[
https://issues.apache.org/jira/browse/DERBY-5425?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Knut Anders Hatlen updated DERBY-5425:
--------------------------------------
Attachment: holdable-row-holders.diff.txt
After some more debugging, it turned out this issue happens because the
temporary conglomerates are not holdable over commits, as suggested above. When
the temporary row holder spills to disk, all rows are written to it, but they
are lost on commit, so that the scan is emptied and too few rows are returned.
I mentioned in my previous comment that I had tried making TemporaryRowHolder's
conglomerate holdable over commit with no luck. However, with the attached
patch (holdable-row-holders.diff.txt) the repro successfully returns 100000
rows. That patch also makes the scans in TemporaryRowHolderResultSet holdable.
This patch is not intended for commit, since it unconditionally changes the
holdability of all temporary row holders, but at least it shows that the lack
of holdability is what's causing this bug.
> Updateable holdable ResultSet terminates early after 65638 updates
> ------------------------------------------------------------------
>
> Key: DERBY-5425
> URL: https://issues.apache.org/jira/browse/DERBY-5425
> Project: Derby
> Issue Type: Bug
> Components: JDBC
> Affects Versions: 10.7.1.1
> Environment: ------------------ Java Information ------------------
> Java Version: 1.6.0_26
> Java Vendor: Sun Microsystems Inc.
> Java home: D:\Program Files (x86)\Java\jre6
> Java classpath: .;..\derby.jar
> OS name: Windows 7
> OS architecture: x86
> OS version: 6.1
> Java user name: Andrew
> Java user home: D:\Users\Andrew
> Java user dir: E:\workspace\DerbyBug\bin
> java.specification.name: Java Platform API Specification
> java.specification.version: 1.6
> java.runtime.version: 1.6.0_26-b03
> --------- Derby Information --------
> JRE - JDBC: Java SE 6 - JDBC 4.0
> [E:\workspace\DerbyBug\derby.jar] 10.7.1.1 - (1040133)
> ------------------------------------------------------
> ----------------- Locale Information -----------------
> ------------------------------------------------------
> Reporter: Andrew Johnson
> Labels: derby_triage10_9
> Attachments: DerbyBug.java, holdable-row-holders.diff.txt
>
>
> After at least 65638 updates to an indexed column have been done via an
> updateable holdable resultset and the transaction is committed
> ResultSet.next() returns false even if more rows exist to be returned.
> The following program should output "Total: 100000" but instead outputs
> "Total: 65638".
> import java.sql.Connection;
> import java.sql.DriverManager;
> import java.sql.PreparedStatement;
> import java.sql.ResultSet;
> import java.sql.SQLException;
> import java.sql.Statement;
> public class DerbyBug {
> public static void main(String[] args) throws ClassNotFoundException,
> SQLException {
> Class.forName("org.apache.derby.jdbc.EmbeddedDriver");
> Connection conn =
> DriverManager.getConnection("jdbc:derby:TestDB;create=true");
> conn.setAutoCommit(false);
>
> Statement createStmt = conn.createStatement();
> createStmt.executeUpdate("CREATE TABLE test (a INT)");
> createStmt.executeUpdate("CREATE INDEX idxa ON test(a)");
> createStmt.close();
>
> PreparedStatement insertStmt = conn.prepareStatement("INSERT
> INTO test(a) VALUES (?)");
>
> for (int i = 0; i < 100000; ++i) {
> insertStmt.setInt(1, i);
> insertStmt.executeUpdate();
> }
> insertStmt.close();
>
> conn.commit();
>
> Statement selectStmt =
> conn.createStatement(ResultSet.TYPE_FORWARD_ONLY, ResultSet.CONCUR_UPDATABLE,
> ResultSet.HOLD_CURSORS_OVER_COMMIT);
> ResultSet rs = selectStmt.executeQuery("SELECT a FROM test FOR
> UPDATE");
>
> int count = 0;
> while (rs.next()) {
> rs.updateInt(1, count);
> rs.updateRow();
> count++;
>
> conn.commit();
> }
>
> rs.close();
> selectStmt.close();
> conn.commit();
> conn.close();
> System.out.println("Total: " + count);
>
> try {
>
> DriverManager.getConnection("jdbc:derby:;shutdown=true");
> } catch (SQLException e) {
> if (!e.getSQLState().equals("XJ015")) {
> throw e;
> }
> }
> }
> }
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
