rafitadiaz opened a new issue, #18511: URL: https://github.com/apache/druid/issues/18511
**Description** When attempting to configure the GCE autoscaler, we encountered a critical exception that prevents the autoscaler from functioning. The issue occurs when the `druid.indexer.autoscale.workerVersion` parameter is not explicitly set in the configuration. The documentation indicates this parameter is **optional**, so its absence should not lead to a failure. Our current configuration for the autoscaler was: ```ini # Required for autoscaling of middle managers druid.indexer.autoscale.strategy=gce druid.indexer.autoscale.doAutoscale=true druid.indexer.autoscale.strategy.type=pendingTaskBased #How often to check whether or not new Middle Managers should be added. druid.indexer.autoscale.provisionPeriod=PT30S # How long can a worker be idle (not a run task) before it can be considered for termination. druid.indexer.autoscale.workerIdleTimeout=PT20M # How often to check when MiddleManagers should be removed. druid.indexer.autoscale.terminatePeriod=PT10M # How long the Overlord will wait around for a MiddleManager to show up before giving up. druid.indexer.autoscale.maxScalingDuration=PT10M #How long a task can be in "pending" state before the Overlord tries to scale up. druid.indexer.autoscale.pendingTaskTimeout=PT30S #druid.indexer.autoscale.workerVersion=0 #If set, will only create nodes of set version during autoscaling. Overrides dynamic configuration. # The port that MiddleManagers will run on druid.indexer.autoscale.workerPort=8091 # An estimation of the number of task slots available for each worker launched by the auto scaler when there are no workers running. Match the value to equal to druid.worker.capacity(CPU -1) druid.indexer.autoscale.workerCapacityHint=1 ``` **Actual Behavior** The service fails with the following stack trace: ``` 2025-09-10T13:29:58,949 ERROR [SimpleResourceManagement-manager--0] org.apache.druid.indexing.overlord.autoscaling.AbstractWorkerProvisioningStrategy - Uncaught exception. org.apache.druid.java.util.common.ISE: No minVersion found! It should be set in your runtime properties or configuration database. ``` We solved the problem adding this line to the configuration ```ini druid.indexer.autoscale.workerVersion="30.0.1" ``` **Expected Behavior** The autoscaler should use a default value (e.g., null) for workerVersion as indicated by the documentation, and the process should not throw an exception. **Notes** 1.- The field is showed as optional on the configuration: * druid.indexer.autoscale.workerVersion: **If set,** will only create nodes of set version during autoscaling. Overrides dynamic configuration. (default null) 2.- The error message and the exception are coming from here: https://github.com/apache/druid/blob/45cf67d09e10aaf493556d2fe512f2034eead897/indexing-service/src/main/java/org/apache/druid/indexing/overlord/autoscaling/ProvisioningUtil.java#L39 **Requested Changes** * **Set a Default Value**: The `druid.indexer.autoscale.workerVersion` property should default to null. This will prevent the ISE exception from being thrown when the parameter is not explicitly configured. * If this is not fix then the **error message should be improved**. The error message `No minVersion found!` is misleading. The code and documentation refer to `workerVersion`, not `minVersion`. It should be updated to clearly state which parameter is missing, such as `No workerVersion found!`, to avoid confusion with `druid.indexer.runner.minWorkerVersion` and other related properties. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
