home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.aix
- Path: sparky!uunet!cs.utexas.edu!geraldo.cc.utexas.edu!portal.austin.ibm.com!awdprime.austin.ibm.com!greenber.austin.ibm.com!jfh
- From: jfh@greenber.austin.ibm.com (John F Haugh II)
- Subject: Re: * IDLE OUT ON RS/6000 *
- Sender: news@austin.ibm.com (News id)
- Message-ID: <C1A3CG.2I0J@austin.ibm.com>
- Date: Fri, 22 Jan 1993 23:17:04 GMT
- References: <727306479@romeo.cs.duke.edu> <C14Dzu.owr@austin.ibm.com> <1993Jan20.104701.4825@pollux.lu.se>
- Organization: AIX Software Support, Austin, Republica de Tejas
- Lines: 41
-
- In article <1993Jan20.104701.4825@pollux.lu.se> gs@balder.tf1.lu.se (Goran Svensson) writes:
- >This has been a loooong story. Months ago I filed an complaint (to IBM)
- >about processes not dying when an telnet connection was broken and my
- >point was that SIGHUP should be delivered when the terminal was disconnected.
- >
- >The answer was that waiting for SIGHUP was wrong. What I should do was
- >to look for the status code on a IOrequest to my terminal and if that was an
- >EOF, then I should go climb a nearby tree. I do not like this scheme, mostly
- >because if I have a CPU bound program it will go on forever.
- >
- >It has something to do with POSIX be unclear or whatever ...
- >What is your reaction? I feel that a SIGHUP should be delivered, but I am
- >not in the position to argue with N.N. (name withhold) is Austin Tejas
-
- This is what POSIX says.
-
- 3.2.2.3
- (4) Termination of a process does not directly terminate its chilren.
- The sending of a SIGHUP signal as described below indirectly
- terminates children in some circumstances.
- (6) If the process is a controlling process, the SIGHUP signal
- shall be sent to each process in the foreground process group of
- the controlling terminal belonging to the calling process.
- (8) If the implementation supports job control, and if the exit of
- the process causes a process group to become orphaned, and if any
- member of the newly-orphaned process group is stopped, then a SIGHUP
- signal, followed by a SIGCONT signal shall be sent to each process
- in the newly-orphaned process group.
-
- OK. If the process which was not dying was a back-ground process, you are
- out of luck unless the backgrounded process was asleep. If the process was
- in the foreground process group, and it wasn't being sent SIGHUP, then you
- have a defect and should call it in.
-
- I did write a quick little testcase, and we do the right thing. If you
- are the foreground process, and your process leader dies, you get a SIGHUP.
- If you are in the background and not stopped, you get no signal.
- --
- John F. Haugh II | Quality is ... knowing who | MaBellNet: (512) 823-8817
- SneakerNet: 042/2F068 | your customer is and what | VNET: HAUGH at AUSVM8
- [ DoF #17 ] [ TSAKC ] | your customer wants. | Disc: I speak 4 me, !IBM.
-