Linux如何观察文件被哪个进程写数据

CentOS7 ASP站长网在某一段时间监控报警磁盘使用率达99%，由于监控属于概要形式信息，没有快照信息的监控（能发现某进程的I/O，CPU消耗情况），所以需要在服务器上去定时执行统计命令获取快照信息。
需要通过iostat -dx -k去查看avgqu-sz、await、svctm、%util；
sar -u查看%iowait、%user；
pidstat -d 查看进程I/O读写的快照信息
 
步骤
生成统计信息文件
cat>/tmp/at_task.sh<<EOF
pidstat -d 2 >/tmp/pidstat_\`date +%F_%T\`.log 2>& 1 &
sar -u 2  >/tmp/sar_\`date +%F_%T\`.log 2>& 1 &
while [ 1 ];do echo -n \`date +%T\` >>/tmp/iostat_\`date +%F\` 2>& 1  && iostat -dx -k 1 1 >>/tmp/iostat_\`date +%F\` 2>& 1; sleep 2; done &
EOF
在while循环中使用iostat的原因是要输出date +%T时间，不然只有数据，没有时间信息也没有什么用
 
使用at 命令定时执行
at 15:14 today -f /tmp/at_task.sh
出现错误
 
Can't open /var/run/atd.pid to signal atd. No atd running?
 
重启atd服务
 
service atd restart
 
重新开启at定时任务
 
at 15:14 today -f /tmp/at_task.sh
job 2 at Wed Mar 13 15:14:00 2019
 
得到如下快照信息
iostat
 
15:13:35Linux 3.10.0-862.14.4.el7.x86_64 (ip-xxxxx)     03/13/2019      _x86_64_        (4 CPU)
 
Device:         rrqm/s   wrqm/s     r/s     w/s    rkB/s    wkB/s avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
vda               0.12     0.07   17.31   19.41   580.79    90.52    36.57     0.09    2.39    4.42    0.57   0.72   2.63
scd0              0.00     0.00    0.00    0.00     0.00     0.00     6.00     0.00    0.28    0.28    0.00   0.25   0.00
sar
 
03:14:00 PM     CPU     %user     %nice   %system   %iowait    %steal     %idle
03:14:02 PM     all      0.25      0.00      0.38      0.00      0.00     99.37
03:14:04 PM     all      1.25      0.13      0.63      0.00      0.00     97.99
03:14:06 PM     all      0.25      0.13      0.50      0.00      0.00     99.12
03:14:08 PM     all      0.50      0.00      0.50      0.63      0.00     98.37
pidstat
 
03:14:00 PM   UID       PID   kB_rd/s   kB_wr/s kB_ccwr/s  Command
03:14:02 PM  5700      9089      0.00      6.00      0.00  uxxx
03:14:02 PM  5700      9140      0.00      6.00      0.00  uxxx
03:14:02 PM  5700      9292      0.00     10.00      0.00  uxxx
03:14:02 PM     0     18084      0.00      2.00      0.00  bash
kill 掉收集信息的命令
 
ps -ef | egrep 'iostat|sar|pidstat|while' | grep -v grep | awk '{print $2}' | xargs -l kill
*但ps -ef | egrep 命令没有获取到while循环的pid，不kill掉该while循环，就会一直对/tmp/iostat_2019-03-13写数据-_-*
 
通过lsof 没有定位到打开文件的进程
 
lsof /tmp/iostat_2019-03-13
[root@ip-10-186-60-117 ~]#
[root@ip-10-186-60-117 ~]#
通过lsof 可以定位到打开mysql-error.log的进程
 
lsof /opt/mysql/data/5690/mysql-error.log
COMMAND   PID                USER   FD   TYPE DEVICE SIZE/OFF     NODE NAME
mysqld  12858 actiontech-universe    1w   REG  253,1     6345 20083533 /opt/mysql/data/5690/mysql-error.log
mysqld  12858 actiontech-universe    2w   REG  253,1     6345 20083533 /opt/mysql/data/5690/mysql-error.log
可见，某进程只有一只持有某文件的inode，才可以通过lsof查看文件在被那些进程使用
 
获取写文件的进程号
安装sysemtap
 
yum -y install systemtap
 
SystemTap 是对 Linux 内核监控和跟踪的工具
利用systemtap中的inodewatch.stp工具来查找写文件的进程号
 
得到文件的inode
 
stat -c '%i' /tmp/iostat_2019-03-13
4210339
获取文件所在设备的major，minor
 
ls -al /dev/vda1
brw-rw---- 1 root disk 253, 1 Jan 30 13:57 /dev/vda1
得到写文件的pid
 
stap /usr/share/systemtap/examples/io/inodewatch.stp 253 1  4210339
 
Checking "/lib/modules/3.10.0-862.14.4.el7.x86_64/build/.config" failed with error: No such file or directory
Incorrect version or missing kernel-devel package, use: yum install kernel-devel-3.10.0-862.14.4.el7.x86_64
根据系统内核版本在kernel-devel rpm build for : Scientific Linux 7网站上下载相应的kernal-devel包
 
wget ftp://ftp.pbone.net/mirror/ftp.scientificlinux.org/linux/scientific/7.2/x86_64/updates/security/kernel-devel-3.10.0-862.14.4.el7.x86_64.rpm
 
rpm -ivh kernel-devel-3.10.0-862.14.4.el7.x86_64.rpm
再次执行stap
 
stap /usr/share/systemtap/examples/io/inodewatch.stp 253 1 4210339
......
Missing separate debuginfos, use: debuginfo-install kernel-3.10.0-862.14.4.el7.x86_64
Pass 2: analysis failed. [man error::pass2]
Number of similar error messages suppressed: 2.
 
安装debuginfo kernal
 
debuginfo-install kernel-3.10.0-862.14.4.el7.x86_64
  Verifying  : kernel-debuginfo-common-x86_64-3.10.0-862.14.4.el7.x86_64                        1/3
  Verifying  : yum-plugin-auto-update-debug-info-1.1.31-50.el7.noarch                           2/3
  Verifying  : kernel-debuginfo-3.10.0-862.14.4.el7.x86_64                                      3/3
 
Installed:
  kernel-debuginfo.x86_64 0:3.10.0-862.14.4.el7                                                     
  yum-plugin-auto-update-debug-info.noarch 0:1.1.31-50.el7                                          
 
Dependency Installed:
  kernel-debuginfo-common-x86_64.x86_64 0:3.10.0-862.14.4.el7    
再次执行stap
 
stap /usr/share/systemtap/examples/io/inodewatch.stp 253 1  4210339
ERROR: module version mismatch (#1 SMP Tue Sep 25 14:32:52 CDT 2018 vs #1 SMP Wed Sep 26 15:12:11 UTC 2018), release 3.10.0-862.14.4.el7.x86_64
WARNING: /usr/bin/staprun exited with status: 1
添加 -v查看详细报错
stap -v  /usr/share/systemtap/examples/io/inodewatch.stp 253 1  4210339
Pass 1: parsed user script and 471 library scripts using 240276virt/41896res/3368shr/38600data kb, in 300usr/20sys/320real ms.
Pass 2: analyzed script: 2 probes, 12 functions, 8 embeds, 0 globals using 399436virt/196284res/4744shr/197760data kb, in 1540usr/560sys/2106real ms.
Pass 3: using cached /root/.systemtap/cache/f5/stap_f5c0cd780e8a2cac973c9e3ee69fba0c_7030.c
Pass 4: using cached /root/.systemtap/cache/f5/stap_f5c0cd780e8a2cac973c9e3ee69fba0c_7030.ko
Pass 5: starting run.
ERROR: module version mismatch (#1 SMP Tue Sep 25 14:32:52 CDT 2018 vs #1 SMP Wed Sep 26 15:12:11 UTC 2018), release 3.10.0-862.14.4.el7.x86_64
WARNING: /usr/bin/staprun exited with status: 1
Pass 5: run completed in 0usr/10sys/38real ms.
Pass 5: run failed.  [man error::pass5]
修改
 
vim /usr/src/kernels/3.10.0-862.14.4.el7.x86_64/include/generated/compile.h
 
#define UTS_VERSION "#1 SMP Tue Sep 25 14:32:52 CDT 2018"
改为
#define UTS_VERSION "#1 SMP Wed Sep 26 15:12:11 UTC 2018"
 
rm -rf  /root/.systemtap/cache/f5/stap_f5c0cd780e8a2cac973c9e3ee69fba0c_7030*
参考SystemTap - 安装 https://www.linuxidc.com/Linux/2019-03/157818.htm
 
再次执行
 
stap /usr/share/systemtap/examples/io/inodewatch.stp 253 1 4210339
 
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4671) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4677) vfs_write 0xfd00001/4210339
iostat(4683) vfs_write 0xfd00001/4210339
............
可见已经得到了写/tmp/iostat_date +%F 文件的进程号，但进程号一直在打印出来，因为后台进程iostat -dx -m 的在while循环中的，每隔sleep 2s 后就会执行一次iostat 产生新的pid。
***那要怎样才能让iostat -dx -m 停止写/tmp/iostat_date +%F 文件了？除了重启大法好 $_$***
 
rm -rf 也不能终止后台的while iostat进程写文件，删除了文件后，while循环又会生成新的文件
 
rm -rf  /tmp/iostat_2019-03-1*
 
stat /tmp/iostat_2019-03-1*
  File: ‘/tmp/iostat_2019-03-13’
  Size: 146700      Blocks: 512        IO Block: 4096   regular file
Device: fd01h/64769d    Inode: 4210339     Links: 1
Access: (0644/-rw-r--r--)  Uid: (    0/    root)   Gid: (    0/    root)
Access: 2019-03-14 16:07:26.211888899 +0800
Modify: 2019-03-14 16:18:17.854019793 +0800
Change: 2019-03-14 16:18:17.854019793 +0800
正确做法
 
cat>/tmp/iostat.sh<<EOF
while [ 1 ];do echo -n \`date +%T\` >>/tmp/iostat_\`date +%F\` 2>& 1  && iostat -dx -m 1 1 >>/tmp/iostat_\`date +%F\` 2>& 1; sleep 2; done &
EOF
 
at  now + 1 minute  today