Linux性能观测——dstat命令详解

dstat

dstat 用来替换vmstat、iostat、netstat、nfsstat和ifstat这些命令的工具,可以说是一个全能的系统信息统计工具。

相关命令

Linux存储性能观测——iostat命令详解
Linux存储性能观测——vmstat命令详解
Linux存储性能观测——free命令详解

安装

安装方式：

yum install -y dstat #Centos
sudo apt-get install dstat #Ubuntu

安装出现的问题以及解决方案：

# Ubuntu下的操作
$ sudo apt-get install dstat
Reading package lists... Done
Building dependency tree
Reading state information... Done
E: Unable to locate package dstat
# Ubuntu下找不到包的时候，只需要执行如下命令即可
$ sudo apt-get update

常用命令形式

$ dstat # 默认参数是-cdngy
You did not select any stats, using -cdngy by default.  
----total-cpu-usage---- -dsk/total- -net/total- ---paging-- ---system--
usr sys idl wai hiq siq| read  writ| recv  send|  in   out | int   csw 
  3   0  96   0   0   0| 436k  879k|   0     0 |   0    72B|4284  3494 
  7   0  92   0   0   0|   0   136k|  52k  327k|   0     0 |7019  4807 
  7   0  91   2   0   0|   0    26M|  32k   83k|   0     0 |6583  5640 
  7   0  93   0   0   0|   0  4086k|  37k  105k|   0     0 |5172  3559 
  7   0  92   0   0   0|   0   363k|  49k  227k|   0     0 |6626  4492 

total-cpu-usage：

usr	sys	idl	wai	hiq	siq
用户时间占比	系统时间占比	空闲时间占比	等待时间占比	硬中断	软中断

dsk/total：

read	writ
磁盘的读总数（KB或者MB）	磁盘的写总数（KB或者MB）

net/total：

recv	send
网络设备接收的数据总数（KB或者MB）	网络设备发送的数据总数（KB或者MB）

paging(系统的分页活动):

in	out
换入次数	换出次数

system（系统统计）：

int	csw
终端次数	上下文切换次数

英文说明

DSTAT(1)                                                                                                                                                                                                                      

NAME
       dstat - versatile tool for generating system resource statistics

SYNOPSIS
       dstat [-afv] [options..] [delay [count]]

DESCRIPTION
       Dstat is a versatile replacement for vmstat, iostat and ifstat. Dstat overcomes some of the limitations and adds some extra features.

       Dstat allows you to view all of your system resources instantly, you can eg. compare disk usage in combination with interrupts from your IDE controller, or compare the network bandwidth numbers directly with the disk
       throughput (in the same interval).

       Dstat also cleverly gives you the most detailed information in columns and clearly indicates in what magnitude and unit the output is displayed. Less confusion, less mistakes, more efficient.

       Dstat is unique in letting you aggregate block device throughput for a certain diskset or network bandwidth for a group of interfaces, ie. you can see the throughput for all the block devices that make up a single
       filesystem or storage system.

       Dstat allows its data to be directly written to a CSV file to be imported and used by OpenOffice, Gnumeric or Excel to create graphs.

       Note
       Users of Sleuthkit might find Sleuthkit’s dstat being renamed to datastat to avoid a name conflict. See Debian bug #283709 for more information.

OPTIONS
       -c, --cpu
              enable cpu stats (system, user, idle, wait, hardware interrupt, software interrupt)

       -C 0,3,total
              include cpu0, cpu3 and total (when using -c/--cpu)

       -d, --disk
              enable disk stats (read, write)

       -D total,hda
              include total and hda (when using -d/--disk)

       -g, --page
              enable page stats (page in, page out)

       -i, --int
              enable interrupt stats

       -I 5,10
              include interrupt 5 and 10 (when using -i/--int)

       -l, --load
              enable load average stats (1 min, 5 mins, 15mins)

       -m, --mem
              enable memory stats (used, buffers, cache, free)

       -n, --net
              enable network stats (receive, send)

       -N eth1,total
              include eth1 and total (when using -n/--net)

       -p, --proc
              enable process stats (runnable, uninterruptible, new)

       -r, --io
              enable I/O request stats (read, write requests)

       -s, --swap
              enable swap stats (used, free)

       -S swap1,total
              include swap1 and total (when using -s/--swap)

       -t, --time
              enable time/date output

       -T, --epoch
              enable time counter (seconds since epoch)

       -y, --sys
              enable system stats (interrupts, context switches)

       --aio  enable aio stats (asynchronous I/O)

       --fs, --filesystem
              enable filesystem stats (open files, inodes)

       --ipc  enable ipc stats (message queue, semaphores, shared memory)

       --lock enable file lock stats (posix, flock, read, write)

       --raw  enable raw stats (raw sockets)

       --socket
              enable socket stats (total, tcp, udp, raw, ip-fragments)

       --tcp  enable tcp stats (listen, established, syn, time_wait, close)

       --udp  enable udp stats (listen, active)

       --unix enable unix stats (datagram, stream, listen, active)

       --vm   enable vm stats (hard pagefaults, soft pagefaults, allocated, free)

       --plugin-name
              enable (external) plugins by plugin name, see PLUGINS for options

       Possible internal stats are
              aio, cpu, cpu24, disk, disk24, disk24old, epoch, fs, int, int24, io, ipc, load, lock, mem, net, page, page24, proc, raw, socket, swap, swapold, sys, tcp, time, udp, unix, vm

       --list list the internal and external plugin names

       -a, --all
              equals -cdngy (default)

       -f, --full
              expand -C, -D, -I, -N and -S discovery lists

       -v, --vmstat
              equals -pmgdsc -D total

       --bits force bits for values expressed in bytes

       --float
              force float values on screen (mutual exclusive with --integer)

       --integer
              force integer values on screen (mutual exclusive with --float)

       --bw, --blackonwhite
              change colors for white background terminal

       --nocolor
              disable colors (implies --noupdate)

       --noheaders
              disable repetitive headers

       --noupdate
              disable intermediate updates when delay > 1

       --output file
              write CSV output to file

       --profile
              show profiling statistics when exiting dstat

PLUGINS
       While anyone can create their own dstat plugins (and contribute them) dstat ships with a number of plugins already that extend its capabilities greatly. Here is an overview of the plugins dstat ships with:

       --battery
              battery in percentage (needs ACPI)

       --battery-remain
              battery remaining in hours, minutes (needs ACPI)

       --cpufreq
              CPU frequency in percentage (needs ACPI)

       --dbus number of dbus connections (needs python-dbus)

       --disk-tps
              per disk transactions per second (tps) stats

       --disk-util
              per disk utilization in percentage

       --dstat
              show dstat cputime consumption and latency

       --dstat-cpu
              show dstat advanced cpu usage

       --dstat-ctxt
              show dstat context switches

       --dstat-mem
              show dstat advanced memory usage

       --fan  fan speed (needs ACPI)

       --freespace
              per filesystem disk usage

       --gpfs GPFS read/write I/O (needs mmpmon)

       --gpfs-ops
              GPFS filesystem operations (needs mmpmon)

       --helloworld
              Hello world example dstat plugin

       --innodb-buffer
              show innodb buffer stats

       --innodb-io
              show innodb I/O stats

       --innodb-ops
              show innodb operations counters

       --lustre
              show lustre I/O throughput

       --memcache-hits
              show the number of hits and misses from memcache

       --mysql5-cmds
              show the MySQL5 command stats

       --mysql5-conn
              show the MySQL5 connection stats

       --mysql5-io
              show the MySQL5 I/O stats

       --mysql5-keys
              show the MySQL5 keys stats

       --mysql-io
              show the MySQL I/O stats

       --mysql-keys
              show the MySQL keys stats

       --net-packets
              show the number of packets received and transmitted

       --nfs3 show NFS v3 client operations

       --nfs3-ops
              show extended NFS v3 client operations

       --nfsd3
              show NFS v3 server operations

       --nfsd3-ops
              show extended NFS v3 server operations

       --ntp  show NTP time from an NTP server

       --postfix
              show postfix queue sizes (needs postfix)

       --power
              show power usage

       --proc-count
              show total number of processes

       --qmail
              show qmail queue sizes (needs qmail)

       --rpc  show RPC client calls stats

       --rpcd show RPC server calls stats

       --sendmail
              show sendmail queue size (needs sendmail)

       --snooze
              show number of ticks per second

       --squid
              show squid usage statistics

       --test show test plugin output

       --thermal
              system temperature sensors

       --top-bio
              show most expensive block I/O process

       --top-bio-adv
              show most expensive block I/O process (incl. pid and other stats)

       --top-childwait
              show process waiting for child the most

       --top-cpu
              show most expensive CPU process

       --top-cpu-adv
              show most expensive CPU process (incl. pid and other stats)

       --top-cputime
              show process using the most CPU time (in ms)

       --top-cputime-avg
              show process with the highest average timeslice (in ms)

       --top-int
              show most frequent interrupt

       --top-io
              show most expensive I/O process

       --top-io-adv
              show most expensive I/O process (incl. pid and other stats)

       --top-latency
              show process with highest total latency (in ms)

       --top-latency-avg
              show process with the highest average latency (in ms)

       --top-mem
              show process using the most memory

       --top-oom
              show process that will be killed by OOM the first

       --utmp show number of utmp connections (needs python-utmp)

       --vmk-hba
              show VMware ESX kernel vmhba stats

       --vmk-int
              show VMware ESX kernel interrupt stats

       --vmk-nic
              show VMware ESX kernel port stats

       --vm-memctl
              show ballooning status inside VMware guests

       --vz-cpu
              show CPU usage per OpenVZ guest

       --vz-io
              show I/O usage per OpenVZ guest

       --vz-ubc
              show OpenVZ user beancounters

       --wifi wireless link quality and signal to noise ratio

ARGUMENTS
       delay is the delay in seconds between each update

       count is the number of updates to display before exiting

       The default delay is 1 and count is unspecified (unlimited)

INTERMEDIATE UPDATES
       When invoking dstat with a delay greater than 1 and without the --noupdate option, it will show intermediate updates, ie. the first time a 1 sec average, the second update a 2 second average, etc. until the delay has been
       reached.

       So in case you specified a delay of 10, the 9 intermediate updates are NOT snapshots, they are averages over the time that passed since the last final update. The end result is that you get a 10 second average on a new
       line, just like with vmstat.

EXAMPLES
       Using dstat to relate disk-throughput with network-usage (eth0), total CPU-usage and system counters:

       dstat -dnyc -N eth0 -C total -f 5
       Checking dstat’s behaviour and the system impact of dstat:

       dstat -taf --debug
       Using the time plugin together with cpu, net, disk, system, load, proc and top_cpu plugins:

       dstat -tcndylp --top-cpu
       this is identical to

       dstat --time --cpu --net --disk --sys --load --proc --top-cpu
       Using dstat to relate cpu stats with interrupts per device:

       dstat -tcyif

BUGS
       Since it is practically impossible to test dstat on every possible permutation of kernel, python or distribution version, I need your help and your feedback to fix the remaining problems. If you have improvements or
       bugreports, please send them to: [1]dag@wieers.com

       Note
       Please see the TODO file for known bugs and future plans.

FILES
       Paths that may contain external dstat_*.py plugins:

       ~/.dstat/
       (path of binary)/plugins/
       /usr/share/dstat/
       /usr/local/share/dstat/

SEE ALSO
   Performance tools
       ifstat(1), iftop(8), iostat(1), mpstat(1), netstat(1), nfsstat(1), nstat, vmstat(1), xosview(1)

   Debugging tools
       htop(1), lslk(1), lsof(8), top(1)

   Process tracing
       ltrace(1), pmap(1), ps(1), pstack(1), strace(1)

   Binary debugging
       ldd(1), file(1), nm(1), objdump(1), readelf(1)

   Memory usage tools
       free(1), memusage, memusagestat, slabtop(1)

   Accounting tools
       dump-acct, dump-utmp, sa(8)

   Hardware debugging tools
       dmidecode, ifinfo(1), lsdev(1), lshal(1), lshw(1), lsmod(8), lspci(8), lsusb(8), smartctl(8), x86info(1)

   Application debugging
       mailstats(8), qshape(1)

   Xorg related tools
       xdpyinfo(1), xrestop(1)

   Other useful info
       collectl(1), proc(5), procinfo(8)

AUTHOR
       Written by Dag Wieers [1]dag@wieers.com

       Homepage at [2]http://dag.wieers.com/home-made/dstat/

       This manpage was initially written by Andrew Pollock [3]apollock@debian.org for the Debian GNU/Linux system.

REFERENCES
       1. dag@wieers.com
          mailto:dag@wieers.com

       2. http://dag.wieers.com/home-made/dstat/
          http://dag.wieers.com/home-made/dstat/

       3. apollock@debian.org
          mailto:apollock@debian.org

  0.7.0                                                                                                       06/15/2010                                                                                                      DSTAT(1)

参考资料

linux manpage dstat