Klaus H. replied to the topic 'Pegasus nyx testing' in the forum. 6 months ago

A minor observation I forgot to mention: After disconnect and reconnect, the speed setting in the GUI did show 900x (that was the speed set before), but the real speed setting was far slower, I suppose only 32x or 64x. Should this GUI field be automatically set to the right value after reconnect?

Kind regards, Klaus

Read More...

Klaus H. replied to the topic 'Pegasus nyx testing' in the forum. 6 months ago

I re-tried a second time, this time everything worked as expected, also immediately after reboot: reboot, disconnect, wait until beep sequence, reconnect, both motors work.
Starting with Windows Unity app and trying reboot there didn't show any issue either.
Updated to latest Unity and Firmware 1.25.2 now.

Kind regards,
Klaus

Read More...

Klaus H. replied to the topic 'Pegasus nyx testing' in the forum. 6 months ago

yes, I did disconnect and reconnect the INDI driver. After this reconnect both motors did not work.

Read More...

Klaus H. replied to the topic 'Pegasus nyx testing' in the forum. 6 months ago

Hello James,
not a problem, as I could recover with power cycling. Firmware version is 1.22. I think it was in siderial tracking mode, before the reboot I tried to switch between the four tracking modes, including King, but I think I switched back to Siderial before trying the reboot command.
Clear skies,
Klaus

Read More...

Klaus H. replied to the topic 'Pegasus nyx testing' in the forum. 6 months ago

Hello James,

I today tested the Pegasus NYX-101 with your latest driver changes (commit b24472523; compiled on ArchLinux via libindi-git AUR recipe). I tested the Elevation limit/horizon limits, flip pier side all worked as expected.
Regarding 'reboot' command I observed that the mount did reboot - disconnected, after some time the beep sequence - but after that both motors did not react anymore, no movement possible at all. Disconnecting and reconnecting the driver did not change anything. The Android app could connect, but also not trigger any motor movement. Mount was not parked, tried to unpark/park but also this was not possible (no click sound). Powering-cycling the NYX mount helped in recovering - after that mount reacted normal.

Kind regards,
Klaus

Read More...

Hello Wolfgang,
I added debug statements in GPhotoCCD::StopStreaming() before and after the liveViewThread.join(), it's called only once, after the 30 frames were taken - at the end of the recording.

To rule out that something is broken in my archlinux/manjaro installations (both of them, PC with AMD Ryzen 5700G, Laptop Lenovo T490s show the problem), I tried something else today:
I installed ubuntu-22.04.3-LTS in gnome-boxes, installed kstars/indi and latest EKOS as mentioned in the ubuntu installation instructions (apt add repository, install indi-full gsc kstars-bleeding), apt dist-upgrade to latest versions, rebooted in the virtual machine.
And then tried in the ubuntu virtual machine both first with the Canon EOS R7 (with GPhoto CCD driver) and then with the Canon EOS 450D (with Canon driver, reporting camera as ' Canon DSLR EOS 450D (PTP mode)) - both drivers crash ... at the end of the recording.
My observations: In the Streaming tab, switching on/off the live streaming works, also multiple times, no crash.
Starting recording with 'Record On' , or 'Record (Duration)', or 'Record (Frames)' buttons works - it records even for minutes without problems - but: as soon as the end of the recording is reached - either end of given duration, or number of frames reached, or I press 'Record off' - the driver crashes reproducibly.

Any hints on how to debug this further? I'm out of ideas for the moment.
Kind regards,
Klaus

Read More...

Hello Wolfgang,

I tried to attach gdb to the indi_gphoto_ccd driver directly now and captured gdb backtraces (just) from the driver crash.
To not garble the forum too much I put it as gitlab snippet here:
indi_gphoto_ccd driver crash with EOS R7 below kstars-git cfc274ad1

I also tried to run with the Canon DSLR driver (which did not work in the past for me), but same result:
indi_canon_ccd driver crash with EOS R7 below kstars-git cfc274ad1

Kind regards,
Klaus

Read More...

Hello Wolfgang,
the code block after 'The logfile looks like this' is the complete content of /home/klaus/.indi/logs/2023-10-08/indi_gphoto_ccd/indi_gphoto_ccd_19:24:33.log
And yes, /home/klaus/indi_2023-10-08/indi_record_2023-10-08@19-25-58.ser exists and is not empty: size 9216000 bytes, opening with siril shows that it contains exactly four frames.
Kind regards,
Klaus

Read More...

Thank you Wolfgang,

now I could add my post that also reproduces the DSLR problem under Linux , hopefully with enough debugging details to give a clue about what goes wrong.

Kind regards, Klaus

Read More...

Hello Ladislav,

I think this might be a general issue with the INDI GPhoto CCD driver, as I have tried to use both Canon EOS 450D and Canon EOS R7, running under both ManjaroLinux and ArchLinux with kstars 3.6.6, both with libindi-2.0.4 and libindi-gphoto 2.0.4.
Connecting the DSLRs works fine, also video preview work using the ' Streaming' 'Stream on' button; but a crash of the INDI GPhoto driver can be reproduced all the time on clicking the 'Record (Frames)' button. After showing a dialog box that the INDI GPhoto driver crashed and a countdown to zero also kstars crashes.

On kstars stdout I see two INDI GPhoto related messages:

  Dispatch command error(-1): INDI: Could not find property DRIVER_INFO in GPhoto CCD
  Dispatch command error(-1): INDI: Could not find property CCD_EXPOSURE in GPhoto CCD

I have re-built kstars from git with debug infos using the AUR (aur.archlinux.org) kstars-git, libindi-git and libindi-gphoto-git recipes:
* kstars 3.6.8 beta (= kstars-git 1:cfc274ad1-1)
* libindi-git v2.0.4.7.gb38785840-1
* libindi-gphoto-git v2.0.4.7.g4056c204-1

Re-runnind under gdb gave me this stacktrace (which occurs as soon as I clicked on 'Record (Frames)' in the GPhoto CCD 'Streaming' tab:
Thread 1 "kstars" received signal SIGSEGV, Segmentation fault.
0x0000000000000000 in ?? ()
(gdb) bt
#0  0x0000000000000000 in  ()
#1  0x0000555555bac6ea in ISD::ConcreteDevice::getNumber(QString const&) const
	(this=this@entry=0x55555a240e80, name=...)
	at /usr/src/debug/kstars-git/kstars/kstars/indi/indiconcretedevice.cpp:107
#2  0x0000555555bc2c7c in ISD::CameraChip::isCapturing() (this=this@entry=0x555558abd3b0)
	at /usr/src/debug/kstars-git/kstars/kstars/indi/indicamerachip.cpp:532
#3  0x0000555555d2d570 in Ekos::Capture::refreshCameraSettings() (this=0x555558fafb50)
	at /usr/src/debug/kstars-git/kstars/kstars/ekos/capture/capture.cpp:708
#4  0x00007ffff4ed0b37 in  () at /usr/lib/libQt5Core.so.5
#5  0x00007ffff4ed2db7 in  () at /usr/lib/libQt5Core.so.5
#6  0x00007ffff4ec35ee in QObject::event(QEvent*) () at /usr/lib/libQt5Core.so.5
#7  0x00007ffff5b7893f in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /usr/lib/libQt5Widgets.so.5
#8  0x00007ffff4e9bbf8 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () at /usr/lib/libQt5Core.so.5
#9  0x00007ffff4eea26b in QTimerInfoList::activateTimers() () at /usr/lib/libQt5Core.so.5
#10 0x00007ffff4eea85a in  () at /usr/lib/libQt5Core.so.5
#11 0x00007ffff3cb2f19 in g_main_dispatch (context=0x7fffe8000ec0) at ../glib/glib/gmain.c:3476
#12 0x00007ffff3d112b7 in g_main_context_dispatch_unlocked (context=0x7fffe8000ec0) at ../glib/glib/gmain.c:4284
#13 g_main_context_iterate_unlocked.isra.0
	(context=context@entry=0x7fffe8000ec0, block=block@entry=1, dispatch=dispatch@entry=1, self=<optimized out>)
	at ../glib/glib/gmain.c:4349
#14 0x00007ffff3cb1112 in g_main_context_iteration (context=0x7fffe8000ec0, may_block=1) at ../glib/glib/gmain.c:4414
#15 0x00007ffff4eeaa1c in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) ()
	at /usr/lib/libQt5Core.so.5
#16 0x00007ffff4e9a904 in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib/libQt5Core.so.5
#17 0x00007ffff4e9bda3 in QCoreApplication::exec() () at /usr/lib/libQt5Core.so.5
#18 0x0000555555741845 in main(int, char**) (argc=<optimized out>, argv=<optimized out>)
	at /usr/src/debug/kstars-git/kstars/kstars/main.cpp:386
(gdb) 

The logfile looks like this:
INFO    0.002644 sec    : Session log file /home/klaus/.indi/logs/2023-10-08/indi_gphoto_ccd/indi_gphoto_ccd_19:24:33.log
INFO    0.740674 sec    : GPhoto CCD is online.
INFO    0.740713 sec    : Detected Canon.Inc Model Canon EOS R7.
INFO    1.022860 sec    : World Coordinate System is enabled.
INFO    1.023798 sec    : Session log file /home/klaus/.indi/logs/2023-10-08/indi_gphoto_ccd/indi_gphoto_ccd_19:24:33.log
INFO    1.023831 sec    : Upload settings set to client only.
INFO    1.025588 sec    : Capture target set to Internal RAM
INFO    1.025601 sec    : Images downloaded from camera will saved in the camera internal storage.
INFO    1.025619 sec    : Force BULB is enabled. All expsures shall be captured in BULB mode except for subsecond captures.
INFO    68.592097 sec   : Starting the video stream with target exposure 0.100000 s (Max theoritical FPS 10)
INFO    85.002103 sec   : Starting video record (Frame count): 5.
INFO    85.002200 sec   : Record file is /home/klaus/indi_2023-10-08/indi_record_2023-10-08@19-25-58.ser
INFO    85.178698 sec   : Waiting for all buffered frames to be recorded

Maybe that helps in tracking down the problem?

Kind regards,
Klaus

Read More...

Dear Indilib.org moderators!

Last weekend I read a problem report in the indilib.org forum from a user that tells that his DSLR camera no long works with the latest indi 2.0.4 driver.
As I also observed that my Canon EOS 450D and EOS R7 DSLRs no longer work with the libindi gphoto driver, I wanted to attach this feedback including stack trace to the post ... this was five days ago.
After I tried to post my reply, realizing that the reply did not show up, I found the info thread in the 'New users' subforum that the first post is moderated.

As I posted the above reply 5d 4h ago, I´d like to know if you did get that post at all, or maybe it was not stored at all as it was only a reply and not an 'initial/new post'?

Kind regards,
Klaus

P.S. Please forgive me if I'm just too impatient and you wanted to acknowledge my first post this weekend :-)

Read More...