Shulou(Shulou.com)06/02 Report--

The requirement is that you can customize the execution command before shutting down or restarting.

Https://opensource.com/life/16/11/running-commands-shutdown-linux

1) use systemd method

In the / lib/systemd/system-shutdown/ directory, store the custom scripts.

Systemd-halt.service

Through ssh's login log

Cat auth.log | grep Accepted

If [!-f'/ root/lastuser.log']; then

Touch / root/lastuser.log

Fi

Mount-oremount,rw /

LOGIN_IP=who am I | awk'{print $NF}'| sed-e's / [()] / / g'

Echo "=" > > / root/lastuser.log

Echo "whoami" > > / root/lastuser.log

Echo "who-m" > > / root/lastuser.log

Echo "who am i" > > / root/lastuser.log

Echo "who am i | awk'{print $NF}'" > > / root/lastuser.log

Echo "whoami date $LOGIN_IP" > > / root/lastuser.log

Mount-oremount,ro /

Whoami= "$(whoami) @ $(echo $SSH_CONNECTION | awk'{print $1}')"

Export PROMPT_COMMAND='RETRN_VAL=$?;logger-p local3.debug "$whoami [$]: $(history 1 | sed" s / ^ [] [0-9] + [] / / ") [$RETRN_VAL]"'

Additionally you can add the line below to your syslog configuration (/ etc/syslog.conf) to redirect the local3 messages to a specific file.

Local3.* / var/log/prompt.log

Today, I looked at a Shell script for system initialization, and there is a command that is configured to log bash commands, as follows:

LOGIN_IP=$ (who am i | awk'{print $NF}')

Export PROMPT_COMMAND=' {msg=$ (history 1 | {read x y; echo $y;}); echo $(date + "% Y-%m-%d% H:%M:%S") [$(whoami) @ $SSH_USER$LOGIN_IP pwd] "$msg" > > / var/log/.history;}'

Where PROMPT_COMMAND is the environment variable of bash, the contents of this variable are executed as a normal bash command, and the execution time is before the bash displays prompt. This is easy to understand, that is, before each command is executed, it is executed first in PROMPT_COMMAND, followed by prompt.

But the msg=$ (history 1 | {read x y; echo $y;}) inside confuses me, and now I'll take it apart.

Let's first understand the history 1 command. Its output is always $COMMAND_Number history 1, which means that its output is always the last command executed.

Bash will first record this command: history 1 | {read x y; echo $y;}? Then history 1 becomes itself without $COMMAND_Number, so the output is the command itself: history 1 | {read x y; echo $y;}?

So the msg=$ (history 1 | {read x y; echo $y;});, the $msg variable always represents the currently executed command?

As far as I understand, the result is obvious, which is that the $msg variable holds the last command executed by bash. But I still don't feel clear, and I still don't understand the combination of history 1 and pipeline.

If the master sees it, please give me some advice on how to understand it.

Https://www.jianshu.com/p/0749ddc6b760

Https://blog.csdn.net/bwlab/article/details/51321308

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.