Configuration | Camunda 8 Docs
operate env variables-
camunda.operate.elasticsearch.numberOfShards
camunda.operate.elasticsearch.numberOfReplicas
Where are these variables used.
We currently see some considerable delays in operate in display of latest process instances. Can we use above variables.
Please advise.
Regards,
Shilpa Chalke
Hi @ShilpaN.Chalke - how long of a delay are you seeing? It is important to understand the architecture of Zeebe, which exports data to a long-term cold-store (by default, Elastic) for Operate and other components to read from. There will always be a small delay. There are several factors that contribute to the length of the delay, many of which can be configured within Zeebe.
I would recommend starting with these resources:
1 Like
We are seeing delays in displaying data which is more than 4 hrs and it started happening recently. We are processing 10k load through the day since last 6 months. We started seeing this error from the month of December.
I also see import error for incidents in operate.
AbstractIncidentPostImportAction : Exception occurred when performing post import for partition 1: Process instance is not yet imported for incident processing. Incident id: xxx, process instance id: yyy.
@ShilpaN.Chalke - this something I would recommend opening a support ticket for. As mentioned before, there are many different elements that all come together that might affect the exporter: Zeebe resources usage, Elastic resource usage, networking, configuration, I/O, and more. Without knowing a lot more about your deployment it’s hard to provide much more help.
Generally speaking, if you already looked at the available configuration, I would next look at resource usage in the Zeebe and Elastic pods and see if you can correlate anything. Also review the logs and events for both, to see if there’s anything causing problems in the cluster.