Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!olivea!tymix!cirrusl!sunstorm!dhesi From: dhesi%cirrusl@oliveb.ATC.olivetti.com (Rahul Dhesi) Newsgroups: comp.unix.xenix.sco Subject: Re: Do I really need to BREAK to change baud rates? Message-ID: <2837@cirrusl.UUCP> Date: 3 Jan 91 01:48:27 GMT References: <158@mnopltd.UUCP> Sender: news@cirrusl.UUCP Organization: Cirrus Logic Inc. Lines: 30 In nelson@sun.soe.clarkson.edu (Russ Nelson) writes: >The problem is that I get random users calling into the BBS, and it's not >practical to tell them that they have to press break... My admittedly limited experience with setting of data rates suggests that most hardware interfaces will detect a "framing error" when the user types at the wrong data rate. A framing error is all that the software needs to switch to a different data rate. In fact, from the device driver's point of view, a "break" and a "framing error" should be identical. I believe this works for Microport System V/AT on standard AT hardware, and for 4.3BSD on a VAX using dz-11 serial port hardware. If for some reason just typing carriage return doesn't cause the driver to switch to a different data rate (before the user has logged in), then your hardware, or the SCO Xenix device driver, is at fault. I would blame the specific hardware or software, not UNIX in general. If the hardware or software does not correctly detect a framing error, then it probably won't detect a break either. All this assumes, of course, that the configuration files for the dial-up ports have entries such that each entry has a pointer to the next entry for a differnt data rate. E.g., you might cycle through 300 -> 1200 -> 2400 -> 300. This will make the initial getty process cycle through these data rates, switching each time it sees a framing error. -- Rahul Dhesi UUCP: oliveb!cirrusl!dhesi