This functionality is experimental and may be changed or removed completely in a future release. Elastic will take a best effort approach to fix any issues, but experimental features are not subject to the support SLA of official GA features.
The following limitations and known problems apply to the 7.3 release of the Elastic data frame analytics feature:
Cross-cluster search limitationedit
Cross-cluster search is not supported in 7.3 for data frame analytics.
Deleting a data frame analytics job does not delete the data frame destination indexedit
The delete data frame analytics job API does not delete the data frame destination index. That index must be deleted separately.
Cannot update a data frame analytics jobedit
You cannot update Data frame analytics configurations. Instead, delete the data frame analytics job and create a new one.
Data frame memory limitationedit
Data frame analytics can analyze data frames that fit into the memory limit dedicated for machine learning processes. For general machine learning settings, see Machine learning settings in Elasticsearch.
Field limitations in data frame analyticsedit
Currently, outlier detection only supports numeric features. Non-numeric fields will be excluded from the analysis by default. If such a field is explicitly included in the analysis, starting the data frame analytics should result in failure. Arrays are not supported either. Documents that contain analyzed fields with unsupported values will be skipped entirely.
Missing values in analyzed fieldsedit
If there are missing values in feature fields (fields that are subjects of the data frame analytics), then the document that contains the fields with the missing values will be skipped during the analysis.