Just thought about something, in the Ekos VM which I'm using, just done an apt-get update/upgrade (do that on a daily basis) and noticed that the indi-eqmod was updated.

However on the Odroid running debian, I use the deb files from the download section and that file doesn't get any new updates.
I believe (from memory as I can't access my Odroid from here), it's version 0.2 only.


edit:
Shall I return to Ubuntu for my Odroid, so that I can get all the latest updates for the Indi packages automatically as they become available?

On my return I will wipe the eMMc and install the official Odroid Ubuntu image instead of that stripped down Debian.
I'm a Debian guy....

I will retest the all thing and report of any issues.

I thought about using the EqMod simulator to demonstrate what I'm seeing on my side.
I've attached a screen recording (using the same tool as you).



EDIT:
33b7583a5f908e61a232-6b8a975948e4f6a1f52...n.com/Ekos-eqmod.mkv

Hi,
If you use the simulator, this behavior is normal as the simulator is started/initialized each time you connect the mount. When using a real mount, the driver tests if the motors have already been initialized when you connect the mount. There should be a message announcing this status, maybe in debug mode, but I think I always display how the driver considers the initial position. The only reason you may find your mount uninitialized is when you power it down. If you want to remember the position between power-ups of the mount, you should use Park/Unpark.
I had a look to your video and you actually use the indi_simulator_telescope, not he simulator part of the indi_eqmod driver, what I said before only concerns the eqmod driver.

Jean-Luc.

Thanks, that explains it for the simulator.



ah I see, I can't remember powering down the scope at all as in turning off the power.
However, now that I think of it, when I kill the indiserver on the Odroid, I can hear the motors being disengaged, as in: you can turn them with your fingers.
When you start the indiserver, you can clearly hear a "clonk" sound, the stepper motors are "hot" and you cannot turn them with your fingers (the pulley).
Maybe this is an issue with the AstroEQ box I'm using to control the EQ3-2 stepper motors.
If it is, then I need to figure out what to do as this is quite a problem for my observatory.


two different things yes, just thought it would work , obviously not.

Had a look to AstroEQ source code, and it seems the only place where motors are disabled (and reinitialized) is when the ATMega firmware recieves the synta 'L' command (instant stop motor). Quickly searched through the indi_eqmod driver code but did not find a place where the driver uses that command (I have thought to the Abort button, but even there, it is a normal stop command). You may look at the serial debug in indi_eqmod driver to check if and when this 'L' command is sent. Otherwise It seems that AstroEQ uses the same status bit for motor initialization.

I've done more testing again today, it works OK as long as the indiserver is still running on the remote server.
As soon as the process is killed, on restarting everything, the scope will not be at the last known location.

I will post a debug log as soon as I can.

I have now everything updated to the latest code, in the Ekos vm and on my Odroid running Ubuntu 14.10 with indi stuff compiled from source (updated to today's code).
The scope location is lost on restart of the indiserver on the Odroid, every time.

I know I can and should do a park, however I'm testing a power cut situation.
My observatory will be located about 400 miles away from my house.... and that will be a very expensive trip just to go and reset the mount position.
If there a way I can do something like:
query the RA/DEC location in real time, dump that into a file
use that file on restart of Ekos to re-inject the value into eqmod in Ekos?

There are 2 situations here: if you only restart the driver/indiserver without powering off the mount, the scope position should not be lost. Is this the case ?
If you power off the mount, encoder positions are lost and RA/DEC are reset to home when reconnecting the driver. And, as you suggest, logging RA/DEC with an indi client seems to be the only simple solution.

If I restart the indiserver on the Odroid, the AstroEQ box cut off the power to the motors.
You can clearly hear it, and you can move the motors with your fingers.


That's I can understand why.


Is this a feature that can be added to the eqmod code?
I guess at the moment, I can't "re-inject" the RA/DEC position into eqmod.

I've contacted Tom, the creator of the AstroEQ box and he replied to my email with the following:

If your Odroid and your mount are on the same power line, it would be better to perform that stuff remotely with a client. Actually you may reinject the encoder values using a Park file, it could be created by hand in this particular situation, it's a simple xml file.