Keyboard Shortcuts
Likes
- Direwolf
- Messages
Search
Re: KISSPORT #### does not override 8001
I bumped into this when trying to run two separate instances of direwolf. One for VHF and one for HF. In that case I would see a collision on port 8001. After digging around a bit I found that the way to turn off the 8001 default is to set KISSPORT 0 before any other KISSPORT settings in the config.?
I also proposed a code change to the KISSPORT default behavior here. https://github.com/wb2osz/direwolf/pull/333 |
Re: KISSPORT #### does not override 8001
开云体育Hello Roger, Ok.. and btw, I've created a Github issue to track this issue for you: ?? --David KI6ZHD ?? On 07/07/2021 06:29 PM, Roger wrote:
Thanks, David |
Re: KISSPORT #### does not override 8001
In addition, I connected both TCP ports using APRSIS32 [0L] KB8RCO-2>APWW11,WIDE1-1,WIDE2-1:@173559h4209.35N/08346.30W1 EXPERIMENTATION [1L] KB8RCO-2>APWW11,WIDE1-1,WIDE2-1:@173559h4209.35N/08346.30W1 EXPERIMENTATION [0L] KB8RCO-2>APWW11,WIDE1-1,WIDE2-1:@173559h4209.35N/08346.30W1 EXPERIMENTATION I enabled 3 RF ports (8001, 8010, 8011) and opened log files for each. There is a delay between 8001 opening and 8010 and 8011 opening - which implies my theory is correct. Although it opens a TCP port and the port is there (available), APRSIS32 does not appear to do anything with 8001. When APRSIS32 beacons, there appears 2x the message for [0L] implying it sends on 8001 and then again on 8010. Robert Giuliano
On Thursday, July 8, 2021, 1:01:28 PM EDT, Rob Giuliano <kb8rco@...> wrote:
I just compiled V1.7 and tried to change the TCP KISS port. As stated, the 'default' port does not change, however when specifying a port, it looked like this:
So it appears you can assign the new port by radio channel. Both appeared to be active, so it appears the issue is with the generic port being set to the default and not re-assigned. So I went with ? KISSPORT 8010 0 ? KISSPORT 8011 1 And 3 KISS TCP ports were open:
So it looks to me like the default KISS TCP port is being opened before the config file is read with the information needed to open it, so a default port is always opened. Robert Giuliano
On Wednesday, July 7, 2021, 9:29:55 PM EDT, Roger <roger@...> wrote:
Thanks, David
I did revert to 1.6 stable version and all is well.? Just thought I'd give the new beta a spin- Appreciate your reply and confirmation. Roger, N1XP On 7/7/21 1:49 PM, David Ranch wrote:
|
Re: KISSPORT #### does not override 8001
I just compiled V1.7 and tried to change the TCP KISS port. As stated, the 'default' port does not change, however when specifying a port, it looked like this:
So it appears you can assign the new port by radio channel. Both appeared to be active, so it appears the issue is with the generic port being set to the default and not re-assigned. So I went with ? KISSPORT 8010 0 ? KISSPORT 8011 1 And 3 KISS TCP ports were open:
So it looks to me like the default KISS TCP port is being opened before the config file is read with the information needed to open it, so a default port is always opened. Robert Giuliano
On Wednesday, July 7, 2021, 9:29:55 PM EDT, Roger <roger@...> wrote:
Thanks, David
I did revert to 1.6 stable version and all is well.? Just thought I'd give the new beta a spin- Appreciate your reply and confirmation. Roger, N1XP On 7/7/21 1:49 PM, David Ranch wrote:
|
Re: KISSPORT #### does not override 8001
开云体育Thanks, DavidI did revert to 1.6 stable version and all is well.? Just thought I'd give the new beta a spin- Appreciate your reply and confirmation. Roger, N1XP On 7/7/21 1:49 PM, David Ranch wrote:
|
TNC exporter: metrics collection and visualization for direwolf
Ross Young
Hi direwolf community,
I'd like to introduce my open source project, , which provides a legible, information-dense and customizable interface to view and analyze TNC activity over time. It is a prometheus exporter and grafana dashboard for direwolf (and other software TNCs) that collects packet metrics using the AGW or KISS TCP/IP interfaces. Please see the project's README file for additional information about requirements, installation, configuration etc. TNC exporter is released under the MIT license. I built this in my spare time over the past few months, mostly because it was an interesting challenge dealing with several technologies I wanted to learn more about. Despite this, I have tried to make it a fully featured, decently documented piece of software that is of use to packet radio operators, particularly folks who would like to track ongoing usage of infrastructure such as digipeaters and iGates. I still have some additional features planned (namely Mic-E and compressed position report parsing) but at this point I think it's quite useful and want to get the word out. I'm looking forward to hearing if folks here find it a good addition to their shack, and I'm happy to provide support or answer questions. Bug reports, suggestions and code contributions are welcome, through the github repo linked above or directly via email if you prefer. Ross KJ7GES |
option digipeater no work in mi direwolf
Hello, I compiled on a raspberry pi zero W direwolf 1.6 as indicated in the manual, my intention is to configure an aprs repeter (digipeater).
The problem is that when I turn on a handheld radio with aprs and it sends its beacom, direwolf if it hears it but does not broadcast the information, when I see the linux shell it says that it does but in reality it does not send it to the radio, it even see on the radio that it is connected to the raspberry that changes the display light because the ptt is activated, but it is as if it does not send the audio, only repeater, because when it sends the audio of the direwolf beacom it does it well. If I activate the igate direwolf option, it registers the data from the handheld radio on the internet without problem and I see it in aprs.fi. This in my understanding is that if it is receiving the packets from the handheld radio well. xe1gnu-1 is direwolf and xe1gnu-7 is other radio I compiled it wrong ?. Do I get a package on the raspberry? Any configuration problems? I have 7 days on that and I have no idea where you could see the problem. Thank you and 73 from xe1gnu is an important project for my SOTA activity. this is my configuration. ADEVICE? plughw:2,0 CHANNEL 0 MYCALL XE1GNU-1 MODEM 1200 PTT GPIO 25 AGWPORT 8000 KISSPORT 8001 PBEACON delay=1 ?every=5 overlay=S symbol="digi" lat=19^28.025N long=099^6.383438W power=25 height=20 gain=4 comment="Iniciando" via=WIDE1-1,WIDE2-1 DIGIPEAT 0 0 ^WIDE[3-7]-[1-7]$|^TEST$ ^WIDE[12]-[12]$ TRACE TTPOINT ?B01 ?37^55.37N ?81^7.86W TTPOINT ?B7495088 ?42.605237 ?-71.34456 TTPOINT ?B934 ?42.605237 ?-71.34456 TTPOINT B901 ?42.661279 ?-71.364452 TTPOINT B902 ?42.660411 ?-71.364419 TTPOINT B903 ?42.659046 ?-71.364452 TTPOINT B904 ?42.657578 ?-71.364602 TTVECTOR ?B5bbbddd ?37^55.37N ?81^7.86W ?0.01 ?mi TTGRID ??Byyyxxx ???37^50.00N ?81^00.00W ?37^59.99N ?81^09.99W TTUTM ?B6xxxyyy ?19T ?10 ?300000 ?4720000 TTCORRAL ??37^55.50N ?81^7.00W ?0^0.02N TTMACRO ?xx1yy ?B9xx*AB166*AA2B4C5B3B0A1yy TTMACRO ?xx2yy ?B9xx*AB170*AA3C4C7C3B0A2yy TTMACRO ?xxyyy ?B9xx*AB180*AA3A6C4A0Ayyy TTMACRO ?z ?Cz this is the sell direwolf -t 2 Dire Wolf version 1.6 Includes optional support for: ?cm108-ptt Reading config file direwolf.conf Audio device for both receive and transmit: plughw:2,0 ?(channel 0) Channel 0: 1200 baud, AFSK 1200 & 2200 Hz, E+, 44100 sample rate / 3. Ready to accept AGW client application 0 on port 8000 ... Ready to accept KISS TCP client application 0 on port 8001 ... [0L] XE1GNU-1>APDW16,WIDE1-1,WIDE2-1:!1928.03NS09906.38W#PHG5140Iniciando XE1GNU-7 audio level = 105(55/32) ??[NONE] ??__||||||_ [0.4] XE1GNU-7>Q9RX0S,WIDE1-1,WIDE2-1:`<0x7f>^C<0x1c><0x1f>SP\`"L6}HG-UV98 2736.8Km 9.1V ?31.5C ?782.1hPa S05 MIC-E, Parking, Mic-Emsg, In Service N 19 28.0300, W 099 06.3900, 0 MPH, course 355, alt 7267 ft HG-UV98 2736.8Km 9.1V ?31.5C ?782.1hPa S05 [0H] XE1GNU-7>Q9RX0S,XE1GNU-1*,WIDE2-1:`<0x7f>^C<0x1c><0x1f>SP\`"L6}HG-UV98 2736.8Km 9.1V ?31.5C ?782.1hPa S05 |
GPIO permission denied, udev racing
This bug was supposedly fixed in direwolf 1.6 by adding 250ms before gpio initialization.??
I'm still observing it on a raspberry pi zero.? Does anyone have a workaround or defensive code/patch to make direwolf more deterministic?? ?I just upped the wait to 500ms to see if that helps, in ptt.c. this is rare, but it still happens,
? |
Re: KISSPORT #### does not override 8001
开云体育Hello Roger, I just tested it and I see what you're seeing.? Ultimately, this seems to be a bug with v1.7 so unless you specifically need the BETA v1.7 release, please use the released v1.6 version. -- $ sudo direwolf -t 0 -d o -a 100 -c /etc/ax25/direwolf.conf Dire Wolf DEVELOPMENT version 1.7 A (Jul? 7 2021) Includes optional support for:? gpsd cm108-ptt Reading config file /etc/ax25/direwolf.conf Audio device for both receive and transmit: plughw:1,0? (channel 0) Channel 0: 1200 baud, AFSK 1200 & 2200 Hz, E+, 44100 sample rate / 3. Ready to accept AGW client application 0 on port 8003 ... Ready to accept KISS TCP client application 0 on port 8001 ...? <--------------------------------- Ready to accept KISS TCP client application 0 on port 8004 ... -- Here is a confirmation of the open ports: -- $ sudo lsof -nPi | grep direwolf direwolf? 792?? root??? 3u? IPv4? 18033????? 0t0? TCP *:8003 (LISTEN) direwolf? 792?? root??? 7u? IPv4? 20010????? 0t0? TCP *:8001 (LISTEN)? <---------------- direwolf? 792?? root??? 9u? IPv4? 18035????? 0t0? TCP *:8004 (LISTEN)? -- This seems like this is a bug introduced in Direwolf v1.7 as I don't see this behavior with Direwolf v1.6: -- $ sudo direwolf -t 0 -d o -a 100 -c /etc/ax25/direwolf.conf Dire Wolf version 1.6 Includes optional support for:? gpsd cm108-ptt Reading config file /etc/ax25/direwolf.conf Audio device for both receive and transmit: plughw:1,0? (channel 0) Channel 0: 1200 baud, AFSK 1200 & 2200 Hz, E+, 44100 sample rate / 3. Ready to accept AGW client application 0 on port 8003 ... Ready to accept KISS TCP client application 0 on port 8004 ...? <-------------------- -- --David KI6ZHD On 07/05/2021 06:28 AM, Roger wrote:
|
Re: Using GPIO pins on other SBCs
The numbering system is typically done through libraries. I would assumevthat since you are usjng something other than a Pi, you are compiling Direwolf on that device.? This could/should have a cross? reference. This site: implies you can use the wiringpi library is used for such things. I would give it a try with the normal designation and use a DMV to check that the proper pin is doing what you expect. Personally, I'd go with a Pi.
|
Re: Using GPIO pins on other SBCs
Thanks Robert,
toggle quoted message
Show quoted text
The information you quote is specific to the RPI and I want to use the "GPIO" config command on a Odroid-C4 (another SBC; see and ). The GPIO 17 command refers to the numbering used on the RPI. i.e GPIO 17 is pin 11 on the 40 pin header. The Odroid-C4 refers to its I/O pins as GPIOX.x (where x is a number). As an example, pin 11 on the C4's 40-pin header is referred to as GPIOX.3 What I need to understand is how direwolf maps the GPIO config commands to the hardware. There has to be an OS intermediary that does this; I just need to find out what that is and adjust it to suit the C4. Regards, Ian On 6/7/21 10:43 pm, Rob Giuliano via groups.io wrote:
Found more information in the manual at: |
Re: Using GPIO pins on other SBCs
Ian, I use a GPIO pin to disable the transmitter. My Igates are at VHF contest sites, and the APRS frequency is close enough to the frequencies used for contesting to interfere with contesting. So I needed a clean way to disable the transmitter.
A webpage I found that discribed simple GPIO control is below. I then wrote bash shell scripts to control a GPIO line. Those scripts are then called through an "at" command to flip the pin to off before and then back on after the contest. The script to turn it on is also run at boot as it powers up disabled. My transmit key circuit is a combination of a pnp and npn transistors so it can be inhibited.
Direwolf has no idea the transmitter is off. I also use GPIO to key the transmitter, as others have already detailed.
Reference page:
Jim Bacher, WB8VSU
wb8vsu@...
On Jul 6, 2021, at 4:16 AM, "Ian (Benny) Bennett" <ibennett@...> wrote: Evening all, |
Re: Using GPIO pins on other SBCs
Found more information in the manual at: Section 9.2.12 (Page 78 & 79) ??? TXINH GPIO 17 ? TXINH GPIO -17 As with PTT, minus in front of the GPIO number means invert the signal. Robert Giuliano
On Tuesday, July 6, 2021, 8:16:45 AM EDT, Rob Giuliano <kb8rco@...> wrote:
If you look through the direwolf.conf file (example file) you will find examples of how to assign functions to pins. In each case, there is an example for a serial port and one for GPIO. Examples: ??? #PTT GPIO 25 ??? #DCD GPIO 24 The number is the GPIO designation (not a pin number)I can't say anything about TX INHIBIT. I've only seen it mentioned in the discussions Robert Giuliano
On Tuesday, July 6, 2021, 4:16:33 AM EDT, Ian (Benny) Bennett <ibennett@...> wrote:
Evening all, ??? I've been reading through the direwolf RPI doco, specifically interested in the use of the GPIO pins for PTT, TX INHIBIT and DCD. ??? I'm planning on using a different SBC (Odroid-C4) which also has exposed GPIO pins (via a 40 pin header) and runs (Ubuntu) Linux. ??? I'm keen to understand how direwolf identifies the RPI GPIO pins so I can (hopefully) adjust the numbering to suit the C4. ??? I did find a "wiringpi" library for the C4 but I can't find any reference to that in the direwolf documentation. ??? I'd give it a try but the C4's are "out of stock" until mid-August, so I'm trying to get a head start by asking here. ??? Thanks in advance. Ian VK1IAN |
Re: Using GPIO pins on other SBCs
If you look through the direwolf.conf file (example file) you will find examples of how to assign functions to pins. In each case, there is an example for a serial port and one for GPIO. Examples: ??? #PTT GPIO 25 ??? #DCD GPIO 24 The number is the GPIO designation (not a pin number)I can't say anything about TX INHIBIT. I've only seen it mentioned in the discussions Robert Giuliano
On Tuesday, July 6, 2021, 4:16:33 AM EDT, Ian (Benny) Bennett <ibennett@...> wrote:
Evening all, ??? I've been reading through the direwolf RPI doco, specifically interested in the use of the GPIO pins for PTT, TX INHIBIT and DCD. ??? I'm planning on using a different SBC (Odroid-C4) which also has exposed GPIO pins (via a 40 pin header) and runs (Ubuntu) Linux. ??? I'm keen to understand how direwolf identifies the RPI GPIO pins so I can (hopefully) adjust the numbering to suit the C4. ??? I did find a "wiringpi" library for the C4 but I can't find any reference to that in the direwolf documentation. ??? I'd give it a try but the C4's are "out of stock" until mid-August, so I'm trying to get a head start by asking here. ??? Thanks in advance. Ian VK1IAN |
Using GPIO pins on other SBCs
Evening all,
I've been reading through the direwolf RPI doco, specifically interested in the use of the GPIO pins for PTT, TX INHIBIT and DCD. I'm planning on using a different SBC (Odroid-C4) which also has exposed GPIO pins (via a 40 pin header) and runs (Ubuntu) Linux. I'm keen to understand how direwolf identifies the RPI GPIO pins so I can (hopefully) adjust the numbering to suit the C4. I did find a "wiringpi" library for the C4 but I can't find any reference to that in the direwolf documentation. I'd give it a try but the C4's are "out of stock" until mid-August, so I'm trying to get a head start by asking here. Thanks in advance. Ian VK1IAN |
Re: USB audio problems; need recovery help
开云体育Sleeve, Shield, regardless.? The way the mechanical attachment of a shielded (sleeved, if you will) cable to the connector nearly demands (and certainly benefits from) the shield being placed in the strain relief crimp at the tail of the connector, with the wires internal to the cable's shield routed to the T and R's.? Swapping the roles of the lower R and the S, retaining S=Ground, would accomplish the same shorting of mike to ground, since the TRRS and TRS (and even TS) plugs are all the same length.I understand that it's the "standard" for many devices, but still believe it was done wrong.? According to at least some "MP3 Players" (remember them?) got it right.? I have my suspicions on why their "standard" didn't stick... Greg? KO6TH Rob Giuliano via groups.io wrote:
|
Re: USB audio problems; need recovery help
TRRS is not Tip-Ring-Ring-SHIELD, but Tip-Ring-Ring-SLEEVE. On most devices, the Sleeve is the mic input.? This allows for it to be compatible between headsets with a mic and without a mic.? Those without a mic will typically short the Sleeve (mic) to ground and indicate to the device that no mic is present. This is true in most current laptops that have switched to the TRRS connector, as well as most tablets and cell phones. Robert Giuliano
On Monday, July 5, 2021, 5:28:03 PM EDT, Greg D <ko6th.greg@...> wrote:
I hope so, too.? It may be a while; the problem appears to have
"gone away" since rebuilding the interface cable, but I've saved the
instructions in a text file on the Pi's desktop in case it "comes
back" at a most inopportune moment. Thanks for the help! Greg? KO6TH p.s.? Note to the folks coming up with connector pinouts:? When you see "Shield" in a TRRS connector's description, please connect it to the Shield, not the Mike-In! Michael - NA7Q wrote:
Just restore. You don't even need to reboot unless something is messed up with the system. |
Re: USB audio problems; need recovery help
开云体育I hope so, too.? It may be a while; the problem appears to have "gone away" since rebuilding the interface cable, but I've saved the instructions in a text file on the Pi's desktop in case it "comes back" at a most inopportune moment.Thanks for the help! Greg? KO6TH p.s.? Note to the folks coming up with connector pinouts:? When you see "Shield" in a TRRS connector's description, please connect it to the Shield, not the Mike-In! Michael - NA7Q wrote:
Just restore. You don't even need to reboot unless something is messed up with the system. |
KISSPORT #### does not override 8001
New install of Direwolf dev1.7 does not allow changing KISSPORT? FROM 8001. |