Github user laurentgo commented on a diff in the pull request:
https://github.com/apache/drill/pull/858#discussion_r123865940
--- Diff:
exec/jdbc/src/main/java/org/apache/drill/jdbc/impl/DrillStatementImpl.java ---
@@ -38,8 +44,12 @@
// methods for compatibility.)
class DrillStatementImpl extends AvaticaStatement implements
DrillStatement,
DrillRemoteStatement {
+ //Not using the DrillbitContext's ExecutorService as this is threadPool
is light-weight (threads wake up to cancel tasks) but needs a low response time
+ private static ExecutorService queryTimeoutTaskPool =
Executors.newCachedThreadPool(new NamedThreadFactory("q-timeout-"));
--- End diff --
Maybe simpler, but I'm doubtful this is less overhead: there's a new
executor and a new thread created for each statement (to compare with just
telling the system how long to wait in DrillCursor, knowing that the i/o system
is using a fixed number of threads vs your proposal which is unbounded).
I'm also not following why you would need a new state, as this would be a
pure client thing.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
