¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io
  1. Bpq32
  2. Messages
Expanded Topics Messages
Date
Re: BBS Error not connecting on all our linbpq nodes with latest 2 versions.

 

So this is what LLM is suggesting me to do for my error, any recommendation from the group before going this route?
?

The error log you're seeing indicates that linbpq crashed with a segmentation fault (SIGSEGV) when it received a connection request via RF. Here¡¯s a breakdown of what could be causing the issue and how to proceed:

? Key Details from the Log:

May 14 14:51:09 kp4dog3 bash[608]: SIGSEGV Received
May 14 14:51:09 kp4dog3 bash[608]: ./linbpq(+0x10585a)[0x59585a]

  • SIGSEGV (Segmentation Fault): This happens when a program tries to access a memory location it's not allowed to. This often indicates a bug, such as a null pointer dereference or buffer overrun.

  • The offset +0x10585a and address 0x59585a refer to where in the binary the crash happened, which could help the developer of linbpq trace the bug if you report it.

? Possible Causes:

  1. Malformed or Unexpected Packet over RF:

    • The BBS might be receiving a connection packet with a format that crashes linbpq due to missing or corrupted data (e.g., bad AX.25 header).

  2. Old or Corrupt bpq32.cfg Configuration:

    • A misconfigured port or application section in your bpq32.cfg might cause a crash if linbpq tries to parse or act on malformed config data triggered by the connection.

  3. Known Bug in linbpq Version:

    • You might be running a version of linbpq with a known bug that gets triggered under certain RF conditions.

  4. Corrupted linbpq Binary or Libraries:

    • If there are filesystem issues (e.g., SD card corruption), the executable or a supporting library (e.g., libc) could be damaged.

  5. Bad Call Sign Format or BBS Name Length:

    • If the incoming connection uses an invalid call sign format or unescaped special characters, it could cause memory handling problems.

? Recommended Actions:

  1. Update linbpq:

    • Ensure you're running the latest linbpq binary from G8BPQ. This is important, as John Wiseman updates the binary often with bug fixes.

  2. Enable Debug Logging:

    • Temporarily increase verbosity in the config by enabling TRACE and DEBUG on the port handling RF. This may give more clues in /var/log/syslog or terminal output about what packet causes the crash.

  3. Check bpq32.cfg:

    • Look for:

      • Unusual APPLICATION definitions.

      • Mismatched PORT or PORTCALL fields.

      • Long ALIAS or BBS names exceeding 6 characters.

      • Any non-printable characters or encoding issues.

  4. Rebuild linbpq:

    • If running from source, do a clean rebuild to rule out binary corruption:

      make clean
make

  5. Run linbpq with gdb (for advanced debugging):

    • If you¡¯re comfortable, you can run linbpq inside gdb:

      gdb ./linbpq
(gdb) run

      Wait for crash, then type:

      (gdb) backtrace

      This will show the function call stack when it crashed.

  6. Check for Hardware Issues:

    • If you are running on a Raspberry Pi, unstable power or SD card corruption can cause random segfaults.

? Optional: Provide to John Wiseman

You can email g8bpq@... with:

  • Description of the crash

  • Your bpq32.cfg file

  • Version and platform of linbpq

  • The exact syslog output

He¡¯s usually very responsive.

Would you like help checking your bpq32.cfg file for possible problems that could trigger this crash?
Re: RF Path

 

Yes, both are set the same. I've been testing for several days, but DW always comes through.

When QtSM and BPQ are running, it will actually digi and send an rf beacon, it seems.?

Sent with secure email.

On Thursday, May 15th, 2025 at 9:57 AM, Lee Bengston via groups.io <kilo5dat@...> wrote:

Given Direwolf is working, I would verify what path is configured in Direwolf and apply it to BPQ if it's different.

Are both QtSM and DW configured to use the same method to key the radio?

73,
Lee K5DAT


On Thu, May 15, 2025 at 8:48?AM Don via <donpjr=[email protected]> wrote:
Hi Mark,
Once I switched from BPQ and QtSM to Direwolf stand-alone last evening, the RF path popped up again. So, other than your suggestions, BPQ is missing something. Who knows?
I was using 
APRSPATH 3=APRS,WIDE2-1 after research showed it was best for a fixed station. I will try your suggestion, however.
Re: RF Path

 

Given Direwolf is working, I would verify what path is configured in Direwolf and apply it to BPQ if it's different.

Are both QtSM and DW configured to use the same method to key the radio?

73,
Lee K5DAT


On Thu, May 15, 2025 at 8:48?AM Don via <donpjr=[email protected]> wrote:
Hi Mark,
?
Once I switched from BPQ and QtSM to Direwolf stand-alone last evening, the RF path popped up again. So, other than your suggestions, BPQ is missing something. Who knows?
?
I was using? 
APRSPATH 3=APRS,WIDE2-1 after research showed it was best for a fixed station. I will try your suggestion, however.
Re: RF Path

 

Hi Mark,
?
Once I switched from BPQ and QtSM to Direwolf stand-alone last evening, the RF path popped up again. So, other than your suggestions, BPQ is missing something. Who knows?
?
I was using? 
APRSPATH 3=APRS,WIDE2-1 after research showed it was best for a fixed station. I will try your suggestion, however.
Re: RF Path

 

I would change the APRSPath to that which is suggested by the docs.
?
APRSPATH 3=APRS,WIDE1-1,WIDE2-1 ; Dest and zero to 8 digis
I don't know where you got your ISFilter from, and I am no expert on filters, but for testing I would change it to something simple...
ISFilter=m/64
?
It will not send to the servers without your passcode. Don't know if you just did not include it or what. APRS Passcodes are not secret..
ISPasscode=8334???
?
Other than those, I see nothing that would stop it from reporting any data received on port 3.? But, if your port 3 is not actually receiving aprs traffic there would be nothing to report to the servers.
?
But.... looking at your raw packets being reported you are showing bothe TCP and RF packets, so I do not see a problem with it.
?
?
--
73,
Mark, N5MDT
Montgomery, Texas
?
?
?
Re: Accountability for NTS messages retrieved from MBO/BBS

 

Yes, see the BPQ Log file.
?
73 - Dave - W7EES
Re: Strange SEG Fault - need help on this one.

 

Well John - this new executable crashes on connect to the BBS alias.? ran it twice to be sure then did a debug capture and put the whole mess into a new file (attached).
--
Charlie (N4NVD)

gdb_output5.txt
gdb_output5.txt

Re: Strange SEG Fault - need help on this one.

 

Thanks John - just implemented the new executable, will test as soon as I finish attaching my bpq.cfg (renamed bpq32.sav so I can clear the passwords)
--
Charlie (N4NVD)

bpq32.sav
bpq32.sav

Re: Strange SEG Fault - need help on this one.

 

¿ªÔÆÌåÓý

Somehow the Mode and Frequency of the incoming connect are getting corrupted.

This version should stop it crashing:



But I'd like to track down the cause of the corruption, so could you send me your bpq32.cfg?

Thanks,
John


On 14/05/2025 21:40, Charlie Hein wrote:

Sorry about that, here are the same two tests, calling the BBS alias directly for #3 and calling the BBS from the node menu on #4
--
Charlie (N4NVD)
Re: RF Path

 

Hey, Mark
?
APRSDIGI
?APRSCall W9JUN-10
?StatusMsg=Madison, IN EM78ht
?
?Symbol=# ? ? ? ? ; Icon to display for station.
?Symset=I ? ? ? ? ; This is a "B" in a Red Diamond.?
?PHG=PHG5000/
?;
?; ? ?Specify a Dest Addreess and Path for each port you want to use for APRS. No dest means receive only port
?; ? ?Ports without an APRSPath statement will not be used by the Digi
?;
?; ? ?Note if you specify APRS as the destination call it will be replaced by the current BPQ32 APRS
?; ? ?identification destination - APBPQ1. If you really want to send APRS, specify APRS-0.?
?
? ?;APRSPath 1= ? ? ? ? ? ? ? ? ? ? ? ? ; Receive only port
? ;APRSPath 7=APRS,WIDE1-1 ? ? ? ? ? ? ; Dest and up to 8 digis
? APRSPath 3=APRS,WIDE2-1
??
? BeaconInterval=15
? BeacontoIS=1 ? ? ? ? ? ? ? ? ? ; Send Beacon every 30 minute. Minimum is 5 mins
? ;MobileBeaconInterval=2 ? ? ? ? ? ? ?; Beacon interval when moving
? MaxAge=60
?; ? ?Define Crossport Digi. For each port, a list of ports to send a packet to (IS means send to APRS-IS)
?; ? ?If a Digimap isn't specified for a port, digi out on the port the packet arrived on,
?; ? ? ? ?and send to APRS-IS (if enabled)
?; ? ?If you want to digi on same port, but not send to APRS-IS, Specify Digimap n=n?
?; ? ?If you dont want to digi on a port, specify Digimap n=
?;
?; ? ?If you only have one APRS port, you probably don't need a Digimap statement
?
?;Digimap 1=9 ? ? ? ? ? ?; Packets from 1 to 9, but not APRS-IS
?Digimap 3=3,IS ? ? ? ; Packets from 7 to 7, 9 and APRS-IS
?;Digimap 9= ? ? ? ? ? ? ; Don't digi packets received on port 9
?
?; ? ?Define Digipeating parameters
?
?TraceCalls=WIDE,TRACE ? ? ? ? ?; Calls for CALLN-n Processing with Trace
?FloodCalls=IN ? ? ? ? ; Calls for CALLN-n Processing without Trace
?DigiCalls=W9JUN-10 ? ? ? ; Calls for Normal (ie no SSID manipulation) Digi
? ? ? ? ? ? ? ? ? ? ? ?; For a Local "Fillin" Digi, you could include WIDE1-1 and leave out TraceCalls and FloodCalls
? ? ? ? ? ? ? ? ? ? ? ??
?;ReplaceDigiCalls ? ? ? ; Replace DigiCalls with APRSCall. Omit if you want the call to be left as it is received.
? ? ? ? ? ? ? ? ? ? ? ??
?;MaxTraceHops=2 ? ? ? ? ; Max value of n in CALLN-n processing. If a packet is received with n greater than
?;MaxFloodHops=2 ? ? ? ? ; this, the value will be replaced BEFORE it is decremented.
?; ? ?Specify position. You can specify a fixed LAT/LONG, or use GPS
?LAT=3847.78N ? ? ? ? ? ; Must be in standard APRS Format (ddmm.mmN/S)
?LON=08522.26W ? ? ? ? ?; Must be in standard APRS Format (dddmm.mmE/W)
?;GPSPort=/dev/ttyACM0 ? ? ? ? ? ? ?; if specified, lat/lon will be taken from a GPS,
?;GPSSpeed=9600 ? ? ? ? ? ? ; overriding the values set above.

?; The Digi can generate Object and Item reports.?
?
?; The format is pretty strict - a single space between param=value pairs,
?; a comma but no spaces in PATH and PORT string.
?
?; Minimum interval is 10 mins. You can have as many OBJECT lines as you need.
?; See thr APRS specification for information on how to format an Object or Item string.
?
?;OBJECT PATH=APRS,WIDE2-2 PORT=3,IS INTERVAL=30 TEXT=;444.80BPQ*111111z3847.78N/08522.26Wr%156 R15m
?
?; IGATE Params. If specified you will gate RF packets and your beacons to APRS-IS
?ISHost=noam.aprs2.net ? ?; APRS-IS Host Name. May be either an IPV4 or an IPV6 Host
?ISPort=14580 ? ? ? ? ? ? ? ?; Normal port for a filtered feed
?ISPasscode= ? ? ? ? ? ? ? ?? ; If you don't already have a passcode I can issue you one.
?;
?; You can specify a filter command to be sent when you log on to APRS-IS. Normally you don't
?; need one. The APRS Mapping Application (BPQAPRS) sets a filter when it starts,
?; and if you aren't running an APRS application, there isn't much point in getting info from APRS-IS
?;
?ISFilter=m/64 b/-CW -d/DL4MDW* -s/w -t/t -t/w ?
?;GateLocalDistance=30
?;GATEDPATH 3=APRS
LOGAPRSIS=1
; Filter Command if needed
Re: Strange SEG Fault - need help on this one.

 

Sorry about that, here are the same two tests, calling the BBS alias directly for #3 and calling the BBS from the node menu on #4
--
Charlie (N4NVD)

gdb_output3.txt
gdb_output3.txt
gdb_output4.txt
gdb_output4.txt

Re: Strange SEG Fault - need help on this one.

 

¿ªÔÆÌåÓý

It looks like the bt command didn't get processed.

When you get --Type <RET> for more, q to quit, c to continue without paging--

type c, then after you get the message 'Thread 1 "linbpq" received signal SIGSEGV, Segmentation fault.' enter the bt.

Thanks,
John

On 14/05/2025 19:23, Charlie Hein wrote:

Hi John - captured the output off the terminal window for two files (attached)...?
?
gdb_output1.txt was when I went out on the VHF port directly to the BBS alias (BCSBBS) via the TAMU-11 digipeater using QTterm TCP?
?
gdb_output2.txt is a call made the same way but calling the node alias (BCS) first, playing back the INFO text, then using the BBS menu option to try to reach the BBS.
?
Hope that this is helpful.
?
--
Charlie (N4NVD)
linbpq and VARA

 

Good day,

??? I have linbpq and VARA running on different computers, I connect VARA no problem, but I have to reload BPQ daylies it lose the TCPIP? connection to my VARA.

Is there a watch dog of some sort in BPQ to re-connect VARA automatically?

--
73 de Jean,
VA2OM / VE2PKT

Sysop de: VE2PKT (BBS), VE2PKT-2 (BPQNode), VE2PKT-4 (XRPI)
: VE2PKT-5 (Jnos), VE2PKT-9 (DXCluster)

RF: 145.05 Mhz (1.2K Bps) Packet, 147.435 Mhz, VARA

Internet:
(XRLin node) - VE2PKT-4
(BPQ BBS ) - VE2PKT
Telnet: ve2pkt.dyndns.org port 2323 (Xrpi Node) - VE2PKT-4
Telnet: ve2pkt.dyndns.org port 9000 (DXCluster) - VE2PKT-9

AMPR Network:
Telnet xrouter.ve2pkt.ampr.org Port 2323 (XRPI Node) = VE2PKT-4
Telnet dxc.ve2pkt.ampr.org Port 9000 (DXCluster) = V2PKT-9
Telnet jnos.ve2pkt.ampr.org Port 23 (Jnos) = VE2PKT-5


E-Mail:
packet: va2om@ve2pkt.#qbc.qc.can.noam
ampr net: va2om@...
Inet: ve2pkt@...

Membre Createur du groupe [email protected]

/g/PacketQuebec

Carte des Station RF au Quebec
Accountability for NTS messages retrieved from MBO/BBS

 

NTS requires accountability for messages relayed through the NTS including the Digital Traffic Network (DTN). If a DTN hub station, using BPQ, auto-forwards a message to a specific station, there is a record as to whom a message was forwarded. However if a destination hub station is queried as to message traffic available for download and the one making the query reads and then kills a message, is there a way for a Sysop to determine who accepted the message? And if not, could that be implemented? Thanks for any help.?
Re: BBS Error not connecting on all our linbpq nodes with latest 2 versions.

 

John
?
I had some family emergency that could not follow trough with your previous request.
?
I did upgrade to the latest version .70 but still the same is occurring, any one trying to connect to the BBS via RF it would not work.
?
I am including the console log of linbpq as previously requested.
?
At the end of the file you will see I did tried to connect to the BBS and it just ended the stream it was around local time 2:50 pm.
?
Thanks
?

status.zip
status.zip

Re: Strange SEG Fault - need help on this one.

 

Hi John - captured the output off the terminal window for two files (attached)...?
?
gdb_output1.txt was when I went out on the VHF port directly to the BBS alias (BCSBBS) via the TAMU-11 digipeater using QTterm TCP?
?
gdb_output2.txt is a call made the same way but calling the node alias (BCS) first, playing back the INFO text, then using the BBS menu option to try to reach the BBS.
?
Hope that this is helpful.
?
--
Charlie (N4NVD)

gdb_output1.txt
gdb_output1.txt
gdb_output2.txt
gdb_output2.txt

Re: Strange SEG Fault - need help on this one.

 

¿ªÔÆÌåÓý

Charlie,

Can you try to get a debug trace of the crash using gdb. See this for details:



Thanks,
John


On 14/05/2025 16:55, Charlie Hein wrote:

Apologies in advance if this is a repeat message, but my earlier reply did not show up and I fear that it was lost.
?
I found an older backup running 6.0.24.40 and when I loaded it up was pleased to discover that it worked okay - in that it did NOT end with a segment fault if someone connected to the BBS on the node's VHF port.??
?
I made a copy of the card for safekeeping and replaced the executable (and the HTML folder) with the newest executable (6.0.24.70) without making any other changes to the configuration, and the problem returned.? Deleting the .70 executable and returning to .40 restored access to the BBS over the VHF port.? Clearly there's been a change that brought on this behavior at some point between when I made the backup of .40 and the release of .70.?
?
@John(G8BPQ):? I'm very sorry that I did not think to check for this sort of odd behavior.? My local VHF port doesn't get a lot of use, and most of that is connections to the node itself. or the chat, both of which do not exhibit this issue.? I would certainly have noticed it sooner if I had more BBS traffic over that port.
?
The node is running on a Pi 3B+ using an NW Digital Radio DRAWS Hat.? ().? The soundmodem is DIREWOLF.??
?
Obviously, I'm glad to help in any way I can - just let me know what you need.?
--
Charlie (N4NVD)
Re: Strange SEG Fault - need help on this one.

 

Apologies in advance if this is a repeat message, but my earlier reply did not show up and I fear that it was lost.
?
I found an older backup running 6.0.24.40 and when I loaded it up was pleased to discover that it worked okay - in that it did NOT end with a segment fault if someone connected to the BBS on the node's VHF port.??
?
I made a copy of the card for safekeeping and replaced the executable (and the HTML folder) with the newest executable (6.0.24.70) without making any other changes to the configuration, and the problem returned.? Deleting the .70 executable and returning to .40 restored access to the BBS over the VHF port.? Clearly there's been a change that brought on this behavior at some point between when I made the backup of .40 and the release of .70.?
?
@John(G8BPQ):? I'm very sorry that I did not think to check for this sort of odd behavior.? My local VHF port doesn't get a lot of use, and most of that is connections to the node itself. or the chat, both of which do not exhibit this issue.? I would certainly have noticed it sooner if I had more BBS traffic over that port.
?
The node is running on a Pi 3B+ using an NW Digital Radio DRAWS Hat.? ().? The soundmodem is DIREWOLF.??
?
Obviously, I'm glad to help in any way I can - just let me know what you need.?
--
Charlie (N4NVD)
Re: RF Path

 

I don't know if this applies to your test setup or not but:

aprsis will reject duped packets automagically

so, if you send the 'same' packet three different ways, only the first will show up on

usually, the 'winner' would be direct internet connection to aprsis.

:)

good luck
kb0wlf


On Wed, May 14, 2025 at 8:30?AM Don via <donpjr=[email protected]> wrote:
Good day and thanks for the response.

I am currently using QtSoundModem which seems to be receiving and transmitting ok. As I love a challenge and do a lot of experimentation, I've been comparing DW as a standalone vs LinBPQ and Qt Sm.

Sent with secure email.

On Wednesday, May 14th, 2025 at 9:20 AM, Lee Bengston via <kilo5dat=[email protected]> wrote:
Looking at , it says that W9JUN-10 is BPQ and that it was heard directly via radio by 7 stations. If it was actually Direwolf transmitting when it was heard, that implies you are using the same callsign/SSID in Direwolf that you are using in BPQ. That may be the problem right there. I would configure Direwolf with a different SSID, and also comment out anything in the dw config that would cause it to transmit on RF, i.e. igating or digipeating. That should remove any conflict with what BPQ is trying to do.

Also, when you first start Direwolf, it should tell you when it receives either a KISS over TCP or AGW connection from BPQ. If you don't see one, then you will never see BPQ transmit on RF.

73,
Lee K5DAT
Re: RF Path

 

I'll do so when I get home this evening. Just taking a chance while I'm away -

I appreciate it

Sent with secure email.

On Wednesday, May 14th, 2025 at 9:22 AM, Mark Taylor <mtaylor@...> wrote:

Don, perhaps if you provided some relevant information we could assist.
?
Your Call, your APRSCall, all of your DigiMap lines, all of your APRSPath lines, and a description of your ports including the port number and description (RF, VHF, etc.)
Maybe your entire APRSDIGI section.
?
Something that might give us a clue to help you get a working configuration that you want.
--
73,
Mark, N5MDT
Montgomery, Texas
?
?
?
61 - 80 of 43797