tree 4934643d99001ce278fd62ae53a48ea7e8612b3d
parent 38ca962f45aa0f3c56843325b26f1ceaa5e86253
author Abdullah Alamoudi <bamousaa@gmail.com> 1519705401 -0800
committer abdullah alamoudi <bamousaa@gmail.com> 1519843475 -0800

[NO ISSUE][ING] Close and return on interrupt of ingestion

- user model changes: no
- storage format changes: no
- interface changes: no

Details:
- Previously, we would still attempt to write through the network
  in case an ingestion task is interrupted.
- The goal was to try and get as much data in as possible but in
  the case where the cluster was in a bad state, this could
  lead to hanging threads.
- After this change, each record reader must implement the stop
  method correctly to allow for graceful stop while interrupts
  will always mean abort the task and return as soon as possible.

Change-Id: I6119617d133fb161a48b39f9812ec79e0189975b
Reviewed-on: https://asterix-gerrit.ics.uci.edu/2428
Sonar-Qube: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Integration-Tests: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Tested-by: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Reviewed-by: Michael Blow <mblow@apache.org>