One .top Role of command

       top The order is Linux Common performance analysis tools , It can display the resource usage of each process in the system in real time , Be similar to Windows Task manager . How to use it is described in detail below .top Is a dynamic display process , That is to say, the current status can be refreshed continuously by pressing the button of the user . If the command is executed in the foreground , It will have exclusive foreground , Until the user terminates the program . To be more precise ,top Command provides real-time status monitoring of system processors . It will be displayed in the system CPU most “ sensitive ” Task list for . The command can be pressed CPU Use . Sort tasks by memory usage and execution time ; And many features of this command can be set through interactive command or in personal customization file .

    Displays information about the processes currently being executed by the system , Including process ID、 Memory usage 、CPU Occupancy rate, etc

Two .top Command related parameters

   top [ Parameters ]

-b  The batch

-c  Show full command

-I  Ignore failure process

-s  Confidentiality mode

-S  Cumulative mode

-i< Time >  Set interval

-u< user name >  Specify user name

-p< Process number >  Specify process

-n< frequency >  Number of cycles

3、 … and .top Use cases of commands

Statistics area ：

The first five lines are the overall statistical information area of the current system . Let’s look at the specific meaning of each line of information .

first line , Task queue information , Same as  uptime  Command execution results , The specific parameters are as follows ：

17:25:49 —  Current system time

up 10 days, 22min—  The system is already running 70 God 16 Hours 44 minute （ During this period, the system did not restart ！）

1 users —  The current is 2 Users log in to the system

load average: 0.00, 0.01, 0.05 — load average The next three numbers are 1 minute 、5 minute 、15 Minute load .

load average The data is every 5 Check the number of active processes per second , The value calculated according to the specific algorithm . If this number is divided by logic CPU The number of , Results higher than 5 It means that the system is overloaded .

The second line ,Tasks —  Mission （ process ）, The details are as follows ：

System now shared 121 A process , Some of them are in operation 1 individual ,120 Sleeping （sleep）,stoped There are 0 individual ,zombie state （ Corpse ） There are 0 individual .

The third line ,cpu State information , The specific attributes are as follows ：

0.1%us —  User space occupancy CPU Percent of .

0.0% sy —  Kernel space footprint CPU Percent of .

0.0% ni —  Change the priority of the process occupied CPU Percent of

99.9% id —  Free CPU percentage

0.0% wa — IO Waiting for occupation CPU Percent of

0.0% hi —  Hard interrupt （Hardware IRQ） Occupy CPU Percent of

0.0% si —  Soft interrupt （Software Interrupts） Occupy CPU Percent of

remarks ： ad locum CPU Usage ratio and windows Different concepts , You need to understand linux Knowledge of system user space and kernel space ！

In the fourth row , Memory status , The details are as follows ：

10074990k total —  Total physical memory

815796k used —  Total memory in use

4967136k free —  Total free memory

4291968k buffers —  Amount of memory cached  

The fifth row ,swap Exchange partition information , The details are as follows ：

4193276k total —  Total exchange area

4193276k used —  Total swap area used

0k free —  k free

4772912k cached —  Total number of swap buffers buffered

remarks ：

Total memory in use in line 4 （used） It refers to the amount of memory controlled by the system kernel , Total free memory （free） It’s the number of cores that haven’t been under its control . Not all the memory included in kernel management is in use , It also includes memory used in the past that can be reused now , The kernel does not return the reusable memory to the free In the middle , So in linux On free Less and less memory , But don’t worry about it .

If you are used to calculating the available memory , Here is an approximate formula ： On the fourth line free +  On the fourth line buffers +  The fifth row cached, According to this formula, the available memory of this server

For memory monitoring , stay top We need to monitor the fifth line all the time swap Swap partitioned used, If this value is constantly changing , It indicates that the kernel is in continuous memory and swap Data exchange of , It’s really out of memory .

Below line 7 ： Processes （ Mission ） Status monitoring of , The project line information is described as follows ：

PID —  process id

USER —  Process owner

PR —  Process priority

NI — nice value . Negative value indicates high priority , A positive value indicates a low priority

VIRT —  Total virtual memory used by the process , Company kb.VIRT=SWAP+RES

RES —  Used by process 、 Physical memory size not swapped out , Company kb.RES=CODE+DATA

SHR —  Shared memory size , Company kb

S —  Process status .D= Uninterrupted sleep  R= function  S= sleep  T= track / stop it  Z= Zombie process

%CPU —  Last updated to now CPU Percentage of time used

%MEM —  Percentage of physical memory used by the process

TIME+ —  Used by process CPU Total time , Company 1/100 second

COMMAND —  Process name （ Command name / Command line ）

Four . Relevant command

Show   Full command

command ：

top -c

top -h

5、 … and . Troubleshooting of thread deadlock command

https://www.jianshu.com/p/8d5782bc596e