home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ferkel.ucsb.edu!taco!rock!stanford.edu!ames!network.ucsd.edu!mvb.saic.com!unogate!news.service.uci.edu!ka.reg.uci.edu!bmarlowe
- From: bmarlowe@ka.reg.uci.edu (Brett Marlowe)
- Subject: Re: Non blocking open of FIFO: Can it be done in perl?
- Nntp-Posting-Host: ka.reg.uci.edu
- Message-ID: <bmarlowe.722105958@ka>
- Newsgroups: comp.lang.perl
- Keywords: O_NDELAY in open on a FIFO
- Lines: 29
- References: <bmarlowe.722046368@ka> <BxwD69.1IH@chinet.chi.il.us>
- Distribution: usa
- Date: 18 Nov 92 17:01:35 GMT
-
- In <BxwD69.1IH@chinet.chi.il.us> les@chinet.chi.il.us (Leslie Mikesell) writes:
-
- >In article <bmarlowe.722046368@ka> bmarlowe@ka.reg.uci.edu (Brett Marlowe) writes:
-
- >>I'm using FIFO as a rendezvouz point between two entirely different
- >>systems. My question is... is the there a way to specify O_NDELAY on
- >>an open call? Alternatively is there a way to make sure there is a
- >>reader on the other end of the FIFO?
-
- >At least on sysVish machines, opening a FIFO for read/write access
- >is non-blocking. As a side effect you won't ever see EOF even
- >if the other processes open and close the fifo repeatedly, so reads
- >will always wait for some data which is what you usually want for the
- >case of one reader process acting as some kind of server to one or more
- >writing process.
-
- >Les Mikesell
- > les@chinet.chi.il.us
-
- SunOS would seem to differ. My implementation blocks on an open until
- a reader starts up and the reader receives an EOF when I close. What
- I am trying to avoid is blocking on open when the reader has died/gone
- away.
-
- -Brett
- --
- -------------------------------------------------------------------------------
- Brett Marlowe | University of California, Irvine
- bmarlowe@uci.edu | Office of the Registrar (making sure you get your bills)
-