[IMPALA-4280] Clean up tasks in Coordinator, QueryExecState and others.

• Type: Task
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: Impala 2.8.0
• Component/s: Backend
• Target Version: Product Backlog
• Assignee: Unassigned
• Reporter: Henry Robinson
• Created: 12/Oct/16 00:01
• Updated: 13/Feb/20 16:20

Description

During the coordinator changes we're implementing in IMPALA-2905 and others, several bugs or opportunities for simplification or clarification came to light. This is an umbrella task for all those issues.

Issue Links

• relates to: IMPALA-9370 Re-factor ImpalaServer, ClientRequestState, Coordinator protocol

Sub-Tasks

• Fix some locking / concurrency issues in ImpalaServer and QueryExecState: Open
• Coordinator::GetNext() should not call WaitForAllInstances() at eos: Open
• Expected end-of-query cancellation shouldn't set Coordinator::query_status_ to CANCELLED: Open
• Move query result caching into Coordinator: Open