Thanks, that worked. One strange thing I'm running into is that the FITSviewer won't debayer the image. The FITS field BAYERPAT is not set, but there's obviously a bayer pattern there. I figure I can correct this manually by adding the field value, but will that work? The files are a bit smaller than they were before the upgrade (they were 60MB, now they are 50)
There's FITS, RAW ... type Light. There's nothing that says mono or color like I've seen for other cameras. Compression is set to none, and in Image Info the bayer offset is 0,0 and the filter field is empty.
Understood, so additional info: My version of indi-full is still at 1.9.7 and refuses to go to 1.9.8, I don't know if this is an issue.
While this is being investigated - and I'll help however I can - what can I do to my FITS header so that say SIRIL and FITSviewer detect the bayer? It appears they depend on the property. Do I just need to set the BAYERPAT property in the FITS header? I know that it is RGGB.
This is resolved. I don't know how it happened exactly, but basically when the camera would crash I tried a bunch of things including deleting the config files.
I found a couple of debian packages to roll bakc to for indi-qhy and used those. They didn't work perfectly but they worked well enough for me to achieve what I needed (now the question is, am I missing information in my images? The images I took before all of this were 60mb, now they're 52 ... maybe I'm just overthinking it).
So, for whatever reason those drivers did not auto-detect/write the RGGB filter field into the configuration.
I deleted the configuration, rebuilt it, and now the filter was in there automatically. It should probably be called 'bayer pattern', not 'filter' under the bayer property.
My images can be fixed by running astfits from gnuastro, so I'm very happy about that.
This was a weird misadventure but I learned a lot.
It would really be good to have a repository available that makes things available for rollbacks though.