The difference between shell redirection &>file, 2>&1, 1>&2

Transferred from: http://www.360doc.com/content/13/0523/16/7044580_287544243.shtml

on the shell: 0 indicates standard input 1 indicates standard output 2 indicates standard error output > default to standard output redirect, same as 1> 2>&1 means redirecting the standard error output to the standard output. &>file means to redirect both the standard output and the standard error output to file To speak with an example: 1. grep da * 1>&2 2. Rm-f $ (Find/-name core) &>/dev/null The above two examples of & how to understand,& is not put in the background to execute it? Cow's solution: 1.&>file or N>&m are an independent redirect symbol, do not separate to understand. 2. Clarify the differences between file and file descriptors. 3.&>file indicates redirection of standard output and error to file For example: Rm-f $ (Find/-name Core) &>/dev/null,/dev/null is a file, this file is very special, all the things passed to it are discarded. 4.n>&m indicates that the file descriptor n becomes a copy of the output file descriptor m. The advantage of this is that sometimes when you look for files, it is easy to generate useless information, such as:2>/dev/null does not display the standard error output, and when you run certain commands, the error message may be important, so that you can check what is wrong, such as:2>& 1 For example: Note that for ease of understanding, an environment must be set up so that the execute grep da * command will have normal output and error output, and then generate three files using the following command, respectively: grep da * > Greplog1 grep da * > Greplog2 1>&2 grep da * > Greplog3 2>&1//grep da * 2> greplog4 1>&2 Results same #查看greplog1会发现里面只有正常输出内容 #查看greplog2会发现里面什么都没有 #查看greplog3会发现里面既有正常输出内容又有错误输出内容

The difference between shell redirection &>file, 2>&1, 1>&2