home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!paladin.american.edu!darwin.sura.net!mips!swrinde!cs.utexas.edu!torn!watserv2.uwaterloo.ca!watserv1!hesse
- From: hesse@watserv1.uwaterloo.ca (BJ Cameron - DCS)
- Newsgroups: comp.unix.aix
- Subject: My backend is out of control (and it's not a pretty sight).
- Message-ID: <BtADCK.IzF@watserv1.uwaterloo.ca>
- Date: 20 Aug 92 14:24:19 GMT
- Organization: University of Waterloo
- Lines: 28
-
-
- I am attempting to replace the standard backend program PIOBE with my own
- homebrew code. For the most part, it works as designed; however, it seems
- to have a bit of a problem detecting that the device it is talking to has
- gone away. By "gone away" I mean telling that the device (in this case
- a printer) has been turned off or that the cable has been torn off, breaking
- the connection between the RS6000 and the printer. The method I use to
- test this "loss of printer" is to tear the cable from the RS6000 side, that way
- not being fooled by some wierd cable design. Anyway, when I use PIOBE as
- the backend the removal of the cable is detected and the queue goes down; when
- I use my backend program the removal is not detected causing data to continue
- to be sent to the serial port.
-
- Now, I have a feeling that my code should be detecting this change of state
- but I'm not real sure how this is done.
- Some questions:
-
- 1. Is there a way (maybe an ioctl call) that will tell me if the DTR or DCD
- signals have gone away?
-
- 2. Is in fact the qdaemon or the device driver telling my program that the
- device has gone away, but my program doesn't understand? if so, which one?
-
- 3. Is there a sample of a backend program that shows how this is detected?
-
- 4. Better yet, is the source for PIOBE available?
-
- Thanks... Bri
-