I have been seeing this behavior for quite sometime on my D700. Bill
On Fri, 2006-09-15 at 11:10 -0600, Tom Russo wrote: > I'm trying to diagnose and fix an extremely annoying behavior that I have > always seen when connecting my laptop running xastir to the D700 in my > vehicle. > > I just made a commit to interface.c that fixed the problem I was seeing below, > but I don't much like what the solution was. This mail is intended to > document the symptom and explain the fix. > > Xastir sends "MYCALL xxxxx-xx" every time it is getting ready to transmit > a posit, followed by "UNPROTO...", then followed by the actual posit data. > Both "MYCALL" and "UNPROTO" are preceded by a control C (to get the TNC > back into command mode). (ASIDE: The control-C before UNPROTO seems like it > might be unnecessary, since the first one should have gotten us into command > mode, but that's beside the point). > > But almost every single time the "MYCALL" string is sent, my d700 replies > "EH?" instead of "MYCALL was xxxxx-xx". This has been happening from the > very first day I ever connected my laptop to my D700 about 2 years ago. The > first time it happened the result was that the still-new TNC didn't know my > callsign and SSID, and the posit went out with "NOCALL" as my callsign. I > fixed it then by manually connecting to the TNC with a terminal program and > typing "MYCALL KM5VY-8" --- that stored my call sign permanently in the > nonvolatile memory, and so ever since then these "EH?" instances have been > unimportant, since until yesterday I only ever connected a single laptop to > the thing with a single SSID. > > It became significant again yesterday when I tried to connect a different > laptop with a different SSID --- the MYCALL line that xastir sends is, > about 99 times out of a hundred, getting an "EH?" and the MYCALL value isn't > getting reset, with the result that my KM5VY-5 laptop's posits were going out > with KM5VY-8's callsign/ssid in the packet. Once in a while, though, it DOES > work. It appears to be random when it does, but it is extremely rare. > > Since seeing this problem, I've tried now and then to figure out how to make > it go away. > > I have toyed in vain with inter-character pacing values --- setting a very > long character pacing has NO effect on this behavior. > > A few weeks ago I even tried changing the interface.c code to pause for > a second or more immediately after sending a control C to the TNC, on the > assumption that it was possible that the d700 pauses to think a while after > being told to go into command mode (there are other places where it is > necessary to insert stupid delays so the D700 can work), but this also had no > effect. No amount of pause up to *5 seconds* after the first control C had > any impact. Scratch one theory. > > And today I finally just tried something obvious: I hacked interface.c and > added a carriage return to the string sent to the TNC, right before the > MYCALL. This has, in quite a bit of testing, completely eliminated the > "rejected MYCALL" instances. > > Now the "EH?" is coming as a result of the leading carriage return instead. > That's why I don't much like the fix: it merely pushes the error message > from the TNC onto a meaningless, null command instead of the important MYCALL > command. > > Clearly, the D700 (or at least *MY* D700) has some Issues with the first > command given after being returned to command mode from CONVERSE mode with > a control-C. I have no idea what those Issues are. > > I committed my changes to interface.c, because it seems a completely benign > thing to have an extra carriage return before the MYCALL for a regular TNC, > and it seems to be a necessary work-around for the D700. > > So here's what I'm actually asking: > > - Do any other d700 users see "EH?" in their View Incoming Data windows > right before "UNPROTO was ..." and a posit transmission by your station? > - Does anyone have a workaround that I should know about? > - Does anyone have a guess as to why such a huge fraction of these > MYCALL commands would be failing on a D700? > - Anyone want to hazard a guess as to what might make it stop doing this? > > This note also serves as warning to users of non-D700 TNCs to be on the > lookout > just in case the leading carriage return I added to the MYCALL command is > NOT as benign as I thought it should be. Please let me know if you see any > strange behaviors resulting from this extra character in the outgoing command > stream. > -- William Gery - KA2FNK Regional Systems Manager National Weather Service Central Region Headquarters 7220 NW 101st Terrace Kansas City, MO 64153-2371 _______________________________________________ Xastir mailing list [email protected] http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir
