Process 1077 detached

感谢大佬的回复，一直持续输出这东西看情况是好像是业务上某个定时任务，好像一直循环操作redis，我应该知道问题所在了，请问大佬这个可以排查到具体代码吗？虽然这个输入也可以排查但是用到相同的key很多，排查起来比较慢。

[NathanFreeman]

看起来Swoole\Crorutine\Socket::recvAll()每次只使用一个长度为1的缓冲区去获取数据，每次就获取1字节的数据

[matyhtf]

类似的问题，可以使用 strace、gdb（zbacktrace）、perf 等工具进行跟踪调试

php-src 和 swoole-src 提供的 gdbinit 脚本，里面有很多工具可以查看相关的信息