Good idea about the kstars log, my first thought was to be able to see what the low layer was doing so I went with Indi. I run the Indi stack from init.d so it's separate from kstars.?
This is my understanding of the wild slew problem: first failure happens with a pulse during a guide. This results in a slew aborted in the mount, and Indi not knowing that it was aborted. Subsequent slews issued from this bad state end up going heywire. I tested this last night and have seen the behavior before. I did many successful slews followed by one slew while guiding and it failed.?
Jamie Amendolagine