datawave

datawave copied to clipboard

title should be pretty self-explanatory. This would happen in the cancel call in CachedResultsQueryService.

jwomeara

There is a MonitorTask for the cached results service which will evict stale entries from the cache. We should evaluate whether we should add logic specifically to evict failed entries...

jwomeara

The CachedResultsQueryService `load` call will stay open until the load action has completed (i.e. until the query has run to completion). We don't necessarily have to do that though, and...

jwomeara

Consider whether we should add this functionality to the cached results query service.

jwomeara

The executor task is currently creating an executor event every time a checkpoint is created. Once the query microservices are deployed, we should determine whether or not we should be...

jwomeara

Both the MapReduceQueryController and QueryController have endpoints for resetting/restarting the query. For QueryController, the call is to `reset`, but for MapReduceQueryController the call is to `restart`. It would be nice...

jwomeara

We need to figure out what to pass for the query when auditing for oozie-based queries. This should be updated in MapReduceQueryManagementService.java when the oozie workflow is submitted.

jwomeara

For the QueryExecutorBean, when we fail during a createAndNext, or next call we return a DefaultEventQueryResponse containing the encountered exceptions. For the query microservices we are returning a VoidResponse instead...

jwomeara

IdTranslatorBean uses `translateId` for the singular call, and `translateIDs` for the batch call. We have recreated these paths exactly in QueryController, but should we use similar case-notation for each call?

jwomeara

Currently, our map reduce query jobs notify the query and mapreduce microservices that the job is complete through the use of a built-in status callback that passes the mapreduce job...

jwomeara