I just installed Ubuntu 15.04, indi-full and kstars-bleeding to try out KStars/Ekos with INDI in the hopes of being able to setup a remote imaging solution using a Raspberry Pi. I have an SBIG ST2000XCM and I was able to use Ekos to capture an image from the main CCD chip but when I try to capture an image from the guiding CCD the download never completes and sometimes the indi_sbig_ccd driver crashes with a SIGSEGV in INDI::CCD::getMinMax().

I turned on logging in the INDI server and I can see that the driver is reporting bad values for the width and height of the guide chip. The guide chip in my camera is 657 pixels wide and 495 pixels high with each pixel being 7.4 x 7.4 u. In one case the driver, via the debug messages and via the INDI control panel, reported the guide chip was 46962 by 20096.

This sounds a little like bug report #62 (Device SBIG CCD guider). In that report the user was able to override the initial invalid values in the INDI Control Panel. I tried that but as soon as I started the guide exposure the values reverted back to the original invalid values.

I have attached a debug log from my last attempt. The log does say that the download completed, but I got the "KStars detected indi_sbig_ccd crash" message the the Capture radio button stayed yellow on the Guide tab.

The INDI library shows v1.1.0 and code $Rev$ Protocol 1.7

Has anyone else seen this or any ideas how I might be able to circumvent this?

Thanks

Mark

The reason is that it detected "Extrenal tracking CCD" from the SBIG library but apparently there isn't, and when querying for data, it is successful but the values are bogus. I'll try to work something out, thanks for the detailed report.

Ok, I added a hack to invalidate such values, check the PPA update tomorrow and please report back.

Thanks for the quick response. I will try out the new INDI driver after work. The ST2000XCM does have an interface for an external guider, but I did not have one attached to it. That may explain why the camera reported an external guider, but since none was attached uninitialized (i.e random) values were passed back to the driver.

Thanks again and I will let you know how the testing goes.

Mark

I was able to test your patch and it worked. The proper values are being displayed in the Control Panel and the guide exposure download completes and is displayed in the FITS viewer.

Thanks !

Mark