Wanted to take? moment to report something that just started to appear when starting Direwolf. I have not previously seen these messages regardong the CM108 PTT dll and thought it best to report here.

Thinking I'd mangled the config file I did a complete reinstall and overwrote the original and am experienceing the same message.

An additional step was reinstalling Direwolf off the root of C and setting up a new config file.

Same results.

FWIW the PTT is being asserted, the radio transmits.

-Chris

开云体育

Hello John,

just another possibility to consider. eas2aprs.py specifies channel 0 as the output for the aprs packet constructed from reading the eas input created by the sdr.

Stepping back a moment, I re-read your original email and I think I was confused.? Are you wanting to do what this following this doc proposes or are you trying to do something different?

??


I ask because your original email specifically wanted to use APRX and I had assumed you wanted it to connect to the remote Direwolf KISS-TCP interface.? That won't be enough to convert the EAS message into an APRS packet.? The above URL talks about the other needed glue including kiss-util (comes with Direwolf), eas2aprs.py (different Git project), and dsame.py (different Git project).? The secret sauce in this one document's approach is the kiss-util command which silently DEFAULTS to connecting to a KISS-TCP port on *locahost* of the current machine but also creates a software "shared memory" connector which the eas2aprs.py + dsame programs do all the protocol translation.

So.. understanding all that... why are you wanting to use APRX to support your project goal?? Best I can tell, Direwolf should be able to do all of this w/o APRX but I have a concern with your design.? You specifically want to receive KISS-TCP packets from Computer-A but you want to *transmit* them from Computer-B.? I don't think that the kissutil program can support that split RX vs. TX KISS-TCP role.? Your original email stated:

?? "For the time being, I wish to maintain the 2 computer setup; EAS reception and processing on computer A, with aprs digipeating and internet gating on computer B. There are a number of reasons for the 2 computer setup; unless I am advised that this is not possible, I would to continue with that plan"


I believe there are other KISS-TCP "mux" programs out there that could be considered to see if you could split the KISS packet traffic but I think it will greatly complicate the setup.

Does the TCP port receive output when DW has no output for a receive-only device?

Yes.. direwolf will always allow remote KISS-TCP connections as long as it's configured to do so.? The actual traffic that comes/goes through that port is secondary.?

Is there a way in direwolf config to construct a second device that receives input from eas2aprs.py and produces output on the TCP port?

eas2aprs.py interacts with direwolf only via the kissutil program.?

--David
KI6ZHD

开云体育

Hello John,

after aprx restart on Computer B (*.153):

netstat -a |grep 8001
tcp ? ? ? ?0 ? ? ?0 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ?

This doesn't make sense.? Why is port 8001 listening?? Try the following:

?? - kill the aprx process so it's not running

stopped aprx with systemctl.

Ok

?

?? - run "lsof -nPi" and see what is listening on port 8001

?

nothing listed but I remembered that DW on computer B opens a KISSPORT on 8001 in the config file. I edited /etc/dorewolf.conf on computer B to remove the KISSPORT command, and restarted DW and aprx on computer B after restarting everything (DW+kissutils+eas2apys.py) on computer A.

Why is Direwolf running on Computer B?? You made no mention of this before.

now on computer A:

?netstat -a |grep 8001
tcp ? ? ? ?0 ? ? ?0 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ?
tcp ? ? ? ?0 ? ? ?0 localhost:58832 ? ? ? ? localhost:8001 ? ? ? ? ?ESTABLISHED
tcp ? ? ? ?0 ? ? ?0 localhost:8001 ? ? ? ? ?localhost:58832 ? ? ? ? ESTABLISHED

Still not good as there isn't any connection from computer B

over on computer B I get:

netstat -a |grep 8001
tcp ? ? ? ?0 ? ? ?0 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ?
unix ?3 ? ? ?[ ] ? ? ? ? STREAM ? ? CONNECTED ? ? 18001?

(what's with the 18001 port?)

I assume you have a good reason for Computer B to run Direwolf and we can ignore it.? The important thing here is that there aren't any connections from APRX going OUTBOUND to Computer A.

Don't worry about the "unix" socket.. that's internal Linux stuff.

Next, test that the connection can be made from Computer B to Computer A, run:

?? telnet 192.168.1.153 8001

RPi do not come with telnet so after installing:

telnet 192.168.1.153 8001
Trying 192.168.1.153...
Connected to 192.168.1.153.
Escape character is '^]'.

it connects, apparently.

You should get a connection but no real specific response.? If it doesn't connect,? you probably have a firewall enabled on Computer A and port 8001 needs to be opened up.

Ok.. that's good.? It seems that something is wrong with APRX.? It's either not starting or not initiating a connection to Computer A.? I recommend you manually run it from the command prompt (don't use systemd) and see if it's throwing any errors.? If that doesn't help, enable additional logging and see what's happening.

--David
KI6ZHD

开云体育

Hello John,

I edited aprx.conf so the relevant section looks like:

<interface>
? ?tcp-device 192.168.1.153 8001 KISS
? ?tx-ok true
</interface>

Ok.

I restarted direwolf + kissutil+eas2aprs.py on computer A (192.168.1.153).
I then restarted aprx after the edit above on computer B (192.168.1.16).
I checked that the aprx computer system B was running both by RF and aprs-is ....OK

After you restarted APRX on Computer B,

I sent the test message from computer A.
Then did netstat. Here are the before and after results:
before executing test message:
after direwolf + kissutil+eas2aprs.py restart on computer A(*.16):
netstat -a |grep 8001
tcp ? ? ? ?0 ? ? ?0 0.0.0.0:8001 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ??
tcp ? ? ? ?0 ? ? ?0 localhost:8001 ? ? ? ? ?localhost:44302 ? ? ? ? ESTABLISHED
tcp ? ? ? ?0 ? ? ?0 localhost:44302 ? ? ? ? localhost:8001 ? ? ? ? ?ESTABLISHED

after aprx restart on Computer B (*.153):
netstat -a |grep 8001
tcp ? ? ? ?0 ? ? ?0 0.0.0.0:8001 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ?

This doesn't make sense.? Why is port 8001 listening?? Try the following:

?? - kill the aprx process so it's not running
?? - run "netstat -a | grep 8001" and see if nothing is listed.? If something IS listed
?? - run "lsof -nPi" and see what is listening on port 8001


Next, test that the connection can be made from Computer B to Computer A, run:

?? telnet 192.168.1.153 8001

You should get a connection but no real specific response.? If it doesn't connect,? you probably have a firewall enabled on Computer A and port 8001 needs to be opened up.

--David

Hi David,
please see below....
I assume you are running this on "computer A". ?Correct?

YES
I edited aprx.conf so the relevant section looks like:

<interface>
? ?tcp-device 192.168.1.153 8001 KISS
? ?tx-ok true
</interface>
I restarted direwolf + kissutil+eas2aprs.py on computer A (192.168.1.153).
I then restarted aprx after the edit above on computer B (192.168.1.16).
I checked that the aprx computer system B was running both by RF and aprs-is ....OK
I sent the test message from computer A.
Then did netstat. Here are the before and after results:
before executing test message:
after direwolf + kissutil+eas2aprs.py restart on computer A(*.16):
netstat -a |grep 8001
tcp ? ? ? ?0 ? ? ?0 0.0.0.0:8001 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ??
tcp ? ? ? ?0 ? ? ?0 localhost:8001 ? ? ? ? ?localhost:44302 ? ? ? ? ESTABLISHED
tcp ? ? ? ?0 ? ? ?0 localhost:44302 ? ? ? ? localhost:8001 ? ? ? ? ?ESTABLISHED
after aprx restart on Computer B (*.153):
netstat -a |grep 8001
tcp ? ? ? ?0 ? ? ?0 0.0.0.0:8001 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ??
after executing test message:
netstat -a |grep 8001 (on Computer A)
tcp ? ? ? ?0 ? ? ?0 0.0.0.0:8001 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ??
tcp ? ? ? ?0 ? ? ?0 localhost:8001 ? ? ? ? ?localhost:44302 ? ? ? ? ESTABLISHED
tcp ? ? ? ?0 ? ? ?0 localhost:44302 ? ? ? ? localhost:8001 ? ? ? ? ?ESTABLISHED
netstat -a |grep 8001 on Computer B (*.153):
tcp ? ? ? ?0 ? ? ?0 0.0.0.0:8001 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN

I don't see the connection on computer B (*.16).
Note no change on computer B (the aprx machine). Also no easmessages sent by either RF nor internet.
John

开云体育

Hello John,

On computer A, with IP 192.168.1.153, the current direwolf configuration file (eas.conf) looks like:

Ok

with kissutil and eas2aprs.py running. and sending the test message, I get this output:

I assume you are running this on "computer A".? Correct?

Attached to KISS TCP client application 0 on port 8001 (radio channel 0) ...

Ready to accept KISS TCP client application 1 on port 8001 (radio channel 0) ...
[0L] NX2I>APZEAS::NWS ? ? ?:[1/4] The National Weather Service in Kansas City/Pleasant Hill, MO has issued a
[0L] NX2I>APZEAS::NWS ? ? ?:[2/4] Required Weekly Test valid until 01:30 PM for the following counties in
[0L] NX2I>APZEAS::NWS ? ? ?:[3/4] Kansas: Johnson, Leavenworth, Miami, Wyandotte, and for the following counties
[0L] NX2I>APZEAS::NWS ? ? ?:[4/4] in Missouri: Cass, Clay, Jackson, Platte. (KEAX/NWS)

This all looks fine where the top line is showing your kissutil running on Computer A, connecting to Computer A

furthermore on computer A,? I see the TCP port running:
netstat -a |grep 8001
tcp ? ? ? ?0 ? ? ?0 0.0.0.0:8001 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ??
tcp ? ? ? ?0 ? ? ?0 localhost:49548 ? ? ? ? localhost:8001 ? ? ? ? ?ESTABLISHED
tcp ? ? ? ?0 ? ? ?0 localhost:8001 ? ? ? ? ?localhost:49548 ? ? ? ? ESTABLISHED

Yup.. everything is working locally.

All this looks OK to me. However, no signal is transmitted by Computer B/aprx digipeater-Igate. I hear no signal with another radio listening on 144.390 MHz nor is anything reported at aprs.fi.

If my assumptions are correct above, there is no connection to Computer B.?

The addition I have made to the original (fully working) config file for aprx looks like:
<interface>
? ?tcp-device 192.168.1.16 8001 KISS
? ?tx-ok true
</interface>

Per Section 3.6.10 from , the "tcp-device" line is for the REMOTE KISS-TCP IP address.? As such, you should be using:

?? tcp-device 192.168.1.153 8001 KISS


Once you reconfigure and restart APRX, run the "netstat -a | grep 8001" command on both Computer-A and Computer-B.? You should see a KISS-TCP connection to each other on both sides.


--David
KI6ZHD

I apologize for this long post in advance. I am interested in revisiting the original issue described in the beginning post of this thread. An EAS sdr receiver has been operational at my station now since 9/2022. I believe receiving NWS severe weather bulletins is valuable, especially during hurricane season ( a 6 month affair here in central Florida). The setup here has included 2 separate radio systems:
1) an rtl sdr running on a pi4 interfaced to a handheld radio, and
2) a 2 meter mobile interfaced to a different pi3 Igate/digi running aprx.

When an EAS alert is received by the sdr,? direwolf transmits the warning bulletin as an aprs packet over the handlheld. The signal is received at NX2I-11 digi and send both over the radio and aprs-is by aprx. This has worked well for almost 2 years.

A few weeks ago, the handheld died, so I have been looking again into sending the aprs packet directly to aprx without the intermediate step via the radio. My (failed) efforts so far have focused on using tcp to send the aprs (EAS) packet to aprx for igate/digipeater operation. For the time being, I wish to maintain the 2 computer setup; EAS reception and processing on computer A, with aprs digipeating and internet gating on computer B. There are a number of reasons for the 2 computer setup; unless I am advised that this is not possible, I would to continue with that plan.

On computer A, with IP 192.168.1.153, the current direwolf configuration file (eas.conf) looks like:

CHANNEL 0
MYCALL NX2I-6
KISSPORT 8001 0
#PTT GP

#DEVICE sdr
ADEVICE ?stdin ?null
ARATE 24000
#CHANNEL 2
MODEM EAS /1

The command, rtl_fm -f 162.500M |direwolf -c /home/pi/eas.conf -a 30, works and produces the output:

Dire Wolf version 1.7
Includes optional support for: ?cm108-ptt

Reading config file /home/pi/eas.conf
Audio input device for receive: stdin ?(channel 0)
Audio out device for transmit: null ?(channel 0)
Found 1 device(s):
Channel 0: 521 baud, AFSK 2083 & 1563 Hz, A+, 24000 sample rate.
Note: PTT not configured for channel 0. (Ignore this if using VOX.)
Ready to accept AGW client application 0 on port 8000 ...
Ready to accept KISS TCP client application 0 on port 8001 (radio channel 0) ...
? 0: ?Realtek, RTL2838UHIDIR, SN: 00000001

Using device 0: Generic RTL2832U OEM
Detached kernel driver
Found Rafael Micro R820T tuner
Tuner gain set to automatic.
Tuned to 162752000 Hz.
Oversampling input by: 42x.
Oversampling output by: 1x.
Buffer size: 8.13ms
Exact sample rate is: 1008000.009613 Hz
Sampling at 1008000 S/s.
Output at 24000 Hz.

ADEVICE0: Sample rate approx. 24.0 k, 0 errors, receive audio level CH0 39

with kissutil and eas2aprs.py running. and sending the test message, I get this output:



Attached to KISS TCP client application 0 on port 8001 (radio channel 0) ...

Ready to accept KISS TCP client application 1 on port 8001 (radio channel 0) ...
[0L] NX2I>APZEAS::NWS ? ? ?:[1/4] The National Weather Service in Kansas City/Pleasant Hill, MO has issued a
[0L] NX2I>APZEAS::NWS ? ? ?:[2/4] Required Weekly Test valid until 01:30 PM for the following counties in
[0L] NX2I>APZEAS::NWS ? ? ?:[3/4] Kansas: Johnson, Leavenworth, Miami, Wyandotte, and for the following counties
[0L] NX2I>APZEAS::NWS ? ? ?:[4/4] in Missouri: Cass, Clay, Jackson, Platte. (KEAX/NWS)

furthermore on computer A,? I see the TCP port running:
netstat -a |grep 8001
tcp ? ? ? ?0 ? ? ?0 0.0.0.0:8001 ? ? ? ? ? ?0.0.0.0:* ? ? ? ? ? ? ? LISTEN ? ??
tcp ? ? ? ?0 ? ? ?0 localhost:49548 ? ? ? ? localhost:8001 ? ? ? ? ?ESTABLISHED
tcp ? ? ? ?0 ? ? ?0 localhost:8001 ? ? ? ? ?localhost:49548 ? ? ? ? ESTABLISHED

All this looks OK to me. However, no signal is transmitted by Computer B/aprx digipeater-Igate. I hear no signal with another radio listening on 144.390 MHz nor is anything reported at aprs.fi.

The addition I have made to the original (fully working) config file for aprx looks like:
<interface>
? ?tcp-device 192.168.1.16 8001 KISS
? ?tx-ok true
</interface>
Presumably this is to interface to the packet stream coming from the sdr system,? running on 193.168.1.153.
I am? not a networking expert...
I'd appreciate any comments/help that will get this running.
73, John, NX2I

?I bought another rlt-sdr device? named "RTL2832U+RB820T2 USB"? ?Use the same start command , it worked well. this is the running log with it

root@raspberrypi:~# rtl_fm -f 144.640M - | direwolf -c /root/direwolf.conf -r 24000 -D 1 -
Dire Wolf version 1.7
Includes optional support for: ?cm108-ptt

Dire Wolf requires only privileges available to ordinary users.
Running this as root is an unnecessary security risk.

Reading config file /root/direwolf.conf
Audio input device for receive: stdin ?(channel 0)
Audio out device for transmit: null ?(channel 0)
Found 1 device(s):
Channel 0: 1200 baud, AFSK 1200 & 2200 Hz, A+, 24000 sample rate.
Note: PTT not configured for channel 0. (Ignore this if using VOX.)
Ready to accept AGW client application 0 on port 8000 ...
? 0: ?Realtek, RTL2838UHIDIR, SN: 00000001

Ready to accept KISS TCP client application 0 on port 8001 ...
Using device 0: Generic RTL2832U OEM
Detached kernel driver
Found Fitipower FC0012 tuner
Tuner gain set to automatic.
Tuned to 144892000 Hz.
Oversampling input by: 42x.
Oversampling output by: 1x.
Buffer size: 8.13ms
Exact sample rate is: 1008000.009613 Hz
Sampling at 1008000 S/s.
Output at 24000 Hz.
Allocating 15 zero-copy buffers

Now connected to IGate server asia.aprs2.net (150.95.137.86)
Check server status here http://150.95.137.86:14501

[ig] # aprsc 2.1.17-ga839328
[ig] # logresp BH6AJP-15 verified, server T2FUKUOKA
[ig] BH6AJP-15>APDW17:!3254.26NR11545.49E&144.640MHz Testing aprs gate raspberry wit rtl-sdr. Why can not recive anything?

BH6AJP-10 audio level = 42(8/8) ? ?|||||____
[0.2] BH6AJP-10>APX1C5,WIDE1-1:!3254.26N/11544.87Er欢迎来到阜阳,439.650/-9/T88.5 Aprsgate 144.640。 4.1V L


It can recive the aprs radio , i found the log do not have " [R82XX] PLL not locked!" keyword . i have already searched the keywork , someone said? " the rtl-lib is too old"? ?and " the bug with the driver"?

So? how to upgrade the lib or the driver ??

Progress! I got the 1.8 DEV version built and running. I see the GPIO25 line go high when re-transmitting a packet.

To recap: I did not downgrade the Pi firmware, I did not revert to an older version of Raspian. All I needed to do is follow the directions Gordon W4ZD pointed to in the Direwolf user manual to build the DEV version of Direwolf. Don't forget the step "git checkout dev" -- that caused one false start when I built the 1.7 version. Another false start: I forgot to include the gpiod support. What worked for me was "Dire Wolf DEVELOPMENT version 1.8 A"

I am running Bookworm with direwolf 1.8A on a raspberry pi 3b and my config file has ?PTT GPIOD gpiochip0 25?

73s,

Jay KI7JS

Sorry, I just re-read you post and you have the '-' at the end of your direwolf command. I have no other ideas that might help

Here is the startup command I use with my rtl device. It has been working for several years on direwolf 1.3. I think you might be missing '-' at the end of your direwolf command.

rtl_fm -f 144.390M - | direwolf -c sdr.conf -l . -r 24000 -D 1 -

Good luck, Jay KI7JS

?
re: How is the git command to get the DEV release?


At this point you should have the most recent stable version.In some cases, you might want the latest (sometimes unstable) development version to test a bug fix orget a preview of a new (possibly incomplete) feature that will be in the next release. In that case, also type:
git checkout dev

Hi David,

Yes. I agree with you.

I've tried (not with DEV release) putting in cfg the gpiod line, and it gives the error about RTS that I've read here.

How is the git command to get the DEV release?

Many thanks!

73's Rafa EB2DJB

RTL_FM is outputing outputing to standard output and Direwolf is listening on TCP/UDP channel.

Hello
? ??
? ? ?I set up Direwolf on a raspberry pi 4b? + rtl-sdr v4 an an Aprs only rx IGate. I can find the igate on aprs.fi , But? when i tried? send aprs radio with my aprs device,it seemed that my Aprs gate cant recive the radio. I have tried? some? rtl+direwolf img of raspberry and? include manual installation configuration. They have the same result. cant not recive aprs radio.I am sure that my aprs device already send the aprs radio ,because? the other gateway recived the aprs radio.

This is my running log?

root@raspberrypi:~# rtl_fm -f 144.64M ?- | direwolf -c /root/direwolf.conf -r 24000 -D 1 -
Dire Wolf version 1.7
Includes optional support for: ?cm108-ptt

Dire Wolf requires only privileges available to ordinary users.
Running this as root is an unnecessary security risk.

Reading config file /root/direwolf.conf
Audio input device for receive: stdin ?(channel 0)
Audio out device for transmit: null ?(channel 0)
Found 1 device(s):
Channel 0: 1200 baud, AFSK 1200 & 2200 Hz, A+, 24000 sample rate.
Note: PTT not configured for channel 0. (Ignore this if using VOX.)
Ready to accept AGW client application 0 on port 8000 ...
Ready to accept KISS TCP client application 0 on port 8001 ...
? 0: ?RTLSDRBlog, Blog V4, SN: 00000001

Using device 0: Generic RTL2832U OEM
Detached kernel driver
Found Rafael Micro R828D tuner
Tuner gain set to automatic.
[R82XX] PLL not locked!
Tuned to 144892000 Hz.
Oversampling input by: 42x.
Oversampling output by: 1x.
Buffer size: 8.13ms
Exact sample rate is: 1008000.009613 Hz
Allocating 15 zero-copy buffers
[R82XX] PLL not locked!
Sampling at 1008000 S/s.
Output at 24000 Hz.

Now connected to IGate server asia.aprs2.net (150.95.137.86)
Check server status here http://150.95.137.86:14501

[ig] # aprsc 2.1.17-ga839328
[ig] # logresp BH6AJP-15 verified, server T2FUKUOKA
[ig] BH6AJP-15>APDW17:!3254.26NR11545.49E&144.640MHz Testing aprs gate raspberry wit rtl-sdr. Why can not recive anything?

I didn't know is something wrong with the "[R82XX] PLL not locked!"?

This is my direwolf conf?


? GNU nano 5.4 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?/root/direwolf.conf
# 输入设备：监听UDP的7355端口，这样比较省资源且不会出现报错情况。
# 输出设备：因为APRS网关只接收信标，不发射，所以没有输出设备，null即是空。
ADEVICE udp:7355 null
# 呼号及SSID，10代表互联网网关设备。
MYCALL BH6AJP-15
# APRS-IS服务器使用国内的，就近原则。
IGSERVER asia.aprs2.net
# APRS-IS认证呼号-SSID及Passcode，可去22meters.com生成。
IGLOGIN BH6AJP-15 19899
# 网关自身信标，让其他人知道你的APRS iGate处于工作状态。
# every为自身信标上报间隔时间，lat和long为经纬度，comment为说明信息。
PBEACON sendto=IG delay=0:30 every=60:00 symbol="igate" overlay=R lat=32^54.26N long=115^45.49E freq="144.640" comment="Testing aprs gate raspberry wit rtl-sdr. Why can not recive anything?"

By the way , i have already used "rtl_test -p"? to test my sdr , and the result is 8 .? i have tried "rtl_fm -f 144.640M -p 8 - | direwolf -c /root/direwolf.conf -r 24000 -D 1 -"? the result is the same, it seemed that i can't recive
any radio.

Sorry for taking up your time? ? I am a ham from China.Best wish for you 73.

开云体育

Hello Rafa,

Reverting to an older kernel and firmware is not ideal as you will be missing on various security fixes, bug fixes, etc. on both the kernel and the firmware.? For impacted Bookworm users, all you need to do is upgrade to Direwolf 1.8DEV and you should be OK.? This Direwolf branch is quite stable at the moment.

--David
KI6ZHD

I've found a temp solution of the GPIO issue.

?

rpi-update cac01bed1224743104cb2a4103605f269f207b1a #6.1.54

with this command, and a reboot, you restore the function of /sys/class/gpio

?

73's Rafa EB2DJB

开云体育

Hello Robert,

After banging into the "Error writing" issue, I used your document to build direwolf DEV/1.8 on the most recent DRAWS hat image Raspbian GNU/Linux 11 (bullseye)
starting the direwolf service still fails but no longer hangs the service start

Can you be specific on which document you're following?? Btw, I believe this impactful change is only for the Bookworm OS release running the new 6.6.x kernel series. Bullseye with it's 6.1.x kernel series is not impacted.

Jun 05 07:44:07? direwolf[15736]: Error writing "12" to /sys/class/gpio/export, errno=22
Jun 05 07:44:07? direwolf[15736]: Invalid argument
Jun 05 07:44:07? direwolf[15736]: It looks like gpio with sysfs is not supported on this operating system.
Jun 05 07:44:07? direwolf[15736]: Rather than the following form, in the configuration file,
Jun 05 07:44:07 direwolf[15736]: ? ? PTT GPIO ?12
Jun 05 07:44:07 direwolf[15736]: try using gpiod form instead. ?e.g.
Jun 05 07:44:07? direwolf[15736]: ? ? PTT GPIOD ?gpiochip0 ?12
Jun 05 07:44:07 direwolf[15736]: You can get a list of gpio chip names and corresponding I/O lines with "gpioinfo" comma>
Jun 05 07:44:07 systemd[1]: direwolf.service: Main process exited, code=exited, status=1/FAILURE
Jun 05 07:44:07? systemd[1]: direwolf.service: Failed with result 'exit-code'.

You MAY want to add that installing gpiod and manually modifying the direwolf config is needed

I believe the issue here is that you haven't installed and enabled the gpiod system in your Bullseye system.? It might not even be present in that OS.? As such, I recommend you REVERT your configuration to use the non-GPIOD Direwolf configuration on your system and you should be ok.

--David
KI6ZHD