Skills for analyzing program logs on linux/unix

When our application runs on linux, many colleagues are afraid of command line operations. Therefore, to analyze the program logs on linux, the FTP + Ultraedit method is used. This method is not convenient, not real-time enough, and not professional enough. If hundreds of megabytes or even GB of log files are encountered, the efficiency of this analysis method is also very high, especially when tracing logs need to be implemented.

So I will summarize several common commands used to analyze program logs in linux and their combined use, hoping to help you.

Instructions:

More:

Purpose: This command prints the content of a file from the front to the back to the terminal screen each time.

Usage:

More filename

After the screen information is displayed, the 'space' will flip down the screen, 'Carriage return 'will flip down the line, and 'Q' will exit.

Cat:

Purpose: this command is used to combine the content of multiple files and print them to the standard output. However, the most common usage is to display the content of a file, however, when the file size is large, the screen will scroll quickly and the valid information cannot be viewed. We recommend that you use more.

Usage:

Cat filename

Grep:

Purpose: print the lines in a file or standard input that match the specific mode. This command is the most effective method for retrieving specific information from the log file.

Usage:

1. Basic usage

Grep pattern filename

For example:

Grep "error" catalina. out

You can display all rows containing errors in the tomcat log catalina. out.

2. Use a regular expression

Grep-e "pattern" filename

For example:

Grep-e "[eE] rror" catalina. out

You can display all rows containing error or Error in the tomcat log catalina. out.

3. Common options

-Case Insensitive for I matching

-N: displays the row number in the original file before the row is printed.

-V: This option is used to flip the default grep behavior and print the rows that do not contain the specific mode.

Wc:

Purpose: count the number of lines, words, and segments in the file.

Usage:

Wc filename

For example:

[Service @ localhost threshold] $ wc nohup. out

9761028 76738200 1672741676 nohup. out

The three numbers from top to bottom are the number of rows, the number of words, and the number of bytes.

The commonly used parameter of this command is-l, which can count only the number of rows

[Service @ localhost threshold] $ wc-l nohup. out

9761028 nohup. out

Head:

Purpose: display the first N lines of the file

Usage:

Head-num filename

For example:

Head-100 catalina. out

Display the first 100 rows of catalina. out

Tail:

Purpose: display the last N lines of the file, or display the added content in the file in real time.

Usage:

1. display the last N lines of the file

Tail-num filename

For example:

Tail-100 catalina. out

The last 100 rows of the catalina. out file are displayed.

2. added content in the real-time display file

Tail-f filename

For example:

Tail-f catalina. out

Executing this command does not directly return the command line, but prints newly added content in the log file in real time. This feature is very effective for viewing logs. To terminate the output, press Ctrl + C.