Dear Helge,

The hidden files is not my problem. The hidden .indi folder is there, but it contains only the profiles.db.

The other problem is solved. I installed yesterday evening the KStars 2.8.4 on my windows machine and reloaded the latest daily indi release. Remote solving is now working. I have indeed Astronomy under Aux 1 in the profile.

So my problem remains: why is non of the drivers writing its log. The logs folder and the files are not even created. The chmod value of .indi is 775.

Regards, Bert

Well, not to state the obvious, but did you enable logging for the drivers in INDI Control Panel and selected to File?

Hi Jasem,

I noticed the absence of log files a few nights ago when working with the real stuff. But due to bad weather I'm using the simulators (remote on the Pi) to reproduce the problem. To keep things simple I only use three drivers Telescope Simulator, CCD Simulator and Astrometry.

In all three drivers I have checked Debug Enabled (without it the Logging Levels are not shown) and Logging Levels set to Errors, Warnings, Messages and Driver Debug. In the telescope also Scope Verbose.

Log output is set to "To Log File". in all three drivers, but the .indi map on the Pi stays empty but for the profiles.db, so no logs folder nor log files.

Regards, Bert

After getting a new micro SD card and loading it with my Raspberry Jessie image, I suddenly had logs!
I saw that on the pi the log is created in /home/pi/.indi, but when I switch to the sd card with Ubuntu Mate the logs are created in /root/.indi, while my root is /home/bert and this must be the reason that the logs are not created.

The question is now, where (in which file) is this root defined?

I hope that with this explanation an answer will come up.

Regards, Bert

I think you ran INDI Web Manager as root and this is why it was writing to to /root/.indi and not $HOME/.indi

This is why it is recomended NOT to run the Web Manager as root.

Same problem here with two QHYCCD cameras. But they arent working together undependant of the driver settings. If i choose one driver for both...nope.
If i choose qhyccd for CCD an qhyccd for Guider, only one of the cameras operates (if i am lucky).
Seems that EKOS indi is not in a stable condition. Hope that these problems dont occur when you are operating your observatory remote and are a few (hundred) miles
away, which is one of the main reasons for indi.
I havent tried to use a second rpi3 and attach the second camera to that device. Could be a cheap workaround until the driver problems are solved.
Robert

The QHY issue is a problem with the QHY SDK and not an issue with INDI or Ekos. If QHY do not fix their SDK to enable simultaneous operations of multiple QHY cameras (which is possible without issues in other cameras), then we can't do much. AFAIK, QHY is working on this, but contact them for specific details.

I'm a Linux newbie and got confused by the fact that there seem to be two roots. A linux root and a user root. So when knro wrote: "This is why it is recommended NOT to run the Web Manager as root.", I uninstalled indiweb with sudo and reinstalled it for user, and ended up nowhere.

Then I found this item . So I reversed my actions again and reinstalled as root. Problem remained that the .service scripts were nowhere to be found. So I eventually downloaded the .zip from GitHub, changed the user in indiwebmanager-pi.service to my username, RENAMED the file to indiwebmanager.service before copying it to the /etc/systemd/system folder. Finally I got my logs.

This may all be very clear to Linux savvies, but may I ask to have a little consideration with Linux newbies? But, after all, I'm glad its working now.

Regards, Bert

Yeah I will update this so that there is only one service file.. INDI Web shouldn't run as root.

just wanted to add some of my things from my last imaging night
- while doing flats with a wall mounted box my eqmod controlled mount was enabling guiding (I could understand dithering but not guiding while doing flats on a wall position)
- after doing flats the ekos interface was switched to lights but when capturing a frame it actually tried to run a flat frame (going to wall position) - I guess that it somehow did not correctly cancel a strange 2000s exposure in indi and failed to sync the switch from flat to light to indi (after stopping and resetting everything it was working as expected) I probably have to reproduce that to give more helpful feedback
- while imaging and using the internal guider my rpi lost connection from ekos to indi twice (kstars running via vnc and ekos doing everything locally) The second disconnect happened while switching the view in kstars. (probably kstars was busy and delayed ekos too much)

I need to remember to create logs by default.

I know I probably should not jump into this thread, but, it's long overdue for me to say my piece concerning QHY and Linux. For many years, QHY and I have been very close friends, and during those years, I have suggested and pushed for QHY to develop Linux drivers for the various cameras which they created. This has always met with direct opposition because QHY believes that Linux represents a small portion of the user market.

Because of this lack of support from the top, this has always been a luke warm or volunteer approach. For those wishing to use QHY cameras on Linux, I offer this direct quote: "Hi Dan,
we will update the INDI driver ASAP after finishing our current Macos support, we are working on.
Yes, there probably is issue related to run multiple cameras with the current driver/indi, so it is in my
TODO work on that.

Please note, this was developed by another programmer, so we have to rework that.
Sorry for that problems, "

So, I will say this regarding INDI with QHY cameras. Users should use multiple softwares in order to guarantee use of separate QHY drivers. For example: Run KStars with a QHY primary camera using INDI. In a totally separate desktop, run Lin_guider to control a QHY or Orion Guide camera which uses the CCCD firmware driver. You will find this separation works, although total KStars is not achieved.

INDI is a slow mechanism at best when controlling multiple simultaneous tasks. When operated on machinery with limited capacity and multiple USB devices, care should be taken to isolate critical USB pathways in order to prevent traffic collisions. And while I must admit, I have not examined the exact data buss of a Pi2 or Pi3 with regard to serial communications, I have found on the ODroid, that one of the USB serial ports is being shared with the LAN interface. Such design compromises must be considered when making wiring connections.

Dan

Yeah logs would be helpful, check my signature for how to.