I got 6 of 20 images last night, 0 of 20 this night. Of the images last night all was well centered but a couple of them had heavy trailed stars.
When you now have the plate solve function in place don't you have logs from it where you can get all information about which jobs have failed and the offsets of all the images? For example I have noticed that images with the OIII-filter often have problem to get through. Perhaps the PS have problem to recognize the stars on OIII-images? This night there seems to be long periods where the same OIII-images was tried time after time.
There was also a question about the java version of the RTM. I use Internet Explorer and it seems that only the java-version works on that platform, so I hope you can have the java version left.