百度了一下，一般守護(hù)進(jìn)程的工作形式是：監(jiān)聽端口，有數(shù)據(jù)來(lái)了就fork()子進(jìn)程對(duì)數(shù)據(jù)進(jìn)行處理。
小弟現(xiàn)在需要編寫一個(gè)分布式任務(wù)處理的程序，這個(gè)任務(wù)手工執(zhí)行的方式是：
1、普通用戶登錄server；
2、手動(dòng)加載profile以及掛載網(wǎng)絡(luò)文件系統(tǒng)；
3、執(zhí)行程序（因?yàn)槌绦蛟贜FS上面才有）。
這種任務(wù)執(zhí)行方式，怎么用守護(hù)進(jìn)程實(shí)現(xiàn)呢？畢竟手工執(zhí)行任務(wù)需要以普通用戶身份登錄控制終端，而守護(hù)進(jìn)程屬于超級(jí)用戶，并且是不屬于某個(gè)控制終端的。
問(wèn)題是守護(hù)進(jìn)程怎么模擬任務(wù)手工執(zhí)行這三個(gè)步驟？因?yàn)檫@個(gè)任務(wù)需要某個(gè)用戶控制終端所屬的進(jìn)程來(lái)執(zhí)行。
2
0
0
光陰似箭催人老，日月如移越少年。
I don’t understand what you mean. Port monitoring does not distinguish between user permissions. Ordinary users can also send sockets to daemons created by super users. I don’t know where you can’t figure it out.
The above two parts can be done with shell programming
I won’t go into details if ssh is configured and used to log in with a certificate.
Ssh can be followed directly by the file name, which is to execute a certain command after the login is completed
As for hanging on the network file system and executing a specific program, it can be written in the command script executed when logging in
![[Web front-end] Node.js quick start](https://img.php.cn/upload/course/000/000/067/662b5d34ba7c0227.png)
