Camunda 8 architectural queries

  1. Where are the process history and task history stored which we see in Camunda Operate?

  2. Suppose we have a zeebe cluster and processing a workflow. The zeebe instance which was processing the workflow gets down due to some reason. What will happen to the worfklow? If the processing of that workflow is still completed, how does that work internally?

  3. " To ensure fault tolerance, Zeebe replicates data across servers using the [raft protocol]." What kind of data is being replicated? Does zeebe broker store any kind of data? If it does, where does it store the data?

Hi @sayan_b19 - Operate pulls its data from Elasticsearch, where Zeebe exports data to. Internally, Zeebe has an internal data store (using RocksDB), where the data remains until it is exported. The data is divided into partitions, and the brokers use a replication protocol to share the data, so if one goes down, the data should be replicated internally within the cluster.

Thank you so much for the clarification!

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.