Thanks Blueshawk!
It looks like it is working now, I forgot to change my location in KStars.
For the moment I am very happy! I will use the log if I experience trouble...
Regards,
Albert

azwing,

Do you want me to describe my operational (Windows) system or the new system I am building at home?

You can have a look at my pictures, including the remote observatory at:
www.astrobin.com/users/Albert_van_Duin/
You will have to scroll down a bit for the observatory.

My new Linux system at home consists of an old i7 with 8GB RAM, Linux Mint, KStars 2.9.8.
The mount is an old Gemini G41, with OnStep drive V1.16e. Telescope is a 200mm F/2.8 ASA.
The camera is an old SBIG ST10XME, the focus motor is on OnStep (Maxi), or I can use Moonlite or OnFocus.
The roof is a garage door opener operated roll off roof, I can open and close it via Pythonscripts, so I hope to be able to use that with INDI too.
I hope to get it working soon.

Best regards,
Albert

Thanks a lot for doing this and let us know the status.

And thanks to Jasem, for this and everything else.

For me, the clouds rolled in early and I have not seen a clear night since the start of November.

Moreover, my astro laptop's solid state disk died, right when I was trying to change a few things in OnStep's INDI. Now, I am trying to recover the data from it, and it is taking weeks to read the data that is not corrupted.

So, I can't test anything until this recovery is over. Who knows how long it will take ...

The changes I wanted to do was add a button for 6 star align, and implement the tracking rates inside of OnStep rather than relying on inheriting from the generic driver (which does not seem to work). Specifically, I wanted to add the King Rate, together with Full Compensation Tracking. This is not urgent now that Howard has fixed all tracking accuracy issues for the STM32.

A tab would be great. Not essential but nicer.

Also a pull down list is nicer, but again not strictly required.

As an interim step, you can just add buttons for more stars. They don't need to be an exhaustive list. For example, we only need: Align 1, Align 2, Align3, Align 4, Align 6, Align 9, and that is it.

Later it can be refined to be a drop down list or whatever.

As the wisdom goes: do not let perfect stand in the way of good enough.

A button to write the alignment is a good thing, and easy to do.

As I said elsewhere, my astro laptop is out of commission for some time, so I would not be able to test things out for a while. That and the clouds ...

The clouds came. I just pushed something to azwing, with so far only the one change to the alignment: Namely that the kdialog isn't used for the align. (I'm using a mini pc outside (ECS Liva, though a Pi could be used instead, the Liva has GigE, and my pi doesn't) and... that it was popping up and freezing the whole system took me a bit to figure out. I'd gone out in the cold and actually reset the OnStep controller before figuring it out.)

So: Under the Align tab, here's how I see it as best, the popup I don't think should be used:
--Align--
Number of Align Stars: Dropdown: 1, 2, 3, 6, 9
Start Alignment: Button single
Alignment Status: Azwing's one currently on main labeled "Align Process"
Align Current sync: Button single

Instructions: Text box
(Which one, I'll leave to you, for now I'll leave mine as is.)

Write Alignment to EEPROM: Button Single (Might not be present until the results of the Align Process say complete.)
----
If stays cloudy (It's doing this really annoying thing where it's cloudy/not so i can't even take cloud flats.) I might do that tonight.

One other additional thing, I don't know why (Even after trying to look), but I've seen the issue with not connecting. I'm not completely sure of the cause, the reset script that kbahey (I think) posted on the OnStep list works. Not sure of the cause, but it appears only after I moved to CP 203?, Updated OnStep Alpha, Updated INDI, so of course it gives a great many possible problem locations.

@James,

I merged you code just now. Compilation Ok and ... Onstep Crash at startup ....

Ok a little bit more work now


I was thinking about a way to discover OnStep capability as for example your Outputs.
For the time beeing it is not possible and the flexibility of OnStep makes it not easier.
there i way for exapmpe to discover the presence of focusers :FA# / :fA#

IIf the one adding code to OnStep to support an "exotic" capability like for example Inputs/Outputs could provide as well a command to check this capability.

Ok was only a thought .... I have to debug now

After some checks it seems the problem comes from updates in indi itself because the official indi version does the same problem

Here is the info again, for those who did not see my message on the OnStep mailing list:

I wrote a Python3 script that sends one command to /dev/ttyUSB0, then exits.

Copy and paste the contents below to a file called /usr/local/bin/reset.py
Make it executable: chmod +x /usr/local/bin/reset.py

Install python3-serial if not already installed:
sudo apt-get install python3-serial

You can then run this script from the command line instead of cutecom.

You can take this further, to make it totally automated by modifying the Teensy's udev rule to run this command automatically when OnStep is inserted.

This goes into:
/etc/udev/rules.d/49-teensy.rules

And you need to add a RUN+="..."

Here are the contents of reset.py:

===
#!/usr/bin/env python3

import time
import serial

# configure the serial connections (the parameters differs on the device you are connecting to)
ser = serial.Serial(port='/dev/ttyUSB0', baudrate=9600)

ser.isOpen()

def serial_send(s):
string = s.encode('utf-8')
ser.write(string)
time.sleep(0.05)

serial_send(':GVN#')
ser.close()
===

If I may offer an opinion:

Let us not delay features until we get the Ultimate Solution. Things can always be refined later, and it is best if we have some features added with a suboptimal user interface, than waiting forever for the Ultimate Solution ...

But there is hope for doing what you want. Not only from the regular LX200 commands, but there are additional ones, e.g. :GXnn commands return a lot of info.

For example:

:GX09# -> number of aligned stars (easier than parsing :A?# output)

And so on, check Command.ino for more.

Other potentially useful ones :

:GX02# -> Polar align error in altitude
:GX03# -> Polar align error in azimuth
:GX95# -> Auto merdian flip?

Thanks Kbahey,

I still have trouble connecting with Ekos/KStars, both with an Onstep Maxi and Mini PCB, it works only sporadically.
Using the App, the OnStep works fine.

The python script gives an error, and I don't see the problem.
Python3 and Python3-serial are both installed and the script has been made executable.
Please see attachment.

Best regards,
Albert

@James,
I am still struggling with Indi since the merge of yesterdy.
One thing is sure it has nothing to do with an error in the version.
The crash happens with all versions but not on all machines

The faultly machine is an Intel Laptop
The Other machine is an AMD Desktop
Both run Debian Buster, same packages, libraries and kernel version so I woul expect to have the eeror on both, but No!

Symptoms:
When I run indiserver -vvv indi_lx200_OnStep (or run it from within Kstars)
the server crashes (contropl panel is opened but not initialized)
dmesg shows [17443.498606] indi_lx200_OnSt[15817]: segfault at 40 ip 00007feca554697e sp 00007ffcbb3981f8 error 4 in libc-2.27.so[7feca54d9000+146000]

After many time debugging it appear that the crash happens during loading of the xml config file .indi/LX200\ OnStep_config.xml

So effectively when removing the entry for the serial port
<newTextVector device='LX200 OnStep' name='DEVICE_PORT'>
<oneText name='PORT'>
/dev/ttyACM0
</oneText>
</newTextVector>
the server starts and runs fine as long I do not update the config file. I can shutdown and restart the server without crashes.
But as soon I update the file the problem is there again.

If somebody has an idea before I commit .........

I will continue to try to find what causes this behaviour but it is not easy to debug, the debugger crashes too when I arrive at the point of the error.

at least the weather is horrible so I have no frustration on this side

My fault, it's fixed in GIT.