How reliable are named pipes?
in [Unix Programming]
|
From: Peter Olcott on 9 Apr 2010 10:46 "Ersek, Laszlo" <lacos(a)caesar.elte.hu> wrote in message news:Pine.LNX.4.64.1004091445300.25979(a)login01.caesar.elte.hu... > On Fri, 9 Apr 2010, Rainer Weikusat wrote: > >> The main problem of FIFOs is that they are not suitable >> for anything except 1:1 communication because there is >> neither a way to determine which of possibly many >> processes wrote a particular sequence of bytes nor [...] > > I apologize for splitting hairs and parroting banalities, > but if all producers agree to write only messages not > bigger than _POSIX_PIPE_BUF (== 512) bytes, then those > messages will be atomic, in a portable way, and a single > consumer can parse the stream. This enables n-to-1 > communication. (Admittedly, not a very general kind.) > Messages may carry a fixed length header, designating the > complete (or remaining) length of the message, and > identifying the sender. > > lacos Yes, exactly what I am proposing. Would there be any advantage to using some form of sockets instead of named pipes?
From: Rainer Weikusat on 9 Apr 2010 16:44 "Peter Olcott" <NoSpam(a)OCR4Screen.com> writes: > "Ersek, Laszlo" <lacos(a)caesar.elte.hu> wrote: in message >> On Fri, 9 Apr 2010, Rainer Weikusat wrote: >>> The main problem of FIFOs is that they are not suitable >>> for anything except 1:1 communication because there is >>> neither a way to determine which of possibly many >>> processes wrote a particular sequence of bytes nor [...] >> >> I apologize for splitting hairs and parroting banalities, >> but if all producers agree to write only messages not >> bigger than _POSIX_PIPE_BUF (== 512) bytes, then those >> messages will be atomic, in a portable way, and a single >> consumer can parse the stream. [...] > Yes, exactly what I am proposing. Would there be any advantage to > using some form of sockets instead of named pipes? The disadvantage of using AF_UNIX sockets is that 'opening' would be more complicated. Apart from that, there would be only advantages - message size not limited to PIPE_BUF - bidirectional - if datagram-based sockets are used, neither framing nor parsing are necessary to extract messages from a bytestrem - kernel provides sender and recipient addresses automatically - replies can use the same socket(s) as requests
From: Peter Olcott on 9 Apr 2010 16:48 "Rainer Weikusat" <rweikusat(a)mssgmbh.com> wrote in message news:871veowe2j.fsf(a)fever.mssgmbh.com... > "Peter Olcott" <NoSpam(a)OCR4Screen.com> writes: >> "Ersek, Laszlo" <lacos(a)caesar.elte.hu> wrote: in message >>> On Fri, 9 Apr 2010, Rainer Weikusat wrote: >>>> The main problem of FIFOs is that they are not suitable >>>> for anything except 1:1 communication because there is >>>> neither a way to determine which of possibly many >>>> processes wrote a particular sequence of bytes nor >>>> [...] >>> >>> I apologize for splitting hairs and parroting >>> banalities, >>> but if all producers agree to write only messages not >>> bigger than _POSIX_PIPE_BUF (== 512) bytes, then those >>> messages will be atomic, in a portable way, and a single >>> consumer can parse the stream. > > [...] > >> Yes, exactly what I am proposing. Would there be any >> advantage to >> using some form of sockets instead of named pipes? > > The disadvantage of using AF_UNIX sockets is that > 'opening' would be > more complicated. Apart from that, there would be only > advantages > > - message size not limited to PIPE_BUF > - bidirectional > - if datagram-based sockets are used, neither > framing nor > parsing are necessary to extract messages from a > bytestrem > - kernel provides sender and recipient addresses > automatically > - replies can use the same socket(s) as requests Can this work for TCP sockets too?
From: Ian Collins on 9 Apr 2010 17:24 On 04/10/10 08:48 AM, Peter Olcott wrote: > "Rainer Weikusat"<rweikusat(a)mssgmbh.com> wrote: [formatting fixed] >> The disadvantage of using AF_UNIX sockets is that 'opening' would be >> more complicated. Apart from that, there would be only advantages >> >> - message size not limited to PIPE_BUF >> - bidirectional >> - if datagram-based sockets are used, neither framing nor >> parsing are necessary to extract messages from a bytestrem >> - kernel provides sender and recipient addresses automatically >> - replies can use the same socket(s) as requests > > Can this work for TCP sockets too? Can what work? There's quite a list there! -- Ian Collins
From: Peter Olcott on 9 Apr 2010 17:25
"Ian Collins" <ian-news(a)hotmail.com> wrote in message news:829k8rF5njU6(a)mid.individual.net... > On 04/10/10 08:48 AM, Peter Olcott wrote: >> "Rainer Weikusat"<rweikusat(a)mssgmbh.com> wrote: > > [formatting fixed] > >>> The disadvantage of using AF_UNIX sockets is that >>> 'opening' would be >>> more complicated. Apart from that, there would be only >>> advantages >>> >>> - message size not limited to PIPE_BUF >>> - bidirectional >>> - if datagram-based sockets are used, neither >>> framing nor >>> parsing are necessary to extract messages from >>> a bytestrem >>> - kernel provides sender and recipient addresses >>> automatically >>> - replies can use the same socket(s) as requests >> >> Can this work for TCP sockets too? > > Can what work? There's quite a list there! > > -- > Ian Collins Can I use TCP sockets as the kind of sockets that I am using for IPC? |