If you go into the details of the lsusb command, you can extract info on each focuser which makes it unique, sometimes a serial nr. You can then use that to identify each focuser. But I must admit, I have never had to use that. Usually, vendor id and device id are enough to identify a device uniquely. The info you need should be in the tutorial.
I believe (but am not sure) that the combination of port mapping/binding and aliases solves your problem.

Hi!

Thanks for you advice.

But...

First, the persistent port mapping: It turns out my two Moonlight controllers (MyFocuserPro, actually, version 1, but running on the Moonlight protocol) has exactly the same serial header:

udevadm info -a -n /dev/ttyUSB0 | grep '{serial}' | head -n1
ATTRS{serial}=="0000000A0005"

and

udevadm info -a -n /dev/ttyUSB1 | grep '{serial}' | head -n1
ATTRS{serial}=="0000000A0005"

So how can I do a persistant port mapping now.....?

Secondly, I still do not see how this solves the Ekos challenge of Focuser module knowing which of them to use. All the cases I have seen here, only use one focuser, but someone must use two.....? (there was a discussion of N-camera setup in an earlier thread but as I saw it, no real solution)

Magnus

I'm not an expert with the scheduler. But I believe when creating jobs you can choose a specific equipment profile for it. So I would try creating separate profiles for scope 1 and scope 2 and assign them to their respective jobs in the queue.
You will still have to resolve the identity of your moonlight focusers. But one trick may involve using the Custom Drivers feature under Tools>Devices to create a uniquely named driver for the second scope. You should then be able to save its own configuration, such as unique serial port mapping for the custom INDI Driver.

I think it is doable. At least when not used at the same time.

Hi!

Thanks! That makes sense. The profiles then would only include one camera and one focuser, so matching focuser and camera would be solved Great, a step forward. I had not noticed this possibilty!

Remains then how to persistently match focusers to drivers - when their serial headers are identical... hrm... work around that?

Magnus

When you run lsusb. Do they also have the same Vender ID and Product ID?
I expect they do, but it's not a guarantee if they made revisions.

Hi!

Well, I was not thinking straight here....

The two focuser controllers: they have the same vendor ID and serial header, to the best of my knowledge. I can this far find nothing to persistently differentiate them.

However, this is only an issue if they are connected to the SAME R-Pi. When I connect them each to a different R-Pi, that is no longer an issue.

So now I have a functional setup (I think) for two scopes, two cameras and two focusers. Camera 2 + Focuser 2 connected to a separate R-Pi (R-Pi 2), drivers connect as remote device to Ekos running on R-Pi 1. I've created aliases so as to clearly differentiate the two Moonlites (now called SW_Moonlite and C8_Moonlite). I've created two profiles in Ekos on R-Pi 1, one for camera 1, focuser 1, guider and mount, and one for camera 2, focuser 2, guider and mount. Now I can choose which profile to use for a particular job in the Scheduler. This means that I now can set the scheduler to shift camera/scope in the middle of night, without me waking up. YES!

However, the only real reason I now have R-Pi, is to differentiate the two Moonlites (MyFocuserPro with Moonlite driver). It seems this far to work OK, but a bit "heavy" solution for differntiating them. So any idea on how to do it with only ONE R-Pi is still very welcome.

Final small issue: It was years since I started the indiserver manually. I guess I could insert a small command somewhere on R-Pi2 to just start the indiserver with the "C8 Moonlite", with indi_moonlite_driver, automatically on Boot and avoid the need to run Ekos and start it "manually". If anyone can advice me on that, I'd be even more happy than I am right now

Thanks for all assistance with this!

Magnus

With some focusers, when you purchase them you can sometimes choose a model which has a different USB id.

Like here for instance: www.firstlightoptics.com/astronomy-cable...ocus-controller.html

Note that there is the option at purchase to either buy "HitecAstro DC Focuser" or "Additional HitecAstro DC Focuser".

Quoting:
Existing Owners - Multiple Controllers
For existing Hitec DC Focus owners who wish to add a second controller to use alongside their existing one please select the 'Additional' option above.

This allows you to run a second controller in parallel with the existing controller on the same PC. In every other respect HitecDCFocus2 is exactly the same as the original.

Running two of exactly the same device from the same PC is not normally possible as the USB identifier would conflict and the PC would not know which one to talk to. HitecDCFocus2 contains a separate USB identifier so that they can be run side by side.

OK, maybe a stupid idea, but: If they are really absolutely identical in their USB identities, can you run a script that connects to them, and checks the current focus position? If you calibrate them to (very) different values, you can discriminate them by that, and then create a matching link moonlite_[12] -> ttyUSB[xy], and use that in the config?

Hi!

That is a creative idea. THat is actually how I differentiate them manually now. They are at very different number of ticks.

Unfortunately, such a script is a bit beyond my programming skills, though. Maybe I can persuade my daughter, who is an engineer, to look at it....

Magnus

The udev rule should be something likeJust add the proper values for vendor and product. Then the script gets the device name, can check it for the focus setting, and create a proper symlink.
The script itself is somewhat more involved, indeed, and would need the command reference of the focuser... but nothing an engineer can't solve.

I'm glad you came up with a workaround. My next suggestion was going to be a headless raspberry pi zero for a a remote indi server just for the focuser.

Hi again!

Bad news (at least news to me): this does not work. It seems that the "Profile" choice in the Scheduler is not per job.

I now have a list of jobs. I wanted the first one to use Profile 1 (camera 1 + focuser 1 + guidecam)
Second job should then use Profile 2 (camera 2 + focuser 2 + filter wheel + guidecam).

But the profile choice does not follow the job. Meaning that the camera/focuser combination does not shift.

I can set the camera in the Capture module sequence. But how on earth do I make sure the correct Focuser is chosen, if not through the Profile chooser in Scheduler?

(Feels counter intuitive that this feature does not follow the job but is global, when all the other settings there follow the job)

In the Scheduler sequence list, the Profile line is also before the Job section.

Does this make sense? Anyone recognize it?

Magnus