Re: xlog hamlib functionality with model 2
On 08/02/2021 20:00, Andy Stewart wrote: On 2/7/21 7:00 PM, Barry Jackson wrote: On 07/02/2021 17:04, Dr. Karsten Schmidt wrote: Hi Barry, yes I have a similar problem. A "2" in the GUI ist stored as "1" in the config file. -K 73 de dg1vs Gentlemen, Please do not edit the config file directly. Please use the GUI or you will cause yourself problems, as you've noticed. How the data is stored in the xlog.cfg file is of no consequence as long as xlog can read/write it and behave correctly. Through the GUI, I have seen no failure. If you do things through the GUI and see a functional misbehavior, please let me know the steps to reproduce it. When I tried it, there was no problem with xlog 2.0.20 and hamlib4. Thanks, and 73, Andy Hi Andy, I finally got to the bottom of this. There were two serial port parameters that are incorrect in hamlib for my radio, such that I have always had to use the manual entry for stop_bits and handshake in most applications. Do the hamlib default settings overwrite the manual ones I wonder? I finally corrected these in the hamlib sources and rebuilt our hamlib package. Without the manual settings being needed all works correctly as you describe when using the rig selection dialogue. I have sent the corrections upstream to hamlib, so while this fixes my TS-450S issue it won't help Karsten. Thanks for your help, 73 Barry
Re: xlog hamlib functionality with model 2
On 07/02/2021 17:04, Dr. Karsten Schmidt wrote: Hi Barry, yes I have a similar problem. A "2" in the GUI ist stored as "1" in the config file. -K 73 de dg1vs Am 07.02.2021 um 12:37 schrieb Barry Jackson: On 02/02/2021 22:14, Dr. Karsten Schmidt wrote: Hi Graeme, I have the same client... Quisk. Btw. I'm using hamlib 4.0 73, Karsten dg1vs On Wed, 2021-02-03 at 11:00 +1300, Graeme Jury wrote: Hello Karsten, It will be no consolation but I get the same error. I am using an SDR client called Quisk and it communicates without any problems with cqrlog. I have not tried to analyze it any further but will have a go soon. 73, Graeme ZL2APV On 3/02/21 9:29 am, Dr. Karsten Schmidt wrote: Hallo, maybe someone can help me. I’m trying to get xlog hamlib functionality running. My Rig is a Quisk-software-transceiver which is accessible via model 2 from hamlib. The port is localhost and the port 4575. After setting up, it reads out the frequency once. Afterwards xlog reports an error via a dialog box “There was an error while communicating with our rig, hamlib has stopped”. I’m quite sure, that hamlib itself is able to handle to read out the frequency from quisk, since the commad “rigctl –model=2 –rig-file=localhost:4575” works like a charm. Tracing the hamlib traffic via xlog doesn’t work, it’s to fast to enable the hamlib logging. Any hints? Kind regards Karsten 73 de dg1vs I have been trying to report issues with xlog and hamlib-4.x for a while now, but have problems with my mails being bounced, or ignored. I find that if the hamlib model number is correct in the config file then it works, but the the hamlib setup GUI dialogue displays the model number minus one. My correct radio is model 2003, which I have set manually in the config file. It shows in the GUI as 2002 but works. Setting the radio model in the GUI puts 2003 in the GUI, but in the config fle it is stored as 2004. This needs sorting out. Check what is in your config file! Cheers, Barry G4MKT Did putting the correct ID (2) in the config file fix it? Cheers, Barry G4MKT
Re: xlog hamlib functionality with model 2
On 07/02/2021 19:28, Andy Stewart wrote: Hi Andy, I hope you get this :) No special steps to duplicate this - exactly as I described to Karsten. If I put my correct Model ID for TS-450S in the config file as 2003 then run xlog it works OK but the GUI in the Settings->Preferences->hamlib dialogue shows model ID as 2002. If I select the model using the GUI as TS450S then the model is shown in the GUI as 2003, but in the config file it is stored as 2004 and does not work. If I select a TS-440S which is 2002 then it stores 2003 in the the config and my TS-450S works fine! I tried to send some links to lists of system component version numbers etc. in the past but all these were bounced so I will leave it at this for now, but please get back with anything that you need to debug this. Sadly we are now in version freeze for Mageia 8 and this will have to be dealt with as a bug report and update after release. I was hoping to get a fix for this before now but all my attempts to contact you failed. Cheers, Barry G4MKT (Mageia maintainer for xlog) HI Barry, While the behavior is admittedly strange, it is the way that the software works. For some reason, it seem that all values are off by one when the file is written, and changed by one when the config file is read by xlog, thus everything works. But it does NOT work! If a user enters his radio Make/Model in the GUI it puts the wrong hamlib ID in the config. (correct ID + 1) That does not work. It only works with the correct hamlib ID in the config file. Are you using hamlib 4.0 or 4.1 or 4.2 ?? I noticed it in my case with my IC-7300 (model 3073). It writes 3074 in the xlog.cfg file, but it works just fine. I guess the moral of the story is: either don't directly edit the xlog.cfg file, or understand this behavior and edit accordingly. 73, Andy Take another look. Maybe *your* radio works with the next higher hamlib ID? This has been an issue since hamlib-4.0 was introduced. Cheers, Barry G4MKT
Re: xlog hamlib functionality with model 2
On 07/02/2021 17:14, Andy Stewart wrote: I have been trying to report issues with xlog and hamlib-4.x for a while now, but have problems with my mails being bounced, or ignored. I find that if the hamlib model number is correct in the config file then it works, but the the hamlib setup GUI dialogue displays the model number minus one. My correct radio is model 2003, which I have set manually in the config file. It shows in the GUI as 2002 but works. Setting the radio model in the GUI puts 2003 in the GUI, but in the config fle it is stored as 2004. This needs sorting out. Check what is in your config file! Cheers, Barry G4MKT Hi Barry, I have never seen this issue of the model number being reported as -1. Can you please describe the scenario precisely so I can duplicate it? Please note that my email address has changed and *@mval.net will not work. Thanks, and 73, Andy xlog maintainer Hi Andy, I hope you get this :) No special steps to duplicate this - exactly as I described to Karsten. If I put my correct Model ID for TS-450S in the config file as 2003 then run xlog it works OK but the GUI in the Settings->Preferences->hamlib dialogue shows model ID as 2002. If I select the model using the GUI as TS450S then the model is shown in the GUI as 2003, but in the config file it is stored as 2004 and does not work. If I select a TS-440S which is 2002 then it stores 2003 in the the config and my TS-450S works fine! I tried to send some links to lists of system component version numbers etc. in the past but all these were bounced so I will leave it at this for now, but please get back with anything that you need to debug this. Sadly we are now in version freeze for Mageia 8 and this will have to be dealt with as a bug report and update after release. I was hoping to get a fix for this before now but all my attempts to contact you failed. Cheers, Barry G4MKT (Mageia maintainer for xlog)
Re: xlog hamlib functionality with model 2
On 02/02/2021 22:14, Dr. Karsten Schmidt wrote: Hi Graeme, I have the same client... Quisk. Btw. I'm using hamlib 4.0 73, Karsten dg1vs On Wed, 2021-02-03 at 11:00 +1300, Graeme Jury wrote: Hello Karsten, It will be no consolation but I get the same error. I am using an SDR client called Quisk and it communicates without any problems with cqrlog. I have not tried to analyze it any further but will have a go soon. 73, Graeme ZL2APV On 3/02/21 9:29 am, Dr. Karsten Schmidt wrote: Hallo, maybe someone can help me. I’m trying to get xlog hamlib functionality running. My Rig is a Quisk-software-transceiver which is accessible via model 2 from hamlib. The port is localhost and the port 4575. After setting up, it reads out the frequency once. Afterwards xlog reports an error via a dialog box “There was an error while communicating with our rig, hamlib has stopped”. I’m quite sure, that hamlib itself is able to handle to read out the frequency from quisk, since the commad “rigctl –model=2 –rig-file=localhost:4575” works like a charm. Tracing the hamlib traffic via xlog doesn’t work, it’s to fast to enable the hamlib logging. Any hints? Kind regards Karsten 73 de dg1vs I have been trying to report issues with xlog and hamlib-4.x for a while now, but have problems with my mails being bounced, or ignored. I find that if the hamlib model number is correct in the config file then it works, but the the hamlib setup GUI dialogue displays the model number minus one. My correct radio is model 2003, which I have set manually in the config file. It shows in the GUI as 2002 but works. Setting the radio model in the GUI puts 2003 in the GUI, but in the config fle it is stored as 2004. This needs sorting out. Check what is in your config file! Cheers, Barry G4MKT
Re: Xlog and Hamlib-4
On 27/11/2020 02:40, Andy Stewart wrote: On 11/26/20 6:36 PM, Barry Jackson wrote: On 31/10/2020 02:55, Andy Stewart wrote: HI Barry, -- Let me know if you continue to have issues. Hi Andy, OK I finally found time to have another look at this. I started by rebuilding against a recent hamlib snapshot. Hi Barry, I need very specific details in order to attempt to reproduce this problem. * if the xlog version is not 2.0.19, please try again with that version, which is the latest * Exact version of hamlib (ideally, the link from where you downloaded it) * Exact type and version of Linux * If you did any custom configuration when you compiled hamlib and/or xlog, I'll need those instructions as well * I don't have a TS-405S, so I may not be able to reproduce the problem, but I will try. Thanks, and 73, Andy Hi Andy, I have tried to reply to you 4 times now and am about to give up. Mail is just rejected. Latest was returned today after 8 days of attempts. I am sending this as a final test. Do you have another mail account? Cheers, Barry
Re: Xlog and Hamlib-4
With 2003 (TS-450S correct code) in the config file all is working fine,
except for "2002" shown in the Hamlib settings dialogue and the crash on
hitting the 'Select a Radio' button.
Here is the tail of strace after hitting the Settings > Preferences >
Hamlib > Select a Radio Button.
poll([{fd=3, events=POLLIN}, {fd=4, events=POLLIN}, {fd=6,
events=POLLIN}], 3, 0) = 0 (Timeout)
getpid()= 1811321
getpid()= 1811321
getpid()= 1811321
poll([{fd=3, events=POLLIN|POLLOUT}], 1, -1) = 1 ([{fd=3, revents=POLLOUT}])
writev(3,
[{iov_base="\24\0\6\0\236\5\240\2\321\1\0\0\6\0\0\0\0\0\0\0\377\377\377\377",
iov_len=24}, {iov_base=NULL, iov_len=0}, {iov_base="", iov_len=0}], 3) = 24
poll([{fd=3, events=POLLIN}], 1, -1)= 1 ([{fd=3, revents=POLLIN}])
recvmsg(3, {msg_name=NULL, msg_namelen=0, msg_iov=[{iov_base="\1
:3\1\0\0\0\6\0\0\0\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"...,
iov_len=4096}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 36
poll([{fd=3, events=POLLIN|POLLOUT}], 1, -1) = 1 ([{fd=3, revents=POLLOUT}])
writev(3, [{iov_base="\16\0\2\0\236\5\240\2", iov_len=8},
{iov_base=NULL, iov_len=0}, {iov_base="", iov_len=0}], 3) = 8
poll([{fd=3, events=POLLIN}], 1, -1)= 1 ([{fd=3, revents=POLLIN}])
recvmsg(3, {msg_name=NULL, msg_namelen=0,
msg_iov=[{iov_base="\1\30;3\0\0\0\0006\1\0\0\0\0\0\0\330\1b\2\0\0\0\0\0\0\0\0\0\0\0\0",
iov_len=4096}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 32
poll([{fd=3, events=POLLIN|POLLOUT}], 1, -1) = 1 ([{fd=3, revents=POLLOUT}])
writev(3, [{iov_base="(\0\4\0\236\5\240\0026\1\0\0\0\0\0\0",
iov_len=16}, {iov_base=NULL, iov_len=0}, {iov_base="", iov_len=0}], 3) = 16
poll([{fd=3, events=POLLIN}], 1, -1)= 1 ([{fd=3, revents=POLLIN}])
recvmsg(3, {msg_name=NULL, msg_namelen=0,
msg_iov=[{iov_base="\1\1<3\0\0\0\0\316\271*\1\230\1\351\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0",
iov_len=4096}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 32
poll([{fd=3, events=POLLIN|POLLOUT}], 1, -1) = 1 ([{fd=3, revents=POLLOUT}])
writev(3, [{iov_base="(\0\4\0\236\5\240\0026\1\0\0\0\0\0\0",
iov_len=16}, {iov_base=NULL, iov_len=0}, {iov_base="", iov_len=0}], 3) = 16
poll([{fd=3, events=POLLIN}], 1, -1)= 1 ([{fd=3, revents=POLLIN}])
recvmsg(3, {msg_name=NULL, msg_namelen=0,
msg_iov=[{iov_base="\1\1=3\0\0\0\0\316\271*\1\230\1\351\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0",
iov_len=4096}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 32
getpid()= 1811321
getpid()= 1811321
getpid()= 1811321
write(12, "W", 1) = 1
write(12, "W", 1) = 1
futex(0x1876290, FUTEX_UNLOCK_PI_PRIVATE) = 0
futex(0x1876290, FUTEX_UNLOCK_PI_PRIVATE) = 0
fcntl(16, F_SETLKW, {l_type=F_RDLCK, l_whence=SEEK_SET, l_start=360,
l_len=1}) = 0
pread64(16, "\0\0\0\0", 4, 44) = 4
pread64(16, "\214-\0\0", 4, 364)= 4
pread64(16, "`\32\0\0D\0\0\0.\0\0\0\4\0\0\0\30\260\r\23\231\31\1&", 24,
11660) = 24
pread64(16, "freedesktop\0button-pressed\0en_GB"..., 46, 11684) = 46
pread64(16, "\7\354\262_", 4, 11730)= 4
fcntl(16, F_SETLKW, {l_type=F_UNLCK, l_whence=SEEK_SET, l_start=360,
l_len=1}) = 0
recvmsg(3, {msg_namelen=0}, 0) = -1 EAGAIN (Resource
temporarily unavailable)
recvmsg(3, {msg_namelen=0}, 0) = -1 EAGAIN (Resource
temporarily unavailable)
poll([{fd=3, events=POLLIN}, {fd=4, events=POLLIN}, {fd=6,
events=POLLIN}], 3, 490) = 1 ([{fd=3, revents=POLLIN}])
recvmsg(3, {msg_name=NULL, msg_namelen=0,
msg_iov=[{iov_base="\5\1=3|\2569\0026\1\0\0\236\5\240\2\0\0\0\0\323\2K\1;\1b\0\0\1\1\0",
iov_len=4096}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 32
recvmsg(3, {msg_namelen=0}, 0) = -1 EAGAIN (Resource
temporarily unavailable)
recvmsg(3, {msg_namelen=0}, 0) = -1 EAGAIN
(Resou/usr/share/themes/Adwaita/gtk-2.0/assets/button-active.png/usr/share/themes/Adwaita/gtk-2.0/assets/button-active.pngrce
temporarily unavailable)
poll([{fd=3, events=POLLIN|POLLOUT}], 1, -1) = 1 ([{fd=3, revents=POLLOUT}])
writev(3, [{iov_base="$\0\1\0&\5\2\0006\1\0\0", iov_len=12},
{iov_base=NULL, iov_len=0}, {iov_base="", iov_len=0}], 3) = 12
poll([{fd=3, events=POLLIN}], 1, -1)= 1 ([{fd=3, revents=POLLIN}])
recvmsg(3, {msg_name=NULL, msg_namelen=0,
msg_iov=[{iov_base="\1\1?3\0\0\0\0006\1\0\0\316\271*\1\323\2K\1\323\2K\1\0\0\0\0\0\0\0\0",
iov_len=4096}], msg_iovlen=1, msg_controllen=0, msg_flags=0}, 0) = 32
poll([{fd=3, events=POLLIN|POLLOUT}], 1, -1) = 1 ([{fd=3, revents=POLLOUT}])
writev(3, [{iov_base="&\0\2\0\316\271*\1", iov_len=8}, {iov_base=NULL,
iov_len=0}, {iov_base="", iov_len=0}], 3) = 8
poll([{fd=3, events=POLLIN}], 1, -1)= 1 ([{fd=3, revents=POLLIN}])
recvmsg(3, {msg_name=NULL, msg_namelen=0,
msg_iov=[{iov_base="\1\1@3\0\0\0\0006\1\0\0\317\271*\1\323\2K\1?\1\202\0\0\0\0\0\0\0\0\0",
iov_len=4096}],
Re: Xlog and Hamlib-4
On 27/11/2020 02:40, Andy Stewart wrote: On 11/26/20 6:36 PM, Barry Jackson wrote: On 31/10/2020 02:55, Andy Stewart wrote: HI Barry, -- Let me know if you continue to have issues. Hi Andy, OK I finally found time to have another look at this. I started by rebuilding against a recent hamlib snapshot. Hi Barry, I need very specific details in order to attempt to reproduce this problem. * if the xlog version is not 2.0.19, please try again with that version, which is the latest * Exact version of hamlib (ideally, the link from where you downloaded it) * Exact type and version of Linux * If you did any custom configuration when you compiled hamlib and/or xlog, I'll need those instructions as well * I don't have a TS-405S, so I may not be able to reproduce the problem, but I will try. Thanks, and 73, Andy Hi Andy, I just realized that I still had not managed to send you the info you need. I have tried three times but every time the message is rejected as spam. I will try again now if I can remember what this is all about ;) The best way to give you all the info you need is to point you to my local build server repository where you will find: The full package build log: http://mtf.duckdns.org/pub/linux/barjac/distrib/cauldron/x86_64/log/xlog-2.0.19-2.1.mga8_bcj.src.rpm/build.0.20201126222709.log The versioned packages comprising the build chroot: http://mtf.duckdns.org/pub/linux/barjac/distrib/cauldron/x86_64/log/xlog-2.0.19-2.1.mga8_bcj.src.rpm/rpm_qa.0.20201126222709.log Hamlib snapshot git# and date are in the packagename which is listed in the packages list above, and was taken from the hamlib site as a source tarball. Other software like klog, wsjtx, jscall, fldigi are working fine built against it. I am building for the imminent release of Mageia (8 aka Cauldron where the next release is cooked up :) I will see if this mail sends OK. If not I will put it in a tarball and add it as an attachment and try again. Let me know if I can help further with maybe a back trace or strace output. Cheers, Barry
Re: Xlog and Hamlib-4
On 31/10/2020 02:55, Andy Stewart wrote: HI Barry, -- Let me know if you continue to have issues. Hi Andy, OK I finally found time to have another look at this. I started by rebuilding against a recent hamlib snapshot. I then was going to delete the .xlog, but instead looked carefully at the contents. My rig is TS-450S and I know that the new hamlib ID is 2003 In the xlog.cfg it had 2004! I changed the ID in the config file to 2003 and bingo, xlog ran without crash and hamlib works. Now when I go into the Hamlib settings dialogue it displays 2002! If I click on the "Select your Radio" button Bang! it crashes. [baz@jackodesktop .xlog]$ xlog Hash collision!!! Fatal error!! [baz@jackodesktop .xlog] I have seen that error in klog, before Jaime fixed it for hamlib-4 but I don't know what he did. It was related to getting the radio list from Hamlib, which looks like the same place. I Hope that helps, Cheers, Barry G4MKT
Re: Xlog and Hamlib-4
On 31/10/2020 02:55, Andy Stewart wrote: HI Barry, The following stack worked for me: * Andy's Ham Radio Linux, version 24c+ (my next dev release), which is based on Ubuntu 18.04.* * Xlog 2.0.19 * hamlib-4.0-rc2 * ICOM 7300 on /dev/ttyUSB0 * I started with a blank ~/.xlog directory. * I started Xlog and reconfigured hamlib, taking care to choose the updated rig number. * When the POWER entry is listed in the QSO entry window, it will say UNKNOWN if the rig power is set to 0, but otherwise, it reflected the rig power setting. Try ctrl-Y ctrl-K for a quick erase/update of the QSO entry window. I was able to duplicate the condition of hamlib hanging xlog, but that always happens when you turn off the rig with hamlib enabled and Xlog is still running. Xlog eventually comes out of it. This isn't ideal, but it is nothing new for Xlog, either. Turning the rig on again causes the software to be happy again. Let me know if you continue to have issues. 73, Andy Thanks Andy, Sorry for the delay, been busy with other stuff. The Hamlib daily snapshots are here: http://n0nb.users.sourceforge.net/ I have been using these to keep up to date with the latest fixes etc. in the hope that something would solve the problem which is also affecting klog. I will test new builds of xlog in a clean environment, and have noted the rig number changes. I will also play with rigctl, however wsjtx is working OK. Barry
Re: Xlog and Hamlib-4
On 30/10/2020 13:41, Andy Stewart wrote: I recommend moving your config file to a safe place and restarting xlog with no config file, as an experiment. Yes I did this effectively by using a new user and later by just re-naming the ~/.xlog dir. That allows the program to start without crashing, but would lock up when trying to enter settings in hamlib config dialog when using hamlib-4.0 I have not tested xlog with hamlib-4.0. I just built hamlib-4.1 and xlog against it and I can now edit the hamlib settings without a hang or crash, however rig control fails to work at all, even using a manual config string for the port. It just fails to communicate, looking at the hamlib tracker in xlog. wsjtx-2.3.0-rc1 and klog-1.4rc5 are working fine built with hamlib-4.1 so it does look like an xlog issue. I'm not familiar with Mageia. It is an Ubuntu derivative, and if so, from which version of Ubuntu are you seeing an issue? I've not tested anything past 18.04. No, Mageia has no connection to *buntus whatsoever. It was forked from Mandrake/Mandriva in 2010 but is community driven with no commercial ties :) RPM based like Fedora, SuSe etc. I maintain most of our radio related packages and am currently tying up loose ends for the next Mageia release (Mageia 8). I can only test Hamlib with my Kenwood TS450S, but please let me know how/if I can help to resolve this. 73, Barry G4MKT Thanks for the report, and 73, Andy Xlog maintainer
Xlog and Hamlib-4
Hello - first visit here! I am having issues with xlog segfaulting when launched on a system that has a config file that was created in an earlier version. I have narrowed this down (I think) to the fact that the new system has hamlib-4.0.0. I filed a bug report about this but have had no response as yet. https://savannah.nongnu.org/bugs/?59362 By chance I was just looking at this list recent archive and saw some discussion about hamlib and xlog, so here I am. I maintain (or try to) xlog in Mageia. Barry
