Loading ...
Sorry, an error occurred while loading the content.

Re: [linuxham] Serial port behavior on Fedora 18

Expand Messages
  • Dave B
    ... Hi. Not just Fedora 18, it s a universal trait of Winderz too. For this very reason, my PC-Icom C-IV interface, has some crude logic so that the RTS and
    Message 1 of 10 , Feb 1, 2013
    • 0 Attachment
      On 31 Jan 2013 at 0:50, linuxham@yahoogroups.com wrote:

      > ______________________________________________________________________
      > Re: Serial port behavior on Fedora 18
      > Posted by: "Richard Shaw" hobbes1069@... kf5oim
      > Date: Wed Jan 30, 2013 3:07 pm ((PST))
      >
      > On Mon, Jan 28, 2013 at 4:32 PM, <jhhaynes@...> wrote: > I
      > don't know how long this has been going on, but on a Fedora 18 >
      > system I notice during startup that when it says > > Starting login
      > service > > the serial port gets the RTS lead asserted, and it stays
      > on until > after I log in through the console. Which means if that
      > serial port is > used for PTT on a radio and the radio is turned on
      > while the computer is > booting up it will key the transmitter. Which
      > I guess is not serious > but is annoying. > > Now that we have systemd
      > I don't understand how anything works anymore. > > Does someone know
      > what is going on and how to turn it off?
      >
      > I don't have a USB->Serial device yet, but I'm working on homebrewing
      > one for FLdigi and my Ten-Tec 540 so I'm definitely interested in the
      > problem. Let me know if you find anything out.
      >
      > Thanks,
      > Richard

      Hi.

      Not just Fedora 18, it's a universal trait of Winderz too.

      For this very reason, my PC-Icom C-IV interface, has some crude logic so
      that the RTS and DTR lines have to be used in oposituion. 1 0 or 0 1
      in essance. Both will power the interface, but only one of those
      combinations (RTS+ and DTR-) will key the rig. ANY other combination,
      and the PTT stays dormant.

      With both -ve, you do loose control of the rig, but in that case, 99.9%
      of the time, there is no software running that needs to control the rig.

      With a full '3-phase' rectifier (including the TXD line too) and sensible
      value resoviour caps on the derrived power lines, you can do a lot, with
      a little ingenuity.

      Dave Freese long ago implemented the needed controls in both Fldigi and
      Flrig to accommodate that (many thanks sir) and all that works very well
      indeed.

      73.

      Dave G0WBX.
    • jhaynesatalumni
      ... With me on Windows it blips several times during bootup but does not stay on. We shouldn t have to resort to crude logic in Linux - it s all open, in
      Message 2 of 10 , Feb 1, 2013
      • 0 Attachment
        --- In linuxham@yahoogroups.com, "Dave B" wrote:

        > Not just Fedora 18, it's a universal trait of Winderz too.
        >
        > For this very reason, my PC-Icom C-IV interface, has some crude logic

        With me on Windows it blips several times during bootup but does not
        stay on.

        We shouldn't have to resort to crude logic in Linux - it's all open,
        in principle we should be able to figure out what is causing this.
        But I personally don't understand much about how Linux works anymore,
        especially since the change to systemd. So I was hoping somebody who
        does understand it would tell me what's going on here.
      • Richard Shaw
        ... Hmm... I don t think systemd has anything to do with this particular problem unless there s something about your setup I don t understand. I would think
        Message 3 of 10 , Feb 1, 2013
        • 0 Attachment
          On Fri, Feb 1, 2013 at 10:50 AM, <jhhaynes@...> wrote:
          > We shouldn't have to resort to crude logic in Linux - it's all open,
          > in principle we should be able to figure out what is causing this.
          > But I personally don't understand much about how Linux works anymore,
          > especially since the change to systemd. So I was hoping somebody who
          > does understand it would tell me what's going on here.

          Hmm... I don't think systemd has anything to do with this particular
          problem unless there's something about your setup I don't understand.
          I would think the USB serial device would be "initialized" by a udev
          rule, however, the RTS issue may could just be something the chip does
          on it's own? Or it could be the kernel module for the pl2303 kernel
          driver...

          I'll look into it more when I get my device.

          On another subject, I was pretty confused by systemd at first but I've
          come to like it. It's probably OT for this list though so feel free to
          email me directly.

          Richard
        • Dave B
          ... Hi again. The crude (descrete resistor/diode/transistor logic) is a protective measure. Many other pieces of software can access the serial ports,
          Message 4 of 10 , Feb 3, 2013
          • 0 Attachment
            On 2 Feb 2013 at 0:16, linuxham@yahoogroups.com wrote:

            > --- In linuxham@yahoogroups.com, "Dave B" wrote:
            >
            > > Not just Fedora 18, it's a universal trait of Winderz too.
            > >
            > > For this very reason, my PC-Icom C-IV interface, has some crude
            > > logic
            >
            > With me on Windows it blips several times during bootup but does not
            > stay on.
            >
            > We shouldn't have to resort to crude logic in Linux - it's all open,
            > in principle we should be able to figure out what is causing this. But
            > I personally don't understand much about how Linux works anymore,
            > especially since the change to systemd. So I was hoping somebody who
            > does understand it would tell me what's going on here.
            >
            >

            Hi again.

            The "crude" (descrete resistor/diode/transistor logic) is a protective
            measure. Many other pieces of software can access the serial ports,
            usually in error (my finger trouble) so this was done to prevent
            embarrasment as much as anything else.

            However, it's simple (near trivial) to do, and is "cross platform"
            compatable too! :-)

            73.

            Dave G0WBX.

            PS: AFIK, none of the USB<>RS232 converter chipsets, will pulse the DTR
            or RTS lines when they power up/down. I'd put money on it, that the
            designers of such things have a brownout detection system that would
            lockout any I/O drivers until their own internal processors are up to
            speed and in control of the hardware. If not, they should have.

            What the host system drivers do when they load an initialise however, and
            why, is another question. It could even be a simple diagnostic, set and
            readback, to confirm the lines can be controled OK.
          • w1hkj
            ... Seems to be ubiquitous David. Found it running on Ubuntu 12.04 as well. Dave
            Message 6 of 10 , Feb 3, 2013
            • 0 Attachment
              On 02/03/2013 04:59 PM, David A. Ranch wrote:

              I wonder if ModemManager is biting you here.  This is what I have in my HamPacket documentation:

              http://www.trinityos.com/HAM/CentosDigitalModes/hampacketizing-centos.html#2c.kernelnavigatorudev

                         Modem-Manager and Centos 6
                         --------------------------
                         As part of Centos 6's fully automatic NetworkManager system, 
                         modem-manager is a "automatic" tool to initialize modems.  The 
                         problem with this is that when UDEV initializes the US Interface's
                         FSK port, modem-manager will try to send some Hayes AT commands 
                         to ANY serial port.  This is generally recognized as BAD code and
                         there is a bug filed to change this behavior.  Unfortunately, the
                         current state makes the Navigator assert PTT on the rig and starts 
                         transmitting those commands via RTTY!
              
                            mv /usr/sbin/modem-manager /usr/sbin/modem-manager.disabled
                            killall modem-manager

              --David
              Seems to be ubiquitous David.  Found it running on Ubuntu 12.04 as well.

              Dave
            • jhaynesatalumni
              ... That is indeed the source of the trouble; and disabling it as you did gets rid of the problem. Thanks for pointing this out. I guess it would be more
              Message 7 of 10 , Feb 4, 2013
              • 0 Attachment
                --- In linuxham@yahoogroups.com, "David A. Ranch" wrote:
                >
                >
                > I wonder if ModemManager is biting you here.

                That is indeed the source of the trouble; and disabling it as you did
                gets rid of the problem. Thanks for pointing this out. I guess it
                would be more satisfying to find out what is starting modem-manager
                and tell it not to do so, but for now just making it impossible to
                run is a good solution.

                Jim W6JVE
              • jhaynesatalumni
                An interesting side-observation is that a few years back they seem to have decided that nobody was using dialup modems any more, so you had to jump through
                Message 8 of 10 , Feb 4, 2013
                • 0 Attachment
                  An interesting side-observation is that a few years back they
                  seem to have decided that nobody was using dialup modems any
                  more, so you had to jump through extra hoops to get dialup
                  support at system installation time. I guess now they have
                  decided everybody has modems that need to be managed.
                Your message has been successfully submitted and would be delivered to recipients shortly.