Ham Radio Control Libraries developers

John Ronan | 4 Jun 2012 17:42

Picon

yfktest/hamlib

Hi,

I started sorting out a laptop for logging for an upcoming field day in 
a few weeks. So I put Ubuntu 12.04 onto my laptop, installed hamlib and 
yfttest. Hamlib is from the distro, not from source, it is version 1.2.14.

Anyway, to test it all with my Elecraft K3 I tormented a few contesters 
Sunday morning with my out-of-practice morse and after making a few  
contacts I left it all running and went for lunch.

When I came back yfktest was hung solid. After some experimentation I 
realised it was rigctld was dying, so I ran it earlier dumping a log 
output and left it.

yfktest is polling about one per second, so the output is quite long and 
I have it here.

http://shack.ei7ig.org/rigctld.txt

Thoughts appreciated.

Regards
John
EI7IG

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware

(Continue reading)

Nate Bargmann | 4 Jun 2012 20:58

Picon

Re: yfktest/hamlib

* On 2012 04 Jun 10:46 -0500, John Ronan wrote:
> Hi,
> 
> I started sorting out a laptop for logging for an upcoming field day in 
> a few weeks. So I put Ubuntu 12.04 onto my laptop, installed hamlib and 
> yfttest. Hamlib is from the distro, not from source, it is version 1.2.14.
> 
> Anyway, to test it all with my Elecraft K3 I tormented a few contesters 
> Sunday morning with my out-of-practice morse and after making a few  
> contacts I left it all running and went for lunch.
> 
> When I came back yfktest was hung solid. After some experimentation I 
> realised it was rigctld was dying, so I ran it earlier dumping a log 
> output and left it.
> 
> yfktest is polling about one per second, so the output is quite long and 
> I have it here.
> 
> http://shack.ei7ig.org/rigctld.txt
> 
> Thoughts appreciated.

Hi John.

Do you have an estimation on how long things ran before the failure
occured?  Looking at the last part of your output:

rigctl(d): m 'currVFO' '' '' ''
k3_get_mode called
kenwood_get_mode called

(Continue reading)

John Ronan | 4 Jun 2012 23:03

Picon

Re: yfktest/hamlib


> Hi John.
>
> Do you have an estimation on how long things ran before the failure
> occured?  Looking at the last part of your output:
I'm going to say between 10 and 20 minutes (Who would leave a K3 running 
and doing nothing for that long... but thats a different discussion)

I've made a mod to yfktest to increase the delay between reads slightly, 
and it appears to have solved it. I say appears, in that I went for a 
shower, came back and it was still running.
I'll test it again tomorrow evening.

>
> rigctl(d): m 'currVFO' '' '' ''
> k3_get_mode called
> kenwood_get_mode called
> kenwood_safe_transaction called
> kenwood_transaction called
> kenwood_transaction: cmdstr = MD
> TX 3 bytes
> 0000    4d 44 3b                                            MD;
> RX 1 characters
> 0000    3b                                                  ;
> kenwood_transaction: wrong reply �� for command MD
> TX 3 bytes
> 0000    4d 44 3b                                            MD;
> read_string: timedout without reading a character
> TX 3 bytes
> 0000    4d 44 3b                                            MD;

(Continue reading)

John Ronan | 5 Jun 2012 22:05

Picon

Re: yfktest/hamlib


Evening,

Right, I left it running when I called over to EI2HIB's QTH earlier.

This is what I got
-- This first bit is from the end of the rigctld log
kenwood_transaction: wrong reply  for command BW
TX 3 bytes
0000    42 57 3b                                            BW;
read_string: timedout without reading a character
TX 3 bytes
0000    42 57 3b                                            BW;
read_string: timedout without reading a character
TX 3 bytes
0000    42 57 3b                                            BW;
read_string: timedout without reading a character
k3_get_mode: Cannot read K3 BW value
fscanf: Success
Connection closed from 127.0.0.1:53729
-- The line above is when I did a Kill -9 of yfktest.

-- I forgot what port rigctld is listening on so checked it before 
trying to connect.
j0n <at> ds9:~/Documents/Source/yfk/yfktest/trunk$ lsof -i
COMMAND  PID USER   FD   TYPE DEVICE SIZE/OFF NODE NAME
rigctld 2095  j0n    4u  IPv4  12813      0t0  TCP *:4532 (LISTEN)
j0n <at> ds9:~/Documents/Source/yfk/yfktest/trunk$ telnet localhost 4532
Trying ::1...
Trying 127.0.0.1...

(Continue reading)

John Ronan | 18 Jun 2012 10:33

Picon

Re: yfktest/hamlib

On 04/06/12 19:58, Nate Bargmann wrote:
> It would appear as though the serial link failed. What is your 
> hardware setup? Do you need to power reset your K3 after this occurs? 
> Do you have to do anything other than simply restarting rigctld? I say 
> this having run Fldigi, CQRlog, or SO2SDR for hours, even days, 
> without issue. I know, that doesn't help you too much.  73, de Nate >> 

Morning,

I used yfktest yesterday and it worked fine once I put in a large enough 
polling delay (we were operating mostly in 'run' mode so it didn't 
really matter).  I've done extensive testing of the K3 and FT-950 
backends with it in the lead up to yesterdays contest and polling at 
anything less than about 5 seconds intervals does rigctld and the K3 to 
stop responding eventually.

If I get a chance later in the week I'll throw something together to do 
nothing else but poll rigctld and see what happens.

Regards
John
EI7IG

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

(Continue reading)

Daniele Pistollato | 18 Jun 2012 11:35

Picon

R: yfktest/hamlib

Hi John,
this is a "known problem" i've already reported some time ago for K3. A
solution for this issue will make me REALLY happy. 
5 seconds is good enough for reading frequency for qso logs, but it's
totally unuseful when using rigctld for rig control.

I've another problem too... when tuning on 40 meters (and ONLY on 40 meters)
the K3 filters will be set to the NARROWEST set. This happens to you too?
This happens ONLY on 40 meters (not 80, not 30...) and obviously the code
that triggers the event it's the same.

Best
Daniele
IW3HMH

-----Messaggio originale-----
Da: John Ronan [mailto:jpronans <at> gmail.com] 
Inviato: lunedì 18 giugno 2012 10:33
A: hamlib-developer <at> lists.sourceforge.net
Oggetto: Re: [Hamlib-developer] yfktest/hamlib

On 04/06/12 19:58, Nate Bargmann wrote:
> It would appear as though the serial link failed. What is your 
> hardware setup? Do you need to power reset your K3 after this occurs?
> Do you have to do anything other than simply restarting rigctld? I say 
> this having run Fldigi, CQRlog, or SO2SDR for hours, even days, 
> without issue. I know, that doesn't help you too much.  73, de Nate 
> >>

Morning,

(Continue reading)

Nate Bargmann | 18 Jun 2012 13:27

Picon

Re: R: yfktest/hamlib

* On 2012 18 Jun 04:37 -0500, Daniele Pistollato wrote:
> Hi John,
> this is a "known problem" i've already reported some time ago for K3. A
> solution for this issue will make me REALLY happy. 
> 5 seconds is good enough for reading frequency for qso logs, but it's
> totally unuseful when using rigctld for rig control.

Guys, keep observing things and helping me narrow down this problem.  It
is fixable!  So this is only happening with the K3?

> I've another problem too... when tuning on 40 meters (and ONLY on 40 meters)
> the K3 filters will be set to the NARROWEST set. This happens to you too?
> This happens ONLY on 40 meters (not 80, not 30...) and obviously the code
> that triggers the event it's the same.

I submitted a patch to the K3 backend that now processes
RIG_PASSBAND_NORMAL correctly.  It had been ommitted.  Right now the
patch only exists in the Hamlib-1.2.15 branch and I have the source
tarball up at http://n0nb.users.sourceforge.net/  Would you like me to
make a Win32 binary for testing as well?

73, de Nate >>

--

-- 

"The optimist proclaims that we live in the best of all
possible worlds.  The pessimist fears this is true."

Ham radio, Linux, bikes, and more: http://www.n0nb.us

(Continue reading)

Daniele Pistollato | 18 Jun 2012 14:37

Picon

R: R: yfktest/hamlib

Nate, if you make an executable (win32) package i will check it AS SOON AS I
COME BACK HOME.
Note to self: I should make a compiling virtual machine...

73
Daniele

-----Messaggio originale-----
Da: Nate Bargmann [mailto:n0nb <at> n0nb.us] 
Inviato: lunedì 18 giugno 2012 13:27
A: hamlib-developer <at> lists.sourceforge.net
Oggetto: Re: [Hamlib-developer] R: yfktest/hamlib

...
I submitted a patch to the K3 backend that now processes RIG_PASSBAND_NORMAL
correctly.  It had been ommitted.  Right now the patch only exists in the
Hamlib-1.2.15 branch and I have the source tarball up at
http://n0nb.users.sourceforge.net/  Would you like me to make a Win32 binary
for testing as well?

73, de Nate >>

--

-- 

"The optimist proclaims that we live in the best of all possible worlds.
The pessimist fears this is true."

Ham radio, Linux, bikes, and more: http://www.n0nb.us

----------------------------------------------------------------------------

(Continue reading)

Nate Bargmann | 18 Jun 2012 19:14

Picon

Re: R: R: yfktest/hamlib

* On 2012 18 Jun 07:38 -0500, Daniele Pistollato wrote:
> Nate, if you make an executable (win32) package i will check it AS SOON AS I
> COME BACK HOME.

I will certainly do so.

> Note to self: I should make a compiling virtual machine...

I use an Xubuntu 10.10 install in a VM to do the Win32 binary build.
Debian Sid has moved to a much different set of Ming, MingW64 as I
recall, that I didn't get to work in the short time I tried.

73, de Nate >>

--

-- 

"The optimist proclaims that we live in the best of all
possible worlds.  The pessimist fears this is true."

Ham radio, Linux, bikes, and more: http://www.n0nb.us

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Nate Bargmann | 23 Jun 2012 14:36

Picon

Re: R: R: yfktest/hamlib

* On 2012 18 Jun 07:38 -0500, Daniele Pistollato wrote:
> Nate, if you make an executable (win32) package i will check it AS SOON AS I
> COME BACK HOME.

Hi Daniele.

I finally got around to building a Win32 binary of 1.2.15.2~rc2.  It's
at the usual place:

http://n0nb.users.sourceforge.net/

73, de Nate >>

--

-- 

"The optimist proclaims that we live in the best of all
possible worlds.  The pessimist fears this is true."

Ham radio, Linux, bikes, and more: http://www.n0nb.us

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

Daniele Pistollato | 23 Jun 2012 15:17

Picon

Re: yfktest/hamlib

Great Nate :)
Can i redirect some inquiries from users with troubles to you?
Sometime i have no anwer...

Log4om is getting a boost in popularity from usa and from ppl involved
in the old ham radio deluxe project... So expect a boost in hamlib
users.

Many users reports problems connecting to high numbered serial ports.
Lowering the serial port number generally solves... Ideas?

Ideas?

2012/6/23, Nate Bargmann <n0nb <at> n0nb.us>:
> * On 2012 18 Jun 07:38 -0500, Daniele Pistollato wrote:
>> Nate, if you make an executable (win32) package i will check it AS SOON AS
>> I
>> COME BACK HOME.
>
> Hi Daniele.
>
> I finally got around to building a Win32 binary of 1.2.15.2~rc2.  It's
> at the usual place:
>
> http://n0nb.users.sourceforge.net/
>
> 73, de Nate >>
>
> --
>

(Continue reading)

Stephane Fillod | 23 Jun 2012 16:00

Picon

Re: yfktest/hamlib

sab, Jun 23, 2012, Daniele Pistollato skribis:
> Great Nate :)
> Can i redirect some inquiries from users with troubles to you?
> Sometime i have no anwer...
> 
> Log4om is getting a boost in popularity from usa and from ppl involved
> in the old ham radio deluxe project... So expect a boost in hamlib
> users.
> 
> Many users reports problems connecting to high numbered serial ports.
> Lowering the serial port number generally solves... Ideas?
> 
> Ideas?

Known problem, it will be solved in the next release. In the mean time,
prefix the serial port name with \\.\

  http://article.gmane.org/gmane.linux.hams.hamlib.devel/3651

73
--

-- 
Stephane - F8CFE

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

(Continue reading)

John Ronan | 24 Jun 2012 23:54

Picon

Re: R: yfktest/hamlib

On 18/06/12 10:35, Daniele Pistollato wrote:
> Hi John,
Hi,

> this is a "known problem" i've already reported some time ago for K3. A
> solution for this issue will make me REALLY happy.
> 5 seconds is good enough for reading frequency for qso logs, but it's
> totally unuseful when using rigctld for rig control.
See comments below, I didn't get to check the filter settings.
> I've another problem too... when tuning on 40 meters (and ONLY on 40 meters)
> the K3 filters will be set to the NARROWEST set. This happens to you too?
> This happens ONLY on 40 meters (not 80, not 30...) and obviously the code
> that triggers the event it's the same.
>
> Best
> Daniele
> IW3HMH
Purely out of curiosity, I set the following up today before I left, as 
I knew I would not be able to get any decent testing done.

So I ran

rigctld -vvvvvvvvv -m 229 -r /dev/ttyS1 -s 38400

Then I ran a small shell script against it.

--
j0n <at> mini-itx:~$ cat k3test.sh
while true; do
  echo "f" | nc localhost 4532

(Continue reading)

Return

Return to gmane.linux.hams.hamlib.devel.

Project Web Page

Ham Radio Control Libraries developers

Search Archive

Language

Change language

Options

Current view: Threads only / Showing only 20 lines / Not hiding cited text.
Change to All messages, whole messages, or hide cited text.

Post a message
NNTP Newsgroup
Classic Gmane web interface
XML

XML

RSS Feed
List Information

About Gmane