[20140707]linux下使用iotop.txt

生产系统disk io量很大,并且很慢,像通过os命令了解那个进程存在大量io.使用vmstat,dstat,nmon只能了解大概的情况,不能确定那个
进程.

检查安装包发现iotop可以满足需要,我安装的是centos 6.2.

参数如下:
OPTIONS
       --version
              Show the version number and exit

       -h, --help
              Show usage information and exit

       -o, --only
              Only show processes or threads actually doing I/O, instead of showing all processes or threads. This can be dynamically toggled by pressing o.

       -b, --batch
              Turn on non-interactive mode.  Useful for logging I/O usage over time.

       -n NUM, --iter=NUM
              Set the number of iterations before quitting (never quit by default).  This is most useful in non-interactive mode.

       -d SEC, --delay=SEC
              Set the delay between iterations in seconds (1 second by default).  Accepts non-integer values such as 1.1 seconds.

       -p PID, --pid=PID
              A list of processes/threads to monitor (all by default).

       -u USER, --user=USER
              A list of users to monitor (all by default)

       -P, --processes
              Only show processes. Normally iotop shows all threads.

       -a, --accumulated
              Show accumulated I/O instead of bandwidth. In this mode, iotop shows the amount of I/O processes have done since iotop started.

       -k, --kilobytes
              Use kilobytes instead of a human friendly unit. This mode is useful when scripting the batch mode of iotop. Instead of choosing the most appropriate unit iotop will display all sizes in kilobytes.

       -t, --time
              Add a timestamp on each line (implies --batch). Each line will be prefixed by the current time.

       -q, --quiet
              suppress some lines of header (implies --batch). This option can be specified up to three times to remove header lines.
              -q     column names are only printed on the first iteration,
              -qq    column names are never printed,
              -qqq   the I/O summary is never printed.

常用参数:
iotop -o -u oracle -a -k

其中-a 是累积IO.

这样很容易知道那个进程消耗的IO很大.

#  lsof -i -Pn  | grep