6. Conditional Trace - High Availability

If a workflow loses contact with the SC, it tries to reconnect as long as the workflow keeps running. If the entire cluster goes down, all trace filters running on the workflow are stopped and you are required to run the trace filter again. 

As long as one SC is running, the trace data continues to be distributed. However, the data already sent to an SC which fails, is lost. If the Platform is down, the SCs continue to run. The SCs cannot be reached from REST, but new trace data generated from the trace filters that are running still reach the cluster. The data generated is available once again when the Platform is up and running again.