Solaris上有Dtrace可以对如Oracle,MySQL进行调试，那Linux也有，那就是systemtap。如Oracle和MySQL出了问题，发现是system的问题，那只能等官方的说法，很被动。如果有这个工具，那就能挖得更深一点，对解决问题很有帮助。
内核引入了一种Kprobe机制，可以用来动态地收集调试和性能信息的工具，是一种非破坏性的工具，用户可以用它跟踪运行中内核任何函数或执行的指令等。systemtap 是利用Kprobe 提供的API来实现动态地监控和跟踪运行中的Linux内核的工具，相比Kprobe，systemtap更加简单，提供给用户简单的命令行接口，以及编写内核指令的脚本语言。
安装这个功能颇费周折，需要安装两个rpm。要想绕过坑，就要执行stap-prep,看下系统需要安装那个版本。 [root@localhost shm]# stap-prep Need to install the following packages: kernel-debuginfo-2.6.32-431.el6.x86_64 ........................................... 可以到http://debuginfo.centos.org/6/x86_64/下找到相应的版本，kernel-debuginfo-common-x86_64-2.6.32-431.el6.x86_64.rpm这个有260M。 [root@localhost shm]# rpm -ivh kernel-debuginfo-common-x86_64-2.6.32-431.el6.x86_64.rpm Preparing... ########################################### [100%] 1:kernel-debuginfo-common########################################### [100%] 安装kernel-debuginfo-2.6.32-431.el6.x86_64.rpm这个也是，如果发现版本不对，系统会提示你需要安装的版本。 [root@localhost shm]# rpm -ivh kernel-debuginfo-2.6.32-431.el6.x86_64.rpm error: Failed dependencies: kernel-debuginfo-common-x86_64 = 2.6.32-431.el6 is needed by kernel-debuginfo-2.6.32-431.el6.x86_64 [root@localhost shm]# rpm -ivh kernel-debuginfo-2.6.32-431.el6.x86_64.rpm Preparing... ########################################### [100%]1:kernel-debuginfo ########################################### [100%]
测试一下是否安装成功
[root@localhost mysql-5.6.14]# stap -v -e 'probe vfs.read {printf("read performed\n"); exit()}' Pass 1: parsed user script and 112 library script(s) using 206604virt/34432res/3212shr/31532data kb, in 30usr/460sys/607real ms. Pass 2: analyzed script: 1 probe(s), 1 function(s), 4 embed(s), 0 global(s) using 303212virt/132016res/4244shr/128140data kb, in 1180usr/1420sys/5393real ms. Pass 3: translated to C into "/tmp/stapOq9UI9/stap_de5b8012cef23568586d0744abf80b8d_1625_src.c" using 303212virt/132372res/4600shr/128140data kb, in 0usr/0sys/7real ms. Pass 4: compiled C into "stap_de5b8012cef23568586d0744abf80b8d_1625.ko" in 1250usr/600sys/2528real ms. Pass 5: starting run. read performed Pass 5: run completed in 10usr/20sys/337real ms.
测试一下，想知道5s探针一下top 20系统命令。 systemtap_t.sh #!/usr/bin/env stap # This script continuously lists the top 20 systemcalls on the system # global syscalls function print_top () { cnt=0 log ("SYSCALL\t\t\t\tCOUNT") foreach ([name] in syscalls-) { printf("%-20s\t\t]\n",name, syscalls[name]) if (cnt++ == 20) break } printf("--------------------------------------\n") delete syscalls } probe kernel.function("sys_*") { syscalls[probefunc()]++ } # print top syscalls every 5 seconds probe timer.ms(5000) { print_top () } [root@localhost root]# stap systemtap_t.sh SYSCALL COUNT sys_times 100 sys_poll 56 sys_read 42 sys_restart_syscall 34 sys_ppoll 28 sys_rt_sigprocmask 15 sys_select 11 sys_close 5 sys_munmap 5 sys_epoll_wait 4 sys_fcntl 4 sys_write 4 sys_open 4 sys_newfstat 4 sys_mmap 4 sys_mmap_pgoff 4 sys_wait4 3 sys_alarm 3 sys_rt_sigpending 3 sys_nanosleep 3 sys_setresuid 2 -------------------------------------- https://sourceware.org/systemtap/ftp/releases 源码下载。
