tree a1bc3261bc67cd697dfd3be0fa7bb73cd883639b
parent 8078e65c9d3c5b17d9074ce4b77ea8851ec5e4bd
author Murtadha Hubail <mhubail@apache.org> 1537265942 +0300
committer Murtadha Hubail <mhubail@apache.org> 1537285519 -0700

[NO ISSUE][STO] Ensure Files From Failed Bulkload Are Deleted

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

Details:
- If a bulkload is interrupted at the end-stage (i.e. after all
  tuples have been added but before the component is added to
  the index's disk components), then the bulkload operation will
  fail but the files from the failed operation are not deleted.
  This change ensures that if a bulkload fails in such stage,
  the files of the generated component are deleted to avoid
  file is already mapped exception if the bulkload is attempted
  again.
- Ensure disk component is only destroyed once in case of
  bulkload cleanup.
- Stop attempting to write a checkpoint early if the failure was
  due to the thread being interrupted.

Change-Id: I7f63054ac72a9482e779e49eb4da658a08fe7e9d
Reviewed-on: https://asterix-gerrit.ics.uci.edu/2970
Reviewed-by: Murtadha Hubail <mhubail@apache.org>
Sonar-Qube: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Tested-by: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Integration-Tests: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Reviewed-by: Michael Blow <mblow@apache.org>