OK - I have an update. When I type "launchctl list | grep indi", nothing is returned, so it appears there are no hung jobs. However, when I run the indi server, I get the same behavior as before. I briefly get the following message: “Failed to remove server Job!” with a red light. This within 1 second updates to “Server is running (idle)” with a green light. Maybe this is a red herring. After starting the server and the control panel, and typing:

riprock99$ launchctl list | grep indi
407 - eu.cloudmakers.indi.INDI-Server.55296
409 - eu.cloudmakers.indi.indiserver

If I then add the MT focuser, I get:

riprock99$ launchctl list | grep indi
385 - eu.cloudmakers.indi.INDI-Server.55296
388 - 0x7faa42434610.anonymous.indi_microtouch
387 - eu.cloudmakers.indi.indiserver

So far so good. When I go to the control panel, and try to connect, I get the same error as before (see attached picture).



Note that sometimes my computer shows up twice in the Control Panel; for unknown reasons, I got three this time. After 2 minutes or so, I also got this weird message:

Connected to Chaos-Manor-2.95321262.members.btmm.icloud.com.:7624
Can't connected to Chaos-Manor-2.95321262.members.btmm.icloud.com.:7624

However, if I quit the control panel and the server, and start the Craig Stark focus control, it works fine. This is all after a cold power up of all equipment in the dome.

Lastly, to verify that I have the right port, with Craig Stark focus controller running, I started the INDI server, and the control panel, removed the focuser, then re-added it. When I tried to open it, I got: "Error retrieving data from Microtouch, please ensure Microtouch controller is powered and the port is correct."

OK - and even stranger things just happened. For the first time in a week I got the INDI control panel (ICP) to connect to the focuser. To do so, I gave it the wrong port, but ICP connected anyway. All of the values for ticks and temperature were crazy (like a temperature of 1304°C), and it wouldn't move anywhere. When I disconnected, ICP would not close the MFT. Finally, after quiting ICP, I couldn't get the behavior to repeat.

Hi Scott, I don't know what, but something is wrong Does it also happen if you try previous INDI Server version? Peter

I was running 2.2.27 (1.2.0). Fortunately I still had 2.2.26 as a .dmg. I fired it up, and the behavior was identical. At least it can be reproduced! I am sitting in the dome now, and you just replied in near real time. Would it help for me to call you? Or arrange a login or something?

Sincerely,

Scott

Please keep me updated when you find something I can help with. Good Luck.

Marco

It is probably too late here today to start some debugging session, but if you can arrange remote access to your machine, we can do it tomorrow. Peter

Hi guys,

I tried updating to OS X 10.11.6, but it didn't help the focuser problem. AstroDSLR now rocks though! Wish I understood why I can't seem to get to the port with the focuser on it anymore. Craig Starks works when I don't have the focuser added to INDI server. As soon as I add it to the server, Starks version won't connect anymore. But no matter what, the control panel says the resource is busy, check to see if it is powered up, etc. Frustrating!

Also, bad news, I have a conflict, and won't be able to trouble shoot tomorrow (Sunday) afterall. Sorry guys!

Sincerely,

Scott

Hi Scott,

I looked at the screenshot again in the morning and I suspect, that you really use a wrong port. What is the device name? While on linux it is probably something like /dev/tty...., on OSX it should be something like /dev/cu..... otherwise you may get a message like this (tty... has modem control, cu... is naked device). The device name may be different if you connect the focuser to a different machine or hub USB port.

Peter

Hi Peter,

I tried using every /dev/cu* device I have, multiple times, before and after reboots. I think the one I have is correct because when I start the INDI server with the USB port that I think is the right one, the Craig Stark software, which successfully controls the focuser, no longer connects.

Let me know if there is some other trouble shooting path I might take.

Scott

Scott, just to be sure - the best way how to find the correct port name is disconnect and connect focuser and each time execute

ls /dev/cu*

The difference is the correct name

Marco, any other idea what can be wrong? I checked your source and it seems, that it actually connected successfully and failed on ack(), am I right? Strange is, that it doesn't show error message in WriteCmd(). Peter

Peter, I think I know whats going on. I had a signed/unsigned mismatch in the code before PR #48. Position is returned as an unsigned short int (16bits). If the focuser is sitting at a position higher than 32.767 the negative sign is returned and Ack() failes. This is corrected in PR #48. Sorry.

Scott, please try and move the unconnected focuser by hand to a position below 32767 and try to connect. Let me know if this works.

Marco

Yes, I see there is another change 3 days ago...

Scott, please can you try this build? Peter

download.cloudmakers.eu/INDI_Server_2.2.29.dmg