Untitled

以下範例，是先建立一個 loop script，在 Amazon Linux 中，使用 systemctl 命令來運行的簡單範例．

1) 首先創建一個 loop 迴圈的 script 腳本，只是每秒打印時間到特定檔案中。
$ cat /home/ec2-user/sample.sh
#!/bin/sh
while true
do
 date >> /tmp/log.txt
 sleep 1
done
$ chmod 755 /home/ec2-user/sample.sh
$ /home/ec2-user/sample.sh
^C
$ tail -3 /tmp/log.txt
日 5月 22 02:36:26 UTC 2022
日 5月 22 02:36:27 UTC 2022
日 5月 22 02:36:28 UTC 2022

2) 接下來在目錄 /lib/systemd/system/ 下創建一個配置檔，可以參考該目錄下的其他服務配置，來調整你的配置。 記得要重啟載入 $ sudo systemctl daemon-reload。
$ cat /lib/systemd/system/sample.service
[Unit]
Description=sample
After=network.target
[Service]
Restart=always
WorkingDirectory=/usr/bin/
ExecStart=/home/ec2-user/sample.sh
[Install]
WantedBy=multi-user.target
$ sudo systemctl daemon-reload

3) 最後測試一下是否有正常運行。
$ sudo systemctl status sample
● sample.service - sample
 Loaded: loaded (/usr/lib/systemd/system/sample.service; disabled; vendor preset: disabled)
 Active: inactive (dead)
$ sudo systemctl start sample
$ sudo systemctl status sample
● sample.service - sample
 Loaded: loaded (/usr/lib/systemd/system/sample.service; disabled; vendor preset: disabled)
 Active: active (running) since 日 2022–05–22 03:00:28 UTC; 4s ago
 Main PID: 11386 (sample.sh)
 CGroup: /system.slice/sample.service
 ├─11386 /bin/sh /home/ec2-user/sample.sh
 └─11396 sleep 1
5月 22 03:00:28 ip-10–0–200–161.us-west-2.compute.internal systemd[1]: Started sample.
$ tail -f /tmp/log.txt
Sun May 22 03:01:09 UTC 2022
Sun May 22 03:01:10 UTC 2022
Sun May 22 03:01:11 UTC 2022
Sun May 22 03:01:12 UTC 2022

---