Briefly, this error occurs when Elasticsearch is trying to start a datafeed but can’t because some indices are currently being upgraded. This means that the system is in the process of updating the indices, and the datafeed can’t be started until this process is complete. To resolve this issue, you can either wait for the upgrade process to finish or pause the upgrade, start the datafeed, and then resume the upgrade. However, it’s generally recommended to let the upgrade finish to avoid potential data inconsistencies.
This guide will help you check for common problems that cause the log ” Could not start datafeed [” + datafeedId +”] as indices are being upgraded ” to appear. To understand the issues related to this log, read the explanation below about the following Elasticsearch concepts: plugin, indices.
Log Context
Log “Could not start datafeed [” + datafeedId +”] as indices are being upgraded” class name is DatafeedNodeSelector.java. We extracted the following from Elasticsearch source code for those seeking an in-depth context :
public void checkDatafeedTaskCanBeCreated() {
if (MlMetadata.getMlMetadata(clusterState).isUpgradeMode()) {
String msg = "Unable to start datafeed [" + datafeedId +"] explanation [" + AWAITING_UPGRADE.getExplanation() + "]";
LOGGER.debug(msg);
Exception detail = new IllegalStateException(msg);
throw new ElasticsearchStatusException("Could not start datafeed [" + datafeedId +"] as indices are being upgraded";
RestStatus.TOO_MANY_REQUESTS; detail);
}
AssignmentFailure assignmentFailure = checkAssignment();
if (assignmentFailure != null && assignmentFailure.isCriticalForTaskCreation) {
String msg = "No node found to start datafeed [" + datafeedId + "]; " +
