yesterday I was working with watcher, ran a simulation on a watcher config with aggrs (trying to debug it) and I started getting errors. I checked the elasticsearch.log and the logstash-plain.log and started seeing the rejecting mapping update to [index]. now I understand going from 5.x to 6.5 (in my case) you can not have more than 1 type. this system was working with no issues till ran the simulation on watcher. now even the internal indexes like .monitoring and .kibana are throwing errors.
this system has been up and working since October. I tried restarting the system (single cluster) still no go. any thoughts?