-
Notifications
You must be signed in to change notification settings - Fork 693
Closed
Task
Copy link
Labels
area/backendcomponent/data-layerRelated to Data Layer TeamRelated to Data Layer Teamcomponent/operateRelated to the Operate component/teamRelated to the Operate component/teamkind/feature-requestRequest for a feature that is not being worked on yetRequest for a feature that is not being worked on yetsupportMarks an issue as related to a customer support requestMarks an issue as related to a customer support requestversion:8.6.25version:8.7.11
Description
Describe the user problem that you are trying to solve
Operate does not support to reconfigure the number of replicas for a shard if the configuration changes. It also doesn't support changing the number of shards for a template (so that it applies to indexes created in the future). The docs point out this limitation: https://docs.camunda.io/docs/self-managed/operate-deployment/operate-configuration/#settings-for-shards-and-replicas
Our use case requires that we are able to scale up elasticsearch and also change the number of replicas to increase the availability and usage of the nodes.
Describe the solution you'd like to see implemented
- When I restart Operate with a changed configuration of
number_of_replicas
- The it is applied to both, existing templates and existing indexes on startup (for indexes, also taking the
numberOfReplicasForIndices
setting into account)
- The it is applied to both, existing templates and existing indexes on startup (for indexes, also taking the
- When I restart Operate with a changed configuration of
number_of_shards
- Then it is applies to existing templates
- It is not applied to existing indexes (as that would require splitting or shrinking the index, a fairly heavy operation); this limitation is documented
Optimize already has a way to do this.
This then allows us to increase the the elasticsearch pods and also the data replication.
Describe alternatives you've considered
Additional Context
- Context here in #tmp-ha-es -> https://camunda.slack.com/archives/C05P0D3AER2/p1694424927422869
- Blocking migration of existing clusters in PDP ticket https://github.com/camunda/product-hub/issues/667
- https://jira.camunda.com/browse/SUPPORT-23300
Definition of Ready - Checklist
- The issue is not duplicated - the is no other issue about the same user problem in Operate/Tasklist/PDP repository
- The issue has a meaningful title, description and clear user problem
engineering-issue-sync-app
Metadata
Metadata
Assignees
Labels
area/backendcomponent/data-layerRelated to Data Layer TeamRelated to Data Layer Teamcomponent/operateRelated to the Operate component/teamRelated to the Operate component/teamkind/feature-requestRequest for a feature that is not being worked on yetRequest for a feature that is not being worked on yetsupportMarks an issue as related to a customer support requestMarks an issue as related to a customer support requestversion:8.6.25version:8.7.11