Re: grouping packets
On Sun, Aug 13, 2000 at 01:57:57PM +, Robin Gilks wrote: > The *NOS SCC driver (from PE1CHL derivatives) implements a 'mute' or 'txhold' > time which is in essence the 1st slottime which is always waited on prior to > transmission. This allows the first delay to be totally different from > subsequent ones. My home brew TNC uses by default a slottime of 13 (130mS) and > a 1st wait of 35 (350mS) but they are all adjustable using kiss parameter 7 > (non-standard value but supported under many *NOSes) > > If its a KISS mode TNC you are using then its up to the TNC to implement it - > SCC can do it directly and hdlcdrv derived interfaces such as Baycom, g4xyw and > soundcard should also be 'tweakable' under Linux. I'm using the soundmodem.. is it the slot time, or ppersist, or ... which I need to change to implement this? thanks Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
grouping packets
Hi, Can I make the linux kernel hold on to outgoing AX.25 packets for a few hundred milliseconds or more and group any queued packets into a single transmission? How would I do that if possible? thanks Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: debian-hams
On Tue, Aug 08, 2000 at 10:55:03AM +1000, Barrett, Peter G wrote: > i'm trying to subscribe to the debian hams M.L. but no luck. can anyone here > tell me the right way? email [EMAIL PROTECTED] and say "subscribe" in the body. It's a pretty quiet list, so don't expect to hear anything straight away (except for the subscription confirmation). Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: (off-topic) Don't you just hate BADLY configured auto responders !!!
On Mon, Aug 07, 2000 at 10:50:53AM -0300, Rubens Nogueira wrote: > And Raymond, I just ask to you: where are you? Turkey, of course! Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: RE:smdiag
On Sun, Aug 06, 2000 at 11:53:48PM +0100, Raymond Miller wrote: > I am on holiday in Turkey for a fortnight but will be back at work on Monday 21st >August 2000. If it is a personal matter there is someone at home to take your calls. >If it is a business issue then talk to Matt and/or Amanda. And if your email autoresponder drives me nuts, who should I talk to? Please make sure your autoresponder doesn't reply to every mailing list post, or unsubscribe when you go away. thanks, Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: ifconfig sm0 up :-(
On Mon, Aug 07, 2000 at 11:21:37AM +0400, Yuriy N. Kosenko wrote:
> I compiled the Kernel with care of ax25 inside.
> After #setcristall, #sethdlc, #ifconfig and #smmixer commands
> I print: ifconfig sm0
> Reseive:
> sm0 Link encap:AMPR AX.25 HWaddr OH2BNS-1
> inet addr:192.168.1.3 Mask:255.255.255.0
> RUNNING MTU:256 Metric:1
> RX packets:0 errors:0 dropped:0 overruns:0 frame:0
> TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:0
> Interrupt:7 Base address:0x534 DMA chan:1
> (Simpl ifconfig show me only eth0 and loopback. But sm0 is absent.)
It sounds like your I/O and IRQ etc settings ('sethdlc') are
incorrect. After you configure the IP, the interface should
automatically come up and appear in 'ifconfig' (without
parameters). Check to see if configuring the interface
generates any messages in the system log.
Also check the I/O etc for your PTT port. If you use a serial
port, you must make sure the serial driver is not using it as well.
For example, if you use ttyS1 (io 0x2f8) for PTT, run
'setserial /dev/ttyS1 uart none' to disable the serial driver
for that port. Otherwise the interface will not come up.
Hamish
--
Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: NEWQPSK tests
On Mon, Jul 24, 2000 at 09:36:30AM -0700, Ken Koster wrote: > I've been trying to get a couple of the local guys interested enough to do > local testing, but everyone is a bit too busy right now, or they don't have > their PC's tied to their HF rigs yet. Similar story here. Also, not many Linux- using hams either. > I'd be willing to setup a periodic beacon while I'm home in the evening > (approx. 0300-1400 GMT). Since I'm in northwest Washington state picking a > frequency that has half a chance of making it between us is going to be the > difficult part. There should be propagation at some time during that period on 20 metres, around 0900 I would guess. I was intending to run the beacon for a few days while I am at work (2200-0800Z), but could do an evening also (until about 1200Z). The equipment is too noisy to leave running overnight. (The FT-847's cooling fan runs continuously!) Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Query on Rig-control API
On Mon, Jul 24, 2000 at 08:16:53PM +1200, Andrew Sands wrote: > There was minor discussion on the possibility of a generic rig-control API > during the discourse about the creation of a linux/unix based logging > program suite? > As I have the opportunity to have access to some computer controllable rigs > on behalf of their owner whom is currently upgrading his shack, I thought > this a suitable time to put some free(ish) time into producing a reasonable > result. I'm still interested. I haven't done anything on it though. I can contribute code for the FT-847, and perhaps ICOM as well (although I'll probably sell my ICOM soon). I had a few thoughts on the topic.. Radios have lots of different capabilities so the API should probably have a way to find out what capabilities are available. For example, my FT-847 has satellite mode, and you can read the S-meter over the CAT interface. It has two VFOs but you can't switch VFOs through software. It has 1Hz tuning but only 10Hz resolution over CAT. It has all bands 160m through 70cm, and has CW and CW-R modes. My IC-726 has no satellite mode and no S-meter command, but you can switch between the two VFOs over the CI-V interface. It has HF and 6m, but switching between HF and 6m is a bit of an artform. It has 10Hz tuning steps and can be controlled down to 10Hz on the interface. It only has CW mode (LSB) with no reverse. Some ideas. The library could read a config file which listed which radios were connected to which serial ports. Application software could get a list of connected radios that way, and find out what capabilities they had. A few basic commands should be supported on all radios -- get/set frequency and mode, probably nothing more is completely standard. (Early firmware on the FT-847 didn't have any read back at all! Most of those got upgraded though.) Memory handling would be quite complicated and IMHO low priority. If made generic enough I suppose the API could work with memory programming for VHF/UHF FM rigs as well. Just a few random thoughts, Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
[log project] CTY.DAT
I've been working on the logging project where time permits. I've found a couple of really dumb things in CTY.DAT though. For example, the entity "Juan de Nova & Europa" lists for alternate prefixes, J and E. So my code so far thinks every Japanese or E* station qualifies. I've written in ADIF import function. The log data and country data are all imported into Postgres tables using DBI. I'm trying to write a DXCC summary report. The challenge at the moment is to filter the data to prevent this: primary_prefix|call |name | qso_date| qslrdate|band --+-+-+--+--+ 3A|3A2K |Monaco |09-08-2000| |20M 3D2 |3D2AA|Fiji |05-05-2000| |20M 3D2 |3D2AA|Fiji |06-12-2000| |20M 3D2/r |3D2AA|Rotuma |05-05-2000| |20M 3D2/r |3D2AA|Rotuma |06-12-2000| |20M 3D2AA is listed specifically for 3D2/r, but of course also fits 3D2*. G |GU3WHN |England |04-09-1999| |20M G |GU3WHN |England |11-09-1999| |20M G |G0UPF|England |25-02-2000| |20M G |M5AHW|England |08-08-2000| |20M G |GB5HQ|England |09-08-2000| |20M G |M5AHW|England |09-08-2000| |20M GU|GU3WHN |Guernsey |01-09-1999| |20M Same problem here: GU is a subset of G. I think it should be good enough to look for the longest prefix as being the most specific. Just to let you all know I'm still working on it anyway, Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: NEWQPSK tests
On Sat, Jul 22, 2000 at 06:25:43PM +0200, Kai Schulte wrote: > > Is anyone interested in doing some NEWQPSK testing with me on HF > > this weekend? > > How do the tests work, and what band(s) do you use? I'd love to give > psk31 a try but currently I don't have _any_ working short wave antennas, > so this would be good motivation to build/fix one :) Hi Kai, Well, I was going to arrange a time and frequency where we could try to make contact. If nothing else I might leave a beacon running one day this week, if there's anyone hear who would try to copy it. I would be running Tomi's NEWQPSK modem for Linux, transmitting AX.25 packets using the 'beacon' program. Would anyone try to decode this if I set it up? 73, Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
NEWQPSK tests
Hi all, Is anyone interested in doing some NEWQPSK testing with me on HF this weekend? I haven't done any more testing since my beacon experiment of several months ago and I'd like to try it again. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: relevance
On Sat, Jul 15, 2000 at 10:30:03PM -0500, Nate Bargmann wrote: > IMO, I think the reason most WinDOS software is shareware/closed/binary > only stuff has much to do with what you mentioned above. I believe it > also has a more basic reason. To do any serious development on a WinDOS > platform requires a commitment of a substantial amount of money for > compiler, IDE, and developer's kit. After making this purchase most > software authors would like to recoup at least part of the investment. Understandable, but I don't think this is entirely true any more. You can get a port of gcc for Windows (from Cygnus/Red Hat), for example. Of course, we know this but those developers don't. Thanks for your comments, Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: relevance
On Fri, Jul 14, 2000 at 02:51:22PM +, James R. Saker Jr. wrote: > Incidentally, I have to believe that the Linux hams have the most to offer > the future of amateur radio -- the spirit of the open source community and > its accomplishments could make a dramatic impact on the amateur world. Unfortunately, there's plenty of opposition to open source amongst amateurs it seens. There's been some threads about it recently on the APRS mailing list and the AMSAT mailing list. I've been pushing the open source line pretty hard on the APRS list. When I wrote to say that the long-term future of APRS was open source, one troll even wrote back to say that if that's true we're better off without it! I'm astonished that so many hams will post full descriptions of their radio circuits in QST and the like, but so may others won't share their software. 73 Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: kernel comp.
On Wed, Jul 12, 2000 at 08:46:07PM +, Richard Adams wrote: > old image to *.old BUT what if the new image has something missing and does > not boot,? you are then up the paddle without a creek, unless you created > some rescue disk sets, like many unfortunalty dont. Use grub. It can read ext2fs, so you can tell it what kernel to use on the command line (and no predefined list is required). Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
aprsd 2.1.3.vk3sb.1 try 2!
Apologies to everyone who tried to download 2.1.3.vk3sb.1; the file is on the web server now. (Quite a lot of hits in the log already.) Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
aprsd 2.1.3.vk3sb.1
I've updated the sockets-enabled aprsd to version 2.1.3. Now available from http://aprs.cloud.net.au/aprsd.shtml There could be bugs in the serial TNC code. (I'm not sure why anyone would use my mods if they were going to use it with a TNC anyway, but it is supported.) The sockets code seems to be working fine though. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Fw: NEW aprsd Linux APRS Server ready to download
t; > 19) Cleaned up code so it now compiles without warnings with the -Wall > option. > > Added "#define _PTHREADS" and "#define _GNU_SOURCE" to all sources. > > This should make the STL container library thread safe. > > > > 20) Changed instances of gethostbyname2() to the thread safe version, > gethostbyname2_r(). > > Also found and changed some other non thread safe fuctions to the safe > versions. > > > > 21) I fixed a bug in the igate connection thread which resulted in > sockets not being > > closed after a failed attempt to connect. This caused aprsd to > eventually run out of sockets > > if one or more igates were unreachable. When this happened no more > connections could be made. > > This bug has been in all previous versions and I believe it has > caused most of the "lockup" problems. > > > > 22) Added "hub" keyword to aprsd.conf. This is used the same way as > "igate" to define remote > > systems to connect to. The difference is that although many hubs can > be defined only > > one connection will be active at any time. If the connection fails the > next hub will be > > tried in rotation until one accepts a connection. Use "hub" to > connect to the "master" > > aprs servers on port 10152 or 23. > > > > > > > > -- > > Dale Heatherington > > [EMAIL PROTECTED] > > Web Page http://www.wa4dsy.net > > Sent by KMail for Linux > > > > > > --- > > You are currently subscribed to aprssig as: [EMAIL PROTECTED] > > To unsubscribe send a blank email to [EMAIL PROTECTED] > > > > > > > -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Predict not reading TLE file
On Thu, Jun 15, 2000 at 11:36:34AM -0400, John Magliacane wrote: > > I am using two different versions of predict and found the same > > problem in each. If I download the ISS TLEs from ftp.celestrak.com, > > I find that predict finds no satellite in the file. > > Interesting... I was able to duplicate the situation myself, so it looks > as if it's a bug, not a feature. :-) Lemme take a look at the code, and > see if I can find the problem. Could it be a problem with CRLF versus LF line endings? Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: ARRL Poll online
On Wed, May 24, 2000 at 08:40:56AM -0400, Jon Bloom wrote: > What I've been thinking of is a Perl application that uses Perl/Tk to implement the >UI, ImageMagick to display and print the pages and DBI to access the article >database. (DBI would allow easy porting to the database engine of your choice.) I'm >open to other suggestions, though. That sounds excellent. > The AView program is used only with the page-scan products (that use the TIFF/JPEG >images), so that point doesn't really apply to AView. Sorry, I was confusing AView with Media View. AView is fine, since the pages are in the original format. > Sometime later this year I expect we'll have the 1995-1999 QST View CD-ROM >available, at which time all issues of QST from 1915-1999 will be available in >TIFF/JPEG format. Any chance the 1995-1999 set might be in PDF instead? 73, Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: ARRL Poll online
On Tue, May 23, 2000 at 09:10:39AM -0400, Jon Bloom wrote: > > Thank goodness. I was just going to say -- at least use standard formats. > > The QEX 81-98 CD is ugly to view in Linux (uses a mix of TIF and JPG > > files from memory); > > TIFF and JPEG aren't standard formats? Yes they are -- but there's no easy way to browse through the pages. > Actually, I've been thinking about putting something together for Linux to allow >reading the QST (QST View), QEX and NCJ CD-ROM collections. Having the images in >TIFF/JPEG format makes that a lot more feasible. Yep. That sounds great. Perhaps not too hard using Tcl/Tk or Perl/Tk or similar. So the QST View CDs also use the TIF/JPG mix? Glad to hear they're not all in the MS proprietary format. > That's correct; it was made using Microsoft's Media View product (sort of an >enhanced Windows help file). No new ARRL products will use that. Great. > > The AView program is yuck. > > Not a particularly informative comment. Did you have some specific gripe or were you >just trying to be insulting? Sorry, no offense was intended. It is certainly usable. However, I would prefer that the original formatting/layout was maintained, as with PDF or TIF/JPG etc. The QST/NCJ/QEX viewer (looking at 1998 for example) does not preserve the formatting. From memory images are viewed separately too (although I don't have the CD in the drive to confirm that at present.) 73 Hamish (who just join the ARRL a week or two ago, coincidentally) -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: ARRL Poll online
On Mon, May 22, 2000 at 10:31:54AM -0400, Bloom, Jon, KE3Z wrote: > The problem -- which affects not only Linux/Unix products but products for > all non-Windows systems -- is that the total market for ARRL products is > relatively small. That makes it hard to focus on systems other than Windows. > As cross-platform tools and applications become more prevalent, we'll try to > do a better job of supporting other platforms. (That's one reason we've > changed our CD-ROM books to PDF format. Now if Adobe would only provide a Thank goodness. I was just going to say -- at least use standard formats. The QEX 81-98 CD is ugly to view in Linux (uses a mix of TIF and JPG files from memory); I don't think the QST 98 CD (for example) is viewable at all. The AView program is yuck. 73 Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: off topic!
On Mon, May 15, 2000 at 11:09:03AM -0300, Jean Letourneau (VE9PT) wrote: > On Sun, 14 May 2000 [EMAIL PROTECTED] wrote: > > > Hi > > Sorry this is an big off topic but not sure where else to ask > > so any help will be great > > ok i have three machines that are tie together with ethernet using > > thincoax bnc type > > it seems all works ok with two bnc connectors on one machine and the other > > two boxes have an bnc on each to it and then on the other side of the T > > connector it has an termertor but when i take one of them off on one of the >machines and run another thincoax with bnc connecto > > r to it it seems it locks up > > anything from flowing anywhere so what i am thinking is it that you only can > > have up to three boxes tie together with thin coax so any info on that will > > help thanks in advance > > Sorry for the miss spelling of some of this! > > BOB N1UAN Maybe Bob has run a drop lead, as in RT---T---TR PC | PC | | | PC (where he says "but when I take one of them off on one of the machines and run another thincoax with bnc connector to it"). This is not supported and will ruin the network. Just think about the standing waves (SWR etc) on the cable created by the huge impedance bump. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [logging] new schema from CTY.DAT
On Fri, May 12, 2000 at 01:28:29PM +, Wilbert Knol wrote: > The 4U1VIC entries aren't duplicates. One of them (the IAURU Headquarters > entity) has the asterisk in front of it, indicating it is a separate entity *only* >in > CQWW contests. It is not a DXCC entity, and outside CQWW contests it > counts as OE, as indicated by the line without the '*', which credits 4U1VIC > to OE. Don't ask me why! See the readme file that comes with CTY.DAT for > details. Ahh, my mistake. I will have to come up with a solution to that then. > I had a quick look around to see where CT gets the 'subentities' info from (the > stuff about VK6 being Zone 29 etc). It doesn't appear to be in the CTY.DAT > or any other files...if I can't figure it out I will ask on the CT reflector. Great. Thanks Wilbert. We need the QSO table format soon, before work on interfaces can begin. At this point, I'm wondering whether a simple single table might be sufficient. Or do we want a separate table mapping callsign to name, QTH, non-QSO-specific comments etc? Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Thu, May 11, 2000 at 12:04:54PM +0200, Ivo Simicevic wrote: > On Thu, May 11, 2000 at 07:31:48PM +1000, Hamish Moffatt wrote: > > > > At the same time, we've agreed that no single front end will suffice > > for both contesting and casual QSOs. So since we accept that multiple > > front ends can be needed, they can differ in implementation too. I don't > > think a web front end could be fast enough or flexible enough for > > contesting. > > > > Unless it is written in Java :-)) Did somebody say slow? :-) Java is not a bad idea for a front end. There are JDBC drivers for Postgres and MySQL. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Thu, May 11, 2000 at 07:34:58AM +0200, [EMAIL PROTECTED] wrote: > Said this, there is no doubt that PHP will not solve every aspect as > efficiently as any other language. And of course, you could always use another > language to solve certain parts, which in my modest opinion that should be > the best approach. My point is that we should come with a solution that would > fit our needs and be as portable as possible, and we should also strive to > implement only *ONE* solution in order not to spread the efforts. > > And yes, I am also a perl fun (that's why I use modperl in my web server ;). > Although we both should take a look at python, which seems to have some > strengths that perl has not) At the same time, we've agreed that no single front end will suffice for both contesting and casual QSOs. So since we accept that multiple front ends can be needed, they can differ in implementation too. I don't think a web front end could be fast enough or flexible enough for contesting. 73 Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [log project] schema
On Wed, May 10, 2000 at 02:04:29PM +, Wilbert Knol wrote:
> CTY.DAT doesn't contain sub-entity info ("VK6 is in zone 29" etc) so that
> info would have to come from somewhere else. But it's good to have the
> facility designed in.
It could -- their format supports it, but the data just isn't there.
They could just list an additional prefix: VK6(29). There's no way
to specify overrides for time zone or a better description though
(like Western Australia).
> However, the ARRL DXCC list has some additional info: whether the ARRL
> QSL bureau will handle cards to that entity, and whether 3rd party traffic with
> that entity is OK, of possible interest to US ops.
Perhaps.. as long as we can incorporate the VK QSL information as well :-)
Hamish
--
Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Wed, May 10, 2000 at 07:23:09AM +0200, [EMAIL PROTECTED] wrote: > > If someone wants to grab this idea, but advice to use PHP3 language > > instead of Perl. > > Me too. Accessing a database with PHP is really easy, and PHP is also a > language that can be very quickly learnt. I have recently heard people say how much better Perl is that PHP. I investigated mod_perl recently for Apache and I was very impressed with the flexibility. I understand you can embed perl scripts inside .shtml (server-side includes) files. Very impressive. Personally, I have done a lot of work with Perl and none with PHP. So I intend to write my version of the front end in Perl. Thanks, Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [logging] new schema from CTY.DAT
On Tue, May 09, 2000 at 10:33:07PM -0400, Steve Dimse K4HG wrote: > On 5/9/00 9:15 PM Wilbert Knol ([EMAIL PROTECTED]) wrote: > > >Hmmm. Somebody on some oddball IOTA island could still have a perfectly > >ordinary callsign. You can't map prefix <-> IOTA, I think. Any award experts > >amongs us? > > > No, you can't. I live in NA-62, nothing special about my call... I thought that would probably be the case. It's only useful to map prefix -> IOTA if the island is a separate DXCC entity. Thanks, Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [logging] new schema from CTY.DAT
On Tue, May 09, 2000 at 04:58:40PM +0200, Ivo Simicevic wrote: > On Tue, May 09, 2000 at 10:56:36PM +1000, Hamish Moffatt wrote: > > > > I would like to incorporate IOTA data. The challenge will be to > > be able to import it into an already-populated database. Perhaps > > it might be simpler to use a separate set of tables. More work > > at lookup time, though. > > > > If I have understood you corectly, you will have to have separate > table(s) as it will be one-to-many relation (one prefix , couple of > IOTA references), many-to-one (different prefixes, same IOTA) or > many-to-many depending what you need (IOTA for prefix or prefixes > for IOTA). That's correct. I think a map of prefix->IOTA would be the most useful. That same information could be included in the existing prefix table. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
[logging] new schema from CTY.DAT
OK I've been working on the DXCC database again. The CTY.DAT format
is quite good and the web site seems to encourage other people to use
it, so I decided it would be OK.
Their format does allow lat/long, zones and continent to be overriden
on a per-prefix basis, with multiple prefixes per entity. This could
be used for the sub-entities we discussed, although we would have
to merge the data in some how, and also it doesn't allow you to override
the description for a sub-entity.
Here's the schema (SQL, more generic this time):
drop table country;
create table country (
prefix varchar(10) PRIMARY KEY,
name varchar(26),
timezone int2
);
drop table prefix;
create table prefix (
prefix varchar(10) PRIMARY KEY,
cqzone int2,
ituzoneint2,
latitude float4,
longitude float4,
continent varchar(2)
);
Quite a bit simpler, really. The country table contains the primary
prefix, country name, and timezone (which can't be overriden on
a per-prefix basis). Then the prefix table contains the specifics, and
a link to the primary prefix.
The doco for CTY.DAT on the K1EA web site is wrong. It says that
prefixes may be at most 6 characters, but there are things like
'4U50VIC", "3D2AG/P" etc. I allowed 10 instead. It also says that
the lines of additional prefixes begin with &, but none of them do.
There's also one duplicate; 4U1VIC is listed twice, once for 4U1 and
once for OE.
I've attached: a script to convert CTY.DAT to SQL, and the output of
that script. I'll modify the script to use DBI and insert the data
directly into Postgres, probably.
Comments most welcome.
I would like to incorporate IOTA data. The challenge will be to
be able to import it into an already-populated database. Perhaps
it might be simpler to use a separate set of tables. More work
at lookup time, though.
Hamish
--
Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
#!/usr/bin/perl
use strict;
use vars qw(@in @prefixes);
if (not defined $ARGV[0]) {
print "syntax: $0 \n";
exit;
}
open(IN, $ARGV[0]) or
die "$0: could not open $ARGV[0]\n";
while (my $in = ) {
chomp($in);
$in =~ s/\x0D$//; # Remove carriage returns
push @in, $in;
}
close(IN);
my $i;
print "delete from country;\n";
print "delete from prefix;\n";
while ($i <= $#in) {
my ($name, $cqzone, $ituzone, $continent, $lat, $long, $tz, $primary);
my (@prefixes);
($name, $cqzone, $ituzone, $continent, $lat, $long, $tz, $primary)
= split(":", $in[$i]);
$i++;
$name =~ s/^\s+//;
$tz =~ s/^\s+//;
$cqzone =~ s/^\s+//;
$ituzone =~ s/^\s+//;
$lat =~ s/^\s+//;
$long =~ s/^\s+//;
$continent =~ s/^\s+//;
$primary =~ s/^\s+//;
print "insert into country values ('$primary', '$name', $tz);\n";
# Get additional prefixes
while ($in[$i] =~ m/^\s/) {
$in[$i] =~ s/^\s+//;
my $o_cqzone = $cqzone;
my $o_ituzone = $ituzone;
my $o_lat = $lat;
my $o_long = $long;
my $o_continent = $continent;
my $o_prefix;
my @aux = split(",", $in[$i]);
my $aux;
foreach $aux (@aux) {
$aux =~ s/;$//;
if ($aux =~ m#^([\w/]+)#) {
$o_prefix = $1;
}
if ($aux =~ m#\((\d+)\)#) {
$o_cqzone = $1;
}
if ($aux =~ m#\[(\d+)\]#) {
$o_ituzone = $1;
}
if ($aux =~ m#\{(\d+)\}#) {
$o_continent = $1;
}
if ($aux =~ m#<(\d+)/(\d+)\}#) {
$o_lat = $1;
$o_long = $2;
}
print "insert into prefix values ('$o_prefix', $o_cqzone, " .
"$o_ituzone, $o_lat, $o_long, '$o_continent');\n";
}
$i++;
}
}
data.sql.gz
Re: [log project] schema
On Tue, May 09, 2000 at 11:02:01AM +0200, Gerard Parat wrote: > As well DX'ers, since this file is used into CLX packet cluster software. BTW, > there is already a link on the site http://clx.muc.de/ Thanks for the link. I was hoping to use the database format from CLX for the DXCC data, and perhaps the import utility. But there is no source code for CLX on the web site, and no license and copyright documentation in the archive or on the web! So it would not be legal for me to look at those files. It's not even clear if it is legal for me to run the program. :-( I just won't work with non-free software on linux, especially for ham radio. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [log project] schema
On Tue, May 09, 2000 at 05:21:32PM +, Wilbert Knol wrote: > The sub-entities thing would be useful, but it doesn't always work. E.g. in the > US I understand the prefix no longer indicates the State. The same is true in > ZL... a ZL4 *could* now be in ZL1. Perhaps an additional field (boolean) to > indicate the sub-entity is meaningful... I will have a look tonight to see how > CT handles that. OK. I've heard that the US is quite a mess. However, I think that the prefix lookup would just offer a default prefix, and that we could override it on a QSO-by-QSO basis (or callsign-by-callsign basis). Would we want to store sub-entities that were not meaningful? I think for the ZL4 example we might do ZL = New Zealand, with general co-ordinates of ZL ZL1 = blah, with more specific co-ordinates, zone etc ZL2 = blah, with more specific co-ordinates, zone etc ... with no record for ZL4 (so the program would fall back to the ZL record). > Also, an IOTA field would be useful (I think it looks like e.g. 'EU-123') so > perhaps best stored as text. Yes.. lots of work populating that field though. > In addition, beam heading fields would be good... long path and short path > (integers), to be populated by an external bit of code at some later stage. I'm not sure I follow. The latitude and longitude fields should be used along with the user's current latitude and longitude to work out the beam headings. I don't think it makes sense to store beam headings in the database, since they are specific to the viewer. > Well, was the country file copyright? If not, let's use it. It is widely used by If it doesn't say what license it uses, we can't assume we're free to use it. > contesters (who also contribute) and is kept up to date to ensure max points > are claimed. Every man and his dog downloads this stuff before the major > contests. Perhaps double-check with the editor of the cty file (AD1C??) > > Why duplicate what the contesters are already doing. Fine if we can get permission. Would you like to contact them? Thanks, Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [log project] schema
Thanks for your detailed comments. I will read through them later today when I have more time. But first quickly answer on part: On Mon, May 08, 2000 at 09:39:41PM +0200, Goran Sandin wrote: > setup a MySQL-server and got it working in my home network. I tried the > schema Hamish posted, but it looks like MySQL doesn't support the same > data-types as in Hamish schema (or I'm doing something wrong :)), which > database is Hamish schema for? Postgres. I suspected that the TEXT type I used for the country name was Postgres-specific. I suppose we will have to change it to VARCHAR(n), but what is a good n for this application? 73 Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [log project] schema
On Mon, May 08, 2000 at 09:33:16AM +, Richard Adams wrote: > On Mon, 08 May 2000, Hamish Moffatt wrote about, Re: [log project] schema: > > On Mon, May 08, 2000 at 03:59:55PM +, Wilbert Knol wrote: > > > The CTY.DAT list of DX entities is available and regularly updated: > > > http://www.contesting.com/ct/files/ > > > > All the links under /ct are broken?! :-( > > Not quite all. > > http://www.contesting.com/ct/files.kr6x I'm sure I checked there earlier and it was empty -- but I see the files now! There is a newer version of CTY.DAT on the www.k1ea.com web site. However, they have gone to some trouble to compile this data so I think there would be copyright problems with pinching it. It IS very comprehensive, although not as comprehensive as it might be -- for example, for each entity it allows additional prefixes. Additional prefixes may optionally override the zone numbers. However, this facility isn't used to correctly determine the zones for Australia. (We are spread between CQ zones 28, 29 and 30.) So the question is: do we want that facility in our schema? ie having the ability to subdivide an entity, with overridden zone numbers. I think it could be useful. Possible modified schema: create table country ( name text PRIMARY KEY, deletedbool, timezone int2, latitude float4, longitude float4, adifnumint2, subentity bool, parent text ); In this case, if the user enters a callsign 'VK3SB', we cut this down to VK3, and find a record in the prefix table with the country name "Australia - Victoria". We look up "Australia - Victoria" and find that it is a subentity, so we look up it's parent. There are a few advantages to dividing an entity -- mostly, for big countries the data is not accurate for the whole entity. Australia covers many CQ zones, and many time zones, and several degrees of latitude and longitude. We might even be able to calculated the US 'Worked All States' award if we divide the US into 50. They even allow particular callsigns to be exceptions. Like U8MIR is a space station, as was R0MIR, as was VK5MIR (not in South Australia!) 4U1UN is in New York, but 4U1?? is in Switzerland (from memory). We could put whole callsigns in with parent pointing to the parent entity. However, I don't think we should pinch this data. If we can find a free source, terrific; if not, develop it ourselves over time. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [log project] schema
On Mon, May 08, 2000 at 03:59:55PM +, Wilbert Knol wrote: > The CTY.DAT list of DX entities is available and regularly updated: > http://www.contesting.com/ct/files/ All the links under /ct are broken?! :-( Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [log project] schema
On Mon, May 08, 2000 at 03:59:55PM +, Wilbert Knol wrote: > The CLX (DX Cluster for Linux) software already has a country database > (postgreSQL). Perhaps we could have used that. Oops. I'll check it out. > Nothing wrong with what Hamish posted though...I will try to load it tonight. > Would keying by prefix speed up the looking-up process? (we're more likely > to want to know where '7O' is, or '4W') True -- but the DXCC list I used didn't unfortunately didn't map 1:1 from prefix to entity. Some prefixes are used by multiple entities. For example VK9Y (from memory), and perhaps also VK0. I was hoping to key it all by prefix but that didn't seem possible. > The CTY.DAT list of DX entities is available and regularly updated: > http://www.contesting.com/ct/files/ Cool, I'll grab that too. The file on the newsgroups was a real mess. It had an amateur prefix, a list of secondary amateur prefixes, and a list of all ITU-allocated prefixes, for each entity. However, some entities were missing the ITU-allocated list. Others were missing well known amateur allocations -- eg the US only listed K; the others were listed as ITU-allocated but not in use by amateurs! So it couldn't be imported directly. I would really prefer a data source which could be imported automatically by a perl program. > I'm interested. The advantage of staying here would be the possibility of > attracting constructive comments. Thanks for your comments Wilbert. I guess we'll stay here until the interest level rises a bit. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
[log project] schema
Hi all, I've been thinking about the database schema for the logging project. The first thing I wanted to do was the schema for the country database. Here's the SQL I've come up with. I got a list of DXCC countries off the newsgroups. There was one line per country, with a list of prefixes, list of CQ zones, list of ITU zones, list of continents, the timezone offset and latitude/longitude. (Some DXCC entities even have multiple continents!) Eventually that's the same scheme I used for this part of the DB. The country table is keyed on the country name. The prefix table is 1:N country:prefix, same for CQ zones, same for ITU zones, Anyone know any good tools for drawing ER (entity-relationship) diagrams and perhaps generating SQL from them? I've also attached the data I used to populate the table (2K lines). There's a field in there for ADIF zone numbers (see http://www.hosenose.com/ for ADIF details), which I haven't filled in yet. All comments welcome. Next task is the main log schema. Should we start a separate mailing list (at onelist?) for those interested in this project? Hamish drop table country; create table country ( name text PRIMARY KEY, deletedbool, timezone int2, latitude float4, longitude float4, adifnumint2 ); drop table continent; create table continent ( countrytext, continent char(2) ); drop table prefix; create table prefix ( countrytext, prefix varchar(3) ); drop table cqzone; create table cqzone ( countrytext, zone int2 ); drop table ituzone; create table ituzone ( countrytext, zone int2 ); data.sql.gz
Re: Yam.
On Mon, May 01, 2000 at 08:18:05AM +0200, Erik Jakobsen wrote: > bash-2.03# insmod ax25 > Using /lib/modules/2.2.13/misc/ax25.o > bash-2.03# insmod yam > Using /lib/modules/2.2.13/net/yam.o It is better to use 'modprobe yam' -- it will insert all the drivers needed by yam. > bash-2.03# setserial /dev/ttyS0 uart none > bash-2.03# yamcfg yam0 io 0x3f8 irq 4 pers 255 txd 150 This looks OK. > bash-2.03# /sbin/ifconfig yam0 44.145.0.9 netmask 255.255.255.0 hw > ax25 OZ4KK up > Segmentation fault That's bad. I don't know sorry. Try doing one thing at a time to narrow it down: ifconfig yam0 hw ax25 OZ4KK ifconfig yam0 44.145.0.9 netmask 255.255.255.0 down ifconfig yam0 up then see which one fails. You need the down on the second line because this line would normally bring the interface up as well as configuring the IP. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Yam.
On Sun, Apr 30, 2000 at 08:43:03PM +0200, Erik Jakobsen wrote: > linux:~ # yamcfg yam0 io 0x2f8 irq 3 pers 255 > linux:~ # /sbin/ifconfig yam0 44.145.0.9 netmask 255.255.255.0 hw ax25 > OZ4KK up > SIOCSIFFLAGS: Permission denied > SIOCSIFFLAGS: Permission denied > linux:~ # > > > Whats the reason I'm getting that message ?. Probably your serial driver is still using that port. The yam driver uses the serial hardware directly. Before using ifconfig, do setserial /dev/ttyS1 uart none which will release the port. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Wed, May 03, 2000 at 02:39:37PM +0200, Hans-Peter Zorn wrote: > I am using the kde 1.1.2 .debs for potato on a mixed potato/woody > box. OK. I obtained the kde debs from the maintainer's site. Now dhlog won't pass ./configure; it ays it can't find the KDE libraries. There is no --with-kde= parameter to specify the location; the configure script does mention /usr/local/kde, but Debian's KDE is in /usr/include/kde, /usr/lib etc. > I think these are the ones from the kde site. I have only the > libraries because I like the API KDE provides, but not the > entire desktop. For most KDE apps you need only kdelibs. (Which > are LGPLd by the way, therefore the licence is not problematic > as it is for kdebase). That's good news. Thanks. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Sat, Apr 29, 2000 at 12:47:51PM +0200, Hans-Peter Zorn wrote: > Browsing radio.linux.org.au I find at least 3 logging programs based > on MySQL: > > MyVRLog : http://www.computertime.de > DHLog: http://www.home.unix-ag.org/holgu/dhlog/indexe.html > QtLog: http://www.ewetel.net/~dieter.kumpies/soft.htm OK well I've been trying these programs. QtLog is in German and I cannot get MySQL set up correct. MyVRLog is for KDE and also appears to be in German. DHLog is also for KDE. Debian does not have KDE, and the KDE site does not have .debs for the current Debian version. :-( Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Mon, May 01, 2000 at 09:51:07AM +0200, Gerard Parat wrote: > So, there is at least two kinds of logging software that maybe designed > with completly different specifications. I don't think a universal > software is a good solution because of the trade-off. I agree. However, it should be possible to have a common database format for the two of them, with different front ends. They are not mutually exclusive, just different with many things in common. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Sun, Apr 30, 2000 at 09:08:14AM -0500, Nate Bargmann wrote: > On Sun, Apr 30, 2000 at 11:37:02PM +1000, Hamish Moffatt wrote: > > > > All suggestions welcome. Testers welcome too! > > When do we start, Hamish? The first thing we need is a database schema. I'll try to write something this week which addresses my needs, and let others add to it whatever they think is missing. Or somebody else can do the same; I won't get to it for a few days yet. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Sun, Apr 30, 2000 at 01:39:53PM +0100, Geoff Blake wrote: > As I tend to get involved with special event stations (GX0MWT - Marconi > Wireless Telegraph for International Marconi Day yesterday) the provision > of a UPS for the server is less of a problem. However if the "front end" > could work with Lynx :-) and IE99 :-( it would be great. Well, I don't think Lynx would be the front-end of choice for speed! It's a nifty program, but navigating around is not particularly quick. > Also, the British authorities, maybe others too, but I doubt it, require > that if an electronic log is used, it is kept on independant media (a disk > devoted to that purpose?), the ability to mirror to a floppy or even a > tape device may be useful. Sounds like they are very strict about logging there. I don't remember anything in our Australian regulations which requires me to log my transmissions. I do so only for interest's sake. I don't bother logging when I talk to my mate 20km away though, or any local VHF contacts even. > I am not a programmer (to me C is the letter befor D) the only way I can > contibute to this is by suggestion, helpful or otherwise! All suggestions welcome. Testers welcome too! Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Sat, Apr 29, 2000 at 07:08:44PM -0500, Nate Bargmann wrote: > Also, mentioned in this thread was rig control. IIRC, some months > back a post was made to the list of someone writing a library for ICOM > rig control. Perhaps this could be extended to other models. I think > it would be a good idea to develop such a library so that any programmer > doesn't have to re-invent the wheel each time rig control is needed. I have recently been thinking about do exactly that. ICOM's protocol is well documented; I have the CI-V manual in PDF format here. I now use a Yaesu though and haven't investigated the protocol there yet. I do know that ICOM's is very similar between radio models, while Yaesu's differs a lot between models. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Sat, Apr 29, 2000 at 08:50:49PM +0200, Goran Sandin wrote: > Please check www.hosenose.com/adif/adif.html for specification of Amateur > Data Interchange Format. > > There you have a standard of how to interchange data between different > log-programs. It would be very nice if a log-program for Linux also > supported this standard. That will be one of the first things I code, if someone doesn't beat me to it. Currently, I am maintaining my log in ADIF format by hand. :-) Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Sat, Apr 29, 2000 at 08:09:48AM -0300, John Ackermann wrote: > year when we go out for the US Field Day operation, I agonize over > whether to switch to computer logging and sending (there's also an > ergonomics issue -- moving hand from keyboard to paddles is > inconvenient and time consuming; sending by hand, I keep the pencil in > my left hand while sending with my right...). Good point. > This is probably not a good idea. Many (most?) rigs are only optimum > for CW receive when set to CW mode -- operating in SSB will lose a lot > of RX capability (narrow filters, fast AGC, and proper zero beat). Also a good point. There's a Hellschreiber mini-contest on this weekend. Narrow IF filters are desirable to minimise QRM from other modes. My FT-847 will not allow use of CW filters on SSB receive, so I have to operate split (CW narrow on receive, SSB on transmit). Works well enough but is frustrating. At least the FT-847 automatically adjusts the frequency when you switch from SSB to CW; my previous rig (IC-726) did not and I had to adjust it by hand. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Sat, Apr 29, 2000 at 12:47:51PM +0200, Hans-Peter Zorn wrote: > MyVRLog : http://www.computertime.de > DHLog: http://www.home.unix-ag.org/holgu/dhlog/indexe.html > QtLog: http://www.ewetel.net/~dieter.kumpies/soft.htm > > Perhaps you can try to avoid duplicating work. These log programs are > all Qt based, so they may be problematic for some people, but > I am sure that the authors of above programs are interested in > collaborating at the backend side. Backend and GUI should be seperate > anyway in my opinion. We (especially Stefan, DL1ELY) considered > starting a logging project ourselfs some time ago, but we are too > busy with other things currently. Thanks Hans-Peter for the pointers. I have looked at QtLog but all of the program and documentation is in German so I could not get the mysql database created as it wanted. Most menu items would segfault because the database was not set up. Also, I had to spend about an hour changing many lines of source because it won't compile as-is with the latest g++. I haven't looked at DHLog or MyVRLog yet because I think both are quite new. I looked at radio.linux.org.au about a week ago and they weren't there. Most of the logging programs on that database are contest programs, or text mode programs. Neither interest me. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Fri, Apr 28, 2000 at 11:05:11PM +, Wilbert Knol wrote: > However, it would be nice if the logging software offered basic CW features. Well, if we decide to work together on this, I think we will just have to make the framework general enough to support everything people want to work on. Personally, I would never use these features, but my idea of working the CQ WW SSB is to make a handful of contacts one evening then turn it off for the rest of the weekend! > The reason is, that a lot of CW ops use a computer to both send CW and > also log with it. The CW 'keyer' needs to know the call of the guy being > worked, in order to fire it back at him, and also the RST report you have > decided to give him (as typed into the RST-sent field). Hmm. I'm a bit surprised (even shocked) to know that these guys let the computer do most/all of the sending for them. > 1. A CW keyer on the parallel port with programmable speed. Personally, I would prefer to generate the CW as audio tones injected as SSB. > Rig (CAT) control for the major radio interfaces (you'll need a lot of beta- > testers :-( Yep. That's something I was thinking about working on anyway before this logging discussion got started. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Fri, Apr 28, 2000 at 09:34:40PM +1000, Carl Makin wrote: > I'm just about to release my Perl/Tk based Converse client and I don't think > Perl/Tk is as portable to Windows as you'd like. :( Oh. :-( I've seen a fairly simple app in Perl/Tk running on both platforms very smoothly, but nothing more complicated. A little off the topic -- the Perl/Tk tutorial says you can use the form -blah => 'value' as parameters to Perl/Tk functions, but perl with 'use strict' enabled requires me to use '-blah' instead. Is it just me? > What about Java? It's OK. I'd consider it a bit more if all the tools were available open source though (compiler, JVM and standard classes). There is a lot of development work going on but nothing of production quality last time I looked (8-9 months). Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Thu, Apr 27, 2000 at 05:10:34PM -0700, Shawn T. Rutledge wrote: > What form would you like to use for QTH? lat/lon, just the city > and country, grid square, ??? I would think it would be best to allow > any or all of these but not to require them. I like just a text field for city name (or perhaps city name + other info like "Blah Blah 30km W of So-and-So", plus grid square in a separate field (required for VHF+). > Typically in this kind of situation I would provide a drop-down list > on the form with all the known choices so far, and the last choice > is "other", and there is an adjacent text field to type in a new one. > After that it becomes one of the choices so you never have to type it > again. Sounds excellent. > Cool. And maybe the type of radio, antenna, elevation, tower height, > direction the antenna is pointed if it's directional, ... or is a > plain comment field enough for all this? I personally don't tend to log those details at all; when I do, the comment field is enough. But since I'm fussy about a lot of other fields, there could be people fussy about these ones too. > > Contesting? Difficult to build a program that does both contesting > > and casual logging well I think. > > Yeah. If you're in a hurry then maybe web forms are not as efficient > as a conventional GUI. But what else is different? You need auto-incrementing serial numbers for some contests. Co-ordinated between operators in multi-operator stations. Nothing else is different for entry, but more reports are needed. Thanks Shawn, I'll have a look at your web site. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Thu, Apr 27, 2000 at 08:35:10AM -0600, Brian Hall wrote: > Delphi for Linux should be available later this year. I have requested to join > the beta program, but they haven't gotten back to me with an NDA yet. From what > I've read, Delphi for Linux (code name Kylix) will use the QT libraries, so it > will mesh well with KDE. Not sure yet if Kylix will use QT 2.0+ or the 1.4x > series at first. > > I'd suggest writing it in Delphi/Win32, with the idea of porting it when the > Linux version becomes available. Do Delphi apps currently run under Wine? Good suggestion Brian -- however, it's not free (in the open source sense), so it's not of interest to me. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Thu, Apr 27, 2000 at 03:42:40PM -0700, Ken Koster wrote: > On Thu, 27 Apr 2000, Hamish Moffatt wrote: > > Contesting? Difficult to build a program that does both contesting > > and casual logging well I think. > > One frontend yes, one database no. Agreed. I don't think a web front end would be fast enough for contesting. > I can see a front end optimised for casual loggins, one for contesting, one > for detailed logging, one for Net operations, etc. All from the same database > so I can be a casual logger with one interface, and still have a seperate > interface to handle Net operations or the occasional contest. That will require a lot of design work on the database, but I think it is well worth the effort. > Shawn (kb7pwd) has mentioned his nettebook program. While I've not had > a chance to actually run it yet, it looks like maybe a good starting point. > Besides. he obviously understands this database stuff better than I do. :-) > I'm an embedded systems guy, it's not often I need a fancy DB. Well, I'm a hardware designer fresh enough out of university to still remember database design (I think so, anyway :-) Now I'm inspired to start coding. The more I think about it, the more I like the web interface, because it simplifies the UI design so much. (No need to do battle with any windowing toolkit; just generate HTML and let the browser worry about rendering it.) Some things will be challenging to implement. Radio control, for example. Certainly possible, but not as easy to link in perhaps using the web method. Lots of Javascript or Java perhaps. DX cluster support might also be challenging. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Thu, Apr 27, 2000 at 10:36:39AM -0700, Ken Koster wrote: > Using MySQL (or PostgreSQL) is a great idea and one I think is fundemental > to doing it right. Rather than concentrating on what GUI to work with why > don't we identify what elements need to be stored and come up with a database > structure first. Then multiple frontends can be created depending upon an > individuals requirements. > > I'm not a database type of programmer, but I do know that a proper database > structure has to come first. Agreed. Well, the list of fields I want is quite big. None of the Windows programs has all of them, which is part of the reason I'm considering home brew. Things commonly lacking in existing log programs * no separate name and QTH fields (some programs expect you to put these in the comments; I prefer separate fields) * mode field of 3 or 4 characters with a fixed list (I mostly operate Hellschreiber which has a lot of sub-modes like "PSKHELL245" and "FELDHELL", just as SSTV has "Martin M1", "Robot 36" etc, and I would like to log the exact mode used). I also like to log the transmit power used both by my station and the remote station. (This is getting into the very picky category.) One program I do like for Windows has a field for my transmit power, but not the other op's. Contesting? Difficult to build a program that does both contesting and casual logging well I think. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: XML configuration file
On Thu, Apr 27, 2000 at 12:32:37AM +0200, Jens David wrote: > But still I thing this is solely the distro´s problem. Applications and > tools should never read such a file directly or via libax25. They should > use the socket interface as all other protocol families also do. That sounds reasonable. Debian has a nice new /etc/network/interfaces file which describes network interfaces and this information could be incorporated in there. The developer has always been mindful of ham radio needs during discussions. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Thu, Apr 27, 2000 at 07:17:44AM +0930, Andrew Hodges wrote: > One of the applications I am particularly wed to is a log program. I have been > using Logic, and have been perusing the web to locate a linux alternative. Logic is very big and very flexible. I don't think you'll find anything like it. Personally, I have tried the demo and found that the program is so generic that it is difficult to do anything at all with it; everything is abstracted. But I have only spent 5 minutes with the demo version. It is fairly pricey too, and of course non-open-source. > Ideally I would be looking at one that was a front end to MySQL or something > like that, and I am also looking for rig control and access to RAC CDRoms etc. Well, if you decide to code something yourself, please contact me by email as I'm certainly interested. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Log Program for Linux
On Thu, Apr 27, 2000 at 12:59:24AM -0500, Bryn Joynes wrote: > This is not a suggestion but an idea I have had for a while. I also > need a logging program but I also need one that is portable between > Linux and Windows. I was thinking of using MySQL as the database with a > web browser as a front end running perl and perl-dbi scripts. The only > problem is the lack of time at present to sit down and write the code. > > Maybe someone with more programming knowledge might like to run with > this idea. I have been thinking about logging software for linux recently too. I considered the web/database backend approach but thought it might be a bit too slow and not as flexible as a real program. I'm undecided about which is the best GUI to work with. I started hacking some stuff using Perl/Tk, which is not bad. (I use Perl a lot for my day job.) There's also Perl/GTK+, although Perl/Tk is more portable to Windows. Then there's the matter of the backend; Perl has a database interface called DBI which is very flexible; I don't know if it's support on Windows. There is DBI to talk to MySQL, postgres, ODBC etc and even to talk to flat text files. So I am currently considering Perl, Perl/Tk and DBI for the job. Writing GUI applications in C/C++ is not my idea of a fun weekend, no matter which toolkit you use (Motif, GTK+, wxWindows etc). There's a lot of stuff to do. Basic record keeping is one thing, but then there's reports for awards and so on. Still, I haven't found a program that suits me 100% on Windows, and if I'm going to write something it will certainly be on Linux. (Even though I could it in Windows using Delphi in no time.) 73 Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: [ANNOUNCE] new-AX.25 for 2.2.14 Rel. 5
On Wed, Apr 26, 2000 at 07:32:06PM +0200, Joerg Reuter wrote: > We need to differ between the tools, the applications and the library. > My idea is to get rid of some programs altogether (axparms for example) > and provide the functionality through procfs. Some tools (those that > use AF_PACKET to monitor traffic) need to get updated -- the DDI has > changed, there is no prepending KISS command byte anymore. That's > annoying, but only affects very few tools: AFAIK only listen, net2kiss > and ax25rtd. The later will use the netlink interface, I don't know And some applications -- aprsdigi, aprsmon, aprsd at least. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: no sign of radio.linux.org.au?!
On Tue, Apr 25, 2000 at 12:08:56AM +, Terry Dawson wrote: > Hamish Moffatt wrote: > > > > radio.linux.org.au seems to have disappeared from the DNS! > > Does anyone know where it's gone? I can't remember TerryD's email address. > > Sorry. Yes we're experiencing DNS problems. It's a bit of a cock-up > frankly, our primary failed and that wasn't a good time to discover that > our secondary had never actually been configured! > > We're working on getting it back again, but the Easter weekend isn't > making it easy with people away. No problem, Terry. I discovered a way to access it anyway, not that I recommend it -- since radio.linux.org appears to be a virtual host of www.linux.org.au (which IS in the DNS), I added an entry for radio.linux.org.au to my /etc/hosts with www's IP. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
no sign of radio.linux.org.au?!
radio.linux.org.au seems to have disappeared from the DNS! Does anyone know where it's gone? I can't remember TerryD's email address. thanks Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: problem with sound card modules:
On Sun, Apr 23, 2000 at 10:24:16PM -0400, Pat Masterson wrote: > If instead I try "insmod sb io 0x220 irq 5 dma 1 " I get: > /lib/modules/2.2.13/misc/sb.o: init_module: Device or resource busy Pat, The format is "insmod sb io=0x220 irq=5 dma=1"; try that and let us know how you get on. 73 Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: About ARPSD..!!
On Sat, Apr 22, 2000 at 05:01:07PM -0300, Jose R. Marte A. wrote: > Hamish I have made the patch to the aprsd 2.1.1 > for the support of the ax.25 tools in RedHat 6.1 > and in a same way I made it with the APRSD 2.1.0, I will bring out a new version of the aprsd mods for 2.1.1. But it should work OK just patching them as you did. > socket: Invalid argument > > and it exit, I was looking in the file sockets.ccp > in the following line: > > / * Create the receive socket * / > if ((rx_socket = socket(AF_PACKET, SOCKET_PACKET, htons(proto))) == -1) > > if I change the AF_PACKET for the AF_INET, the program works in TX > via RF Link but non RX in the RF Link, in RedHat 6.1 as AF_PACKET yes > it works using the ax25 tools in RX > > My main Question is your patch works for the RedHat 5.2 ?? AF_INET is for TCP/IP and not for AX.25 though. I think the problem is that you are not running aprsd as root -- it must be run as root to create the AF_PACKET socket if I recall correctly? (Just as "listen" must be run as root -- and listen is where I got the above code from originally.) Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Twpsk
On Fri, Apr 21, 2000 at 05:49:29PM -0600, Bryan Curl wrote: > Hope all is well. > Question. Why does my Twpsk receive screen jump to the top all the time? > I t only odes this after I scroll back, once, then try to go to the bottom. > At the end of each received line it jumps to the top. > Just a note: Also notice that CTRL functions dont work if numlock is on. Another PSK31 program you might like to try for Linux is 'phaseshift.' Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Programing in C...
On Thu, Apr 20, 2000 at 08:02:32AM -0700, Michael Anderson wrote:
> if you just want a beep, then try for more portable code:
> #include
>
> void main(void)
> {
> printf("\7");
> }
Actually, printf("\a") would be more portable. \7 assumes your computer
is using the ASCII character set.
If you want to get really picky, a lot of ANSI C compilers will not
compile "void main" because it is not a valid prototype for main.
main must return an int.
Hamish
--
Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: logging
On Mon, Apr 17, 2000 at 08:58:32AM -0700, John van V. wrote: > For the squillion-th time: > Perl is the way to go here. You can use TK/perl for X, or go w/ Gnome's GTk. > My offer still stands !! > I have a perl primer on line: > http://puny.vm.com John -- I program perl every day for a living at the moment, so I'm quite familar with it. But unless there's a core module I don't know, perl out of the box does not do QSO logging! I could certainly roll my own logger if I wanted to, and Perl would be the tool of choice (especially with Tk and DBI), but I was hoping for an application someone had already written. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
logging
At the risk of starting a flamewar (which is what happens on rec.radio.amateur.dx if anyone asks this question), can anyone recommend a good logging program? I've looked on radio.linux.org.au and most are text-mode, or contest focussed. I want something for a casual operator. I want X11, and free. I've seen Qtlog, but I couldn't get the mysql stuff set up correctly and all the program text and documentation is in German. To be really picky, I want the QSO mode field to be big.. I want to store obscure mode names like "PSKHELL245". All the Windows stuff I've seen is limited to 3-4 characters, which assumes the only three valid modes are SSB, CW and RTTY. Or I want the field to be user-sizable. But since I have source code I can fix it myself if necessary. I would like ADIF input, but again I can always hack that up. Awards checking is nice to have. Contest scoring also nice to have. thanks. hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Bad AXPORTS
On Wed, Apr 12, 2000 at 11:57:36PM -0700, [EMAIL PROTECTED] wrote: > sm0 Link encap: AMPR AX.25 HWaddr N7XZV-1 > cat /etc/ax25/axports > # /etc/ax25/axports > 2mN7xZV-1 1200255 2 Test Port ^ Looks OK apart from that. > smdiag. It would be neat to have more modulations standards like newQPSK > etc. newqpsk is a soundmodem for linux too. It runs in user space and talks to the mkiss driver. Once running it's a standard AX.25 device. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: aprsd 2.10.vk3sb.5 with validation source code
Hi Bruno, On Wed, Apr 12, 2000 at 01:08:26PM -0400, Bruno Quesnel wrote: > Great, but I think after discussion with Steve on this subject, he > wanted the whole algorithme to be changed... This is maybe why he > released the code > > Again, I plan to start working on this early next week or the next > after. When I'll start, I'll let you know I'll probobly have a test > version for you to test some day. Will be a pleasure to work with you in > the future weeks to come, if you still want to work with me. Hope > so. We could do some neet things together... I have some ideas for a new validation method. When I have it all detailed I will make it public. Some changes to aprsd will be required but nothing major. Dale is going to publish a 2.1.1 version which has the validation source and some other bug fixes. I will add my changes to that. That would be the best version for you to start with. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
aprsd 2.10.vk3sb.5 with validation source code
I've updated my modified version of aprsd to 2.10.vk3sb.5. It now includes source code for the validation function, as posted by Steve Dimse to this list. Now that source code is available to the whole program, I'll be working on a package of aprsd for Debian GNU/Linux. You can download my modified version (which also includes support for Linux sockets and some wildcard support in the config file) from http://aprs.cloud.net.au/aprsd.shtml 73, Hamish -- Hamish Moffatt VK3SB. CCs of replies on mailing lists are welcome.
Re: Unproto Path in AX25
On Mon, Apr 10, 2000 at 09:23:19AM -0500, John Gorkos wrote: > Unfortunately, with APRS being a UI-only protocol, my beacons don't get > propagated through the APRS system. I need to be able to do something > like this: > beacon -d "APRS via WIDE,WIDE,WIDE" ... That should work fine, use "v" instead of "via". No need to use modified ax25tools. I may have had to patch aprsdigi to get it to compile here, I can't remember. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: testers needed for aprsd mods
On Tue, Apr 11, 2000 at 09:22:59AM +0100, Jorge Matias wrote: > I'm interested. I have already a few APRS stations on the air ready to > help me on that. Jorge, I replied to your email directly but it bounced with an "access denied" message. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: VC and datagram mode?
On Wed, Mar 22, 2000 at 01:12:40PM +0100, Thomas Sailer wrote: > But IP expects the phy layer to have low packet error rate, which > is usually not the case on radio links (IP cannot distinguish > packet loss due to bit errors and due to congestion, and assumes > the latter). > > So for links over more than one radio link, VC with hop2hop ack > (and redundant frame remover) is vastly better than DG. OK. But standard AX.25 doesn't have hop-hop ack, does it? Only NET/ROM does? Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: VC and datagram mode?
On Tue, Mar 21, 2000 at 09:37:54AM -0400, Jose R. Marte A. wrote: > In datagram mode, IP packets are encapsulated in AX.25 > (UI frames) and transmitted without any other link level mechanisms, > such as connections or acknowledgments. > > In vc (virtual circuit) mode, IP packets are encapsulated in AX.25 > (I frames) and are acknowledged at the link level according to the > AX.25 protocol. Link level (i.e. AX.25) connections are opened as > necessary. > > In both modes, ARP is used to map IP to AX.25 addresses. > > I hope this info help you So which is better? I'd guess datagram mode is better, because IP packets are datagrams themselves. IP does not expect the layers below to be reliable. TCP really maps to a connected AX.25 session, and UDP to UI frames. IMHO. Hamish -- Hamish Moffatt VK3SB <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: Slackware
On Wed, Mar 01, 2000 at 08:23:21AM +, Richard Adams wrote: > I belive slackware.org is the home of Walnut Creek and not slackware, > slackware.com is the home of slackware. > They both point to version 7.0. Walnut Creek and Slackware have been very much linked for years. Hamish
Re: Slackware
On Tue, Feb 29, 2000 at 07:08:41AM +, Ken Adams wrote: > direction of this so called Slackware 7.0. Does it use the new 2.4 > kernels? I am using 2.2.6 upgraded by hand to 2.2.13. I think even Linus would be surprised to know that Linux 2.4 is out. They are working on 2.3 now, which is the development version of 2.4. Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: NEW-AX.25 (was: FLEXNET flamewar)
On Thu, Feb 24, 2000 at 10:25:17AM +0100, Hans-Peter Zorn wrote: > format out there. And - I can tell you from experience - the > average Linux beginner of these days doesn't know how to > compile from source. Fortunately this is amateur radio and we have always expected some basic background knowledge. For example you still need to know how to erect an antenna and cable it, how to plug your radio into the power supply and computer etc. hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: QSSTV
On Sun, Feb 13, 2000 at 04:40:30AM +0700, robby wrote: > Hello, > I need to have RPM version of qsstv for redhat 6.1, since I always fails to > compile the tgz version I did a .DEB of it a while ago, maybe you can convert that to RPM. I need to resurrect it, as it was done last October and libraries have changed since then. Hamish vk3sb
Re: Lart as a TNC
On Mon, Feb 07, 2000 at 11:12:37AM -0500, Gregory Maxwell wrote: > Has anyone here considered making a run of Larts > (http://www-ict.its.tudelft.nl/~erik/open-source/LART/) for use as > advanced TNCs? It could be one of the most expensive TNCs ever made! I'm sure it wouldn't come cheap. Is anybody making these commercially? Hamish
Re: axports not found
On Sun, Feb 06, 2000 at 08:57:15PM -0500, John Ackermann wrote: > I just got bit by this kind of problem, too. At least in Debian, the > pre-compiled packages install themselves into the normal /etc/ax25, > /usr/sbin, etc. But roll-your-own packages are using the /usr/local > tree. The new version of call that Bob Meyer put together does this. > I ended making /usr/local/etc/ax25 a symlink to /etc/ax25 and that > solved the problem. The configure script for libax25 has a few parameters which let you tell it where its config files, program files and data bases go (individually). You may need to use something similar when compiling other programs -- tell it the configuration files are in /etc, but you still want the binaries in /usr/local (because it's not a Debian-provided package). By the way if the program is accessing the standard libax25 config files, it should do it using just libax25 -- so it should not need to know the location of the config file directory. I think, anyway. 73 Hamish vk3sb -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: pa4ab
On Mon, Feb 07, 2000 at 12:16:21AM +0700, astari wrote: > i want to ask is can we implement the soundmodem with ssb(single side > band) so our region will be more big ? > > how we can control the outpot of the soundmodem what command to use > smmixer ? and what is the best setting for it ? > > and how should the good outpot from soundmodem if we look it with smdiag > with eye diagram or else how should the picture ? Sure, you can just put the radio on SSB.. if you use 1200 mode, you will end up with real FSK on the air. No need to tell the soundmodem any of this. Unless you want 300 bps -- which AFAIK isn't supported. Sorry I can't help with the eye diagram -- although it was mentioned on this list a while ago that the eye diagram is irrelevant for 1200 anyway. 73 Hamish VK3SB
Re: Routing problem
On Fri, Feb 04, 2000 at 03:28:49PM +0200, Tomi Manninen OH2BNS wrote: > I'm not sure but I don't think the subnet / netmask issue is of relevance > here. The main point of the problem is that Linux 2.2 seems to refuse to > send out an IP datagram on the same interface it came in. At least with > AX.25 interfaces. I tried to test it with ethernet; I think it worked but I couldn't be 100% sure. Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: Routing problem
On Thu, Feb 03, 2000 at 07:36:57PM -, Robert A Jenkins wrote: > For machines within the same subnet to work via an intermediate system, you > must add specific routes to the target machines with the intermediate as a > gateway. I set mine up in rc.local. Well, they are not really on the same subnet then if an active intermediate is required. Can you change the subnet mask to the correct value, which may even be 255.255.255.254? (I don't know that such a mask is actually supported to be honest..) Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: Hardware TNC (Baycom etc.)
On Tue, Feb 01, 2000 at 03:53:58PM +0100, Gerd wrote: > There are some good reasons to still use something like BayCom in our days. > The first reason is: There are still frequencies that one cannot operate more than > 1200 bps on, due to bandwidth limitations, for example. soundmodem supports 1200 AFSK just fine. I use it at home. > The second reason is that the SoundModem driver must be considered experimental, and > is sometimes not easy to set up. Even more, it seems that Thomas's support for that Perhaps it is not easy to set up, but it is quite reliable once configured. I have not had any problems running smdiag on linux 2.2.12, although I did not try to compile it with that kernel (but used an older binary). I think Craig has said it will return in the next ax25-tools. > The fourth reason is that the (few) soundcard designs supposed to work with > SoundModem are not available any more. That is because ISA cards are generally Yes this a real problem. Somebody needs to work on it; it's open source, so it doesn't have to be Tom. Ideally, it would work with the standard sound drivers (either OSS or ALSA). Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: 9600 packet query
On Sun, Jan 30, 2000 at 07:05:48PM +, Ing. Jose A. Amador wrote: > I wonder if it would help to point the interested to > www.febo.com/layer-one, or sort of it, I am typing by heart. John has put > together a number of advices and experimental/measurement reports that is > really interesting to read, since without layer one support we could not > make our software to communicate via radio. That's an excellent web page; I used it a few weeks ago to set the levels on my soundmodem setup. As a result my hit rate into the digipeater is very close to 100% now; before it was pretty poor. He also makes some comments about particular radios -- I have a Kenwood V7A which he talks about specifically. As a result I ended up putting an RC filter for pre-emphasis purposes inside the DIN connector for my TNC :-) I will have to revisit it to see what he says about 9600 FSK. Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
9600 packet query
I confess this is not entirely linux related. I've never been on 9600 packet before but now that I have a 70cm FM rig I was eager to try it out. It's a Kenwood TM-V7A. I have a YAM which I built ages ago and haven't really used. So I connected up the YAM to the V7A (shame on Kenwood for using 6-pin mini din plugs, obscure AND hard to work with). It works great on 1200 packet, but nothing happens on 9600. Well, to be honest it's late at night and there isn't much activity, so I can't really tell what callsigns are valid. But my main question is: I have a 70cm hand held which I tuned up on the same frequency as the V7A. Should I expect to hear anything on that rig when transmitting packet? When the YAM transmits, the V7A transmits OK, but the squelch doesn't always open on the handheld. (It does if I key the microphone on the V7A though.) I figure this is reasonable since it's not really transmitting FM, but FSK. Right? So what should I do with the squelch on the V7A for receiving? thanks Hamish VK3SB -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: SoundModem
On Thu, Jan 27, 2000 at 03:55:15PM -0800, Cathryn Mataga wrote: > Yeah, I run Soundmodem on an SBPro, and I found that the default > mixer settings for that card came out a little screwy. Or it was set > to receive input on the mike input jack rather than the line input jack > -- something like that. Yep -- the default is the mic jack. On SB16, you can have multiple devices selected for recording, so the s= notation is a bit weird. Not that you would want to receive from multiple devices in this case. As I think we've discussed here before, I have to use a modified smmixer on my SBPro packet system, because it doesn't let you set the line in volume, which defaults to -46 dB. Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: SoundModem
On Thu, Jan 27, 2000 at 03:04:54PM +0100, Mitja Bezget wrote: > I set up two soundblasters (modems) on two computers. When I wanted > to test it with a cross-wired audio cable (no radio RTX) it just > didnt work. > When I run a packet thru sm0 I can measure it with an oscilloscope. > It looks pretty nice. But the computer on the other side just does not > respond! I have tried all modulations.. > I wanted to fine tune the i/o levels with smmixer/smdiag but smdiag > didnt work for me. The screen was always blank with a straight horizontal > line (one pixel wide). I tried all command line options but at the end I > gave up and tuned the output level with an oscilloscope. I tried several > settings for input level but it still didnt work. > I also did NOT specify PTT... Is it necessary in my case? > I'm using 2.0.38, AWE64, AWE32, P/133, P2/233, axtools-2.1.42 ... You do not need to specify PTT in this case, the sound modem doesn't mind if you don't have PTT. Perhaps the mixer settings on the receiving sound card are wrong? Either the levels are wrong or the line in is not selected for recording. Something like "smmixer s=line" (on SB16) should fix that. That would be why smdiag doesn't show anything. Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: AX.25 Configuration HOWTO
On Thu, Jan 27, 2000 at 10:18:43AM +, Terry Dawson wrote: > Gerd wrote: > > one there. For people that prefer "make menuconfig" or "make xconfig" the > > make-kpkg tool seems to be not usable. > > Gerd, > make-kpkg doesn't rely on "make config". > > You "make menuconfig", or "make xconfig" first, then you run > 'make-kpkg'. There is no need to use "make config", that is just what > "make-kpkg" uses if it finds a kernel source tree without a > configuration file. That's correct, but there is a catch you have to be wary of. When you run make config|menuconfig|xconfig, the makefile creates symlinks for the asm and system include files for the appropriate architecture. If you run "make-kpkg clean" after configuration, it removes those links -- actually, it runs "make mrproper" but keeps your config file safe (mrproper would usually remove .config too). So then "make-kpkg kernel_image" will fail because of the missing links. There's no dependence on the text-mode config -- I never use it, always menuconfig. Hamish vk3sb -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: AX.25 Configuration HOWTO
On Wed, Jan 26, 2000 at 05:47:46PM +, richard bown wrote: > As for the ramdisk for start up never found it needed on RH, but I'm curious > as to the preference to use bzlilo , instead of bzImage. I've always used > either zImage or bzImage depending on the kernel version, and then edited > lilo.conf manually before running lilo. > I've also noticed that on the 2.2.13 kernel and later ? > there is no need to run make dep, and then make clean. > Any chance of some clarification on these please ? On Debian, 'make-kpkg' is the tool of choice. It compiles the kernel and modules and builds a .deb package, which you then install with dpkg. This makes it absolutely simple to build a kernel package for a different machine -- I prefer to build kernels for most of my machines on one PC, for speed reasons. 73 Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: Sattrack
On Tue, Jan 25, 2000 at 01:01:26PM -0500, Edson Pereira wrote: > Can someone point me to the sources of sattrack-3.1.6 or 3.1.7? Also, in a > previous message there was a mention of a patch for the y2k issue. Where > can I find the patch? You can download my DEBs or RPMs converted from the DEBs at ftp://ftp.rising.com.au/pub/hamish/ Whether or not you'll like my arrangement with the keps is another thing. I'm not even sure I like it myself. I don't think there is a 3.1.7 -- only 3.1.6. Here's the Y2K patch if you want to build it from source. It's from Rob Janssen. Hamish diff -cr SatTrack-orig/src/sattrack/sattime.c SatTrack/src/sattrack/sattime.c *** SatTrack-pe1chl/src/sattrack/sattime.c Sat Jan 4 22:19:08 1997 --- SatTrack/src/sattrack/sattime.c Sat Jan 1 14:19:51 2000 *** *** 269,275 if (gdnY < 50) /* allow 4 or 2 digit year specifications */ gdnY += 2000; else ! if (gdnY < 100) gdnY += 1900; result = (long) gdnY-1901)*1461) >> 2) + monthDays[gdnM-1] + gdnD+365); --- 269,275 if (gdnY < 50) /* allow 4 or 2 digit year specifications */ gdnY += 2000; else ! if (gdnY < 200) gdnY += 1900; result = (long) gdnY-1901)*1461) >> 2) + monthDays[gdnM-1] + gdnD+365); -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: New AX.25 and newest kernel 2.2.14: first test results
On Mon, Jan 24, 2000 at 11:43:06AM +0100, Gerd wrote: > > I'm a Debian developer, I'm not running slink. > So there are plans to use 2.2.x kernels on Debian, too? You can run 2.2 on Debian 2.1 (slink). It works just fine, out of the box; almost all of the tools support 2.2 and 2.0 (deliberately). There may even be 2.2 sources included, although there are no 2.2 compiled kernels. Debian 2.2 (potato) has Linux 2.2.14 (or will do soon). Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: New AX.25 and newest kernel 2.2.14: first test results
On Mon, Jan 17, 2000 at 08:04:16PM +0100, Thomas Pinz wrote: > > connect: No route to host > > > > Here I have no idea how this can be fixed, too. > > there are some changes in the routing behavior... if you add a "axparms -route > add ", it > works... but here the problem appears only with "call", linkt for example have no > problem. I get "socket: invalid argument" when trying to use call or axparms as above. However I'm not 100% sure my recompile of ax25-* actually uses the new code. I did recompile everything but I think /usr/include/netax25/ax25.h is still the original. On Debian that file is provided by libc6-dev and not by libax25-dev it seems, which means it was probably not updated. Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: New AX.25 and newest kernel 2.2.14: first test results
On Thu, Jan 20, 2000 at 06:28:47PM +0100, Gerd wrote: > > > Pentium class computer, preferrably 300 MHz and above > > > 64 MB RAM or more (the more, the better) > > > > These are wildly exaggerated requirements for Linux. > > But in a lot of publications and also in a lot of user statements in the > Usenet, repeatedly, such requirements were announced. Depends what you want to do with it. If you want an environment something like Windows eg: GNOME, X, perhaps a word processor etc, those requirements aren't too inflated. They don't apply in all situations obviously. Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: Sattrack
On Thu, Jan 20, 2000 at 06:09:28PM -, Eamon Skelton wrote: > Has anybody managed to compile SatTrack 3.1.6-7 > for RedHat 6.1? My best effort so far, resulted > in a lot of warnings (Return type of main is not int.) > > The demo program works fine but sattrack shows: > > Elevation 0.000 deg > Azimuth 0.000 deg > Range 0.000 km etc > > for all satellites. What version of gcc do you have? libc6? etc. How did you get these above results (what did you run)? I can tell you what tools I used to compile the RPM. Otherwise I'm not sure what would give the above. By the way, it has Y2K problems. There's a patch in the sources I used to build the RPM. Maybe that's the problem you're having -- some pretty weird stuff happened until this was fixed. The SatTrack sources in there are hacked a bit with regards to directories. I'm not entirely happy with the changes; I've just thought of a better way to do it so I might post something here about it. Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: New AX.25 and newest kernel 2.2.14: first test results
On Tue, Jan 18, 2000 at 08:47:33AM +0100, Gerd wrote: > > I use the new stuff on a 486/66, 16mb ram, disless boot via nfs an suse 6.2. > > Kernel 2.2.14 with 16 MB RAM? Does that really work? Then one can guess that Certainly, my 2.2.12 packet box is a 486/100 with 16mb, also diskless (ie: no swap!) Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: About SINGLE MODE!
On Sat, Jan 15, 2000 at 07:42:33AM +, Richard Adams wrote: > If you are worried about someone rebooting the machine who has access to > the keyboard, then you as root could add the passwd option to /etc/lilo.conf > and rerun lilo, then only root can start the machine up with the defined Besides, there's a handful of ways besides 'linux single' to get single user mode or better. Like 'linux emergency', or 'linux init=/bin/bash' On Debian, single asks you for the root password before it lets you in. (emergency doesn't) Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
weird digipeating problem
Testing some APRS paths here and we've noted a weird digipeating problem. I don't think it's linux particular but I'm not 100% sure. I can talk directly to a standard standalone digipeater called VK3RPP-1. On the other side of town is VK3JFK, a digipeater with an alias of WIDE with callsign substiution enabled, and another digipeater, VK3JFK-1, with the same alias. If I send packets via: VK3RPP-1 VK3JFK VK3JFK-1 VK3RPP-1 it works (packet gets all the way), VK3RPP-1 VK3JFK WIDE VK3RPP-1 it works, VK3RPP-1 VK3JFK-1 WIDE VK3RPP-1 it works, but VK3RPP-1 WIDE VK3JFK-1 VK3RPP-1 doesn't work (doesn't return), nor VK3RPP-1 WIDE VK3JFK VK3RPP-1 doesn't work, nor VK3RPP-1 WIDE WIDE VK3RPP-1. So anywhere that I have WIDE second doesn't work. It appears that the digipeaters do key the radio though. We don't yet know if they actually modulate the signal. JFK and JFK-1 are both running PacComm 5.0 TNC firmware [I think]. Anyone seen anything like this before? A couple of other stations have used these WIDE repeaters and they work OK (and obviously they both do in some paths -- as above). Very strange. My local station here is running the linux AX.25 stack. (Kernel 2.2.12.) VK3RPP-1 is running X1J4 or whatever it's called. Local packets were generated with 'beacon'. thanks Hamish VK3SB -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: AX25 and Redhat 6.1
On Sat, Jan 08, 2000 at 11:43:53PM +, Richard Adams wrote: > Yes it is, for the simple reason, redhat-6.x uses kernels in the 2.2.xx > range. Rh 6.x means 6.0 and 6.1. > Its as easy as that. Don't you also need libc6.1 (glibc2.1) for the new tools? That probably means RH6.1, not 6.0. (But since I don't run RH, I'm guessing. On Debian, you need unstable(2.2) for both 2.2 kernel and 2.1 glibc). Hamish VK3SB -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: new sattrack packages
On Fri, Jan 07, 2000 at 02:43:24PM -0700, Brian Hall wrote: > Where do I download the packages again? ftp://ftp.rising.com.au/pub/hamish/sattrack-rpm/ Hamish -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
Re: new sattrack packages
On Thu, Jan 06, 2000 at 10:31:36PM -, Martyn Kinder G0CZD wrote: > > I will regenerate the RPMs to fix the fakeroot problem. > > Hi Hamish, - Has this been done yet? I am suffering the same problem (SuSE > 6.2). I have just uploaded the new files. Sorry about the delay. > BTW, I had some trouble downloading the file, it downloads as a text file, > not as a binary file.. I suspect this is a problem with your browser; it must not be recognising rpm files as binary. I suggest using a proper FTP program instead of using netscape or whatever. Hamish VK3SB -- Hamish Moffatt Mobile: +61 412 011 176 [EMAIL PROTECTED] Rising Software Australia Pty. Ltd.http://www.risingsoftware.com/ Phone: +61 3 9894 4788Fax: +61 3 9894 3362USA: 1 888 667 7839
