I received the above error on one of my machine learning jobs and I'm looking into some insight into what exactly happened. There was a data interruption in the creation of the new daily index the job is analyzing, but a similar job that uses the same index had no errors
It's possible that the two ML jobs were running on different schedules. So, the first job could encounter the error when there was the data interruption and the other job happened to run when the interruption cleared up?