WebSphere Query Timeout Thread Interruption Control | Generated by AI
In WebSphere Application Server, the queryTimeoutInterruptProcessingMode property is a data source custom property that controls how query timeouts are enforced when an SQL statement exceeds its configured queryTimeout.
Here’s how it works:
Default Behavior
Normally, when a query timeout is reached, WebSphere calls the JDBC driver’s Statement.cancel() method. This only sends a cancel request to the database. Depending on the database and driver, the running SQL may or may not stop immediately. Sometimes it continues consuming resources on the database server until it finishes or until the DB itself decides to terminate it.
Purpose of queryTimeoutInterruptProcessingMode
This property determines what WebSphere itself does in addition to calling Statement.cancel():
-
When set to
true(enabled):- WebSphere interrupts the thread running the query after the timeout.
- This means the application thread will stop waiting for the query, even if the database hasn’t finished canceling it.
- Useful when you want to ensure application responsiveness and not have threads hanging indefinitely.
- Downside: interrupting a thread can cause
SQLExceptions or inconsistent states if the app or driver wasn’t designed to handle it.
-
When set to
false(disabled, default):- WebSphere does not interrupt the thread.
- It relies only on
Statement.cancel()and waits for the driver/database to handle cancellation. - Safer for app stability, but riskier for performance if the DB takes too long to cancel.
Typical Use Cases
-
Enable (
true) if:- You have seen JDBC calls hanging even after timeout.
- You want to guarantee that application threads are freed.
- You are okay handling
InterruptedException-like behavior or extra SQLExceptions in your app.
-
Keep Disabled (
false) if:- Your JDBC driver and DB reliably handle cancellations.
- You don’t want to risk interrupting threads mid-processing.
- Stability and clean DB-driver state are more important than fast thread release.
✅ In short:
queryTimeoutInterruptProcessingMode=true makes WebSphere forcibly break the waiting thread when query timeout is reached. false relies only on the DB driver to cancel the SQL. It’s a trade-off between responsiveness and stability.
Do you want me to also explain how this interacts with statementTimeout and connectionTimeout in WebSphere data sources? That way you’ll see the full picture of timeout handling.