home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.questions
- Path: sparky!uunet!cs.utexas.edu!oakhill!engage.sps.mot.com!fmbutt
- From: fmbutt@engage.sps.mot.com (Farooq M. Butt)
- Subject: Answer to my question about child processes and their deaths
- Message-ID: <1992Sep13.181407.11902@oakhill.sps.mot.com>
- Sender: news@oakhill.sps.mot.com
- Nntp-Posting-Host: 223.3.249.15
- Organization: Motorola RISC Software, Austin, TX
- Date: Sun, 13 Sep 1992 18:14:07 GMT
- Lines: 19
-
- I just wanted to blast you folks a quick note and let you
- know the net.consensus on my question about the best way to be
- informed of a child process' demise:
-
- 1. trap SIGCLD in the parent
- 2. when you get a SIGCLD, you know "a" child is dead. Do either a
- waitpid (nonblocking type) or a wait3 to figure our which kid
- died.
-
-
- Thanks to all the kind souls who answered.
-
- fmb
-
- --
- Standard High-Tech Disclaimer: NOTHING in the above article has the slightest
- relationship to reality. If any reality correspondences are found, please
- notify me IMMEDIATELY. Any threats, abuse or stupidity of any kind is purely
- UNintentional. These are MY Opinions NOT my employer's.
-