VACUUM cannot run inside a transaction block - postgresql

My application has this code:
Connection connection = DriverManager.getConnection(url, userName, password);
connection.setAutoCommit(false);
try {
Statement statement = connection.createStatement();
try {
statement.executeUpdate("TRUNCATE mytable");
} finally {
statement.close();
}
connection.commit();
connection.setAutoCommit(true);
statement = connection.createStatement();
try {
statement.executeUpdate("VACUUM mytable");
} finally {
statement.close();
}
connection.setAutoCommit(false);
connection.commit();
} finally {
connection.close();
}
It fails on statement.executeUpdate("VACUUM mytable"); because VACUUM cannot run inside a transaction block.
I must be missing some notions as I thought that setting the autocommit to true would remove the transaction context (needed for TRUNCATE). Apparently it doesn't. So how do I get out of the transactional context and perform my VACUUM successfully? How should I write my code to not have this transactional context?

Related

How to use entity framework transaction in raw query?

I am using entity framework but doing my operations with raw queries. My operations are like following:
Check if recırd exist with integration_id
Delete record if exit
Insert new record
So I am using transaction
using (var transaction = await _context.Database.BeginTransactionAsync())
{
var isExist = await IsExist(id);
if (isExist)
{
var deleteQuery = "delete from ....";
await _context.Database.ExecuteSqlRawAsync(deleteQuery);
}
var insertQuery = "insert into ...";
await _context.Database.ExecuteSqlRawAsync(insertQuery);
}
if insert operation fails, does deleted record rollback?
UPD: https://learn.microsoft.com/en-us/ef/core/saving/transactions#controlling-transactions
transaction will auto-rollback when disposed if either commands fails
So, my code below may be overkill on the catch side, but Commit is still essential :)
======================
I believe the correct way of using transaction would be following:
using (var transaction = await _context.Database.BeginTransactionAsync())
{
try
{
var isExist = await IsExist(id);
if (isExist)
{
var deleteQuery = "delete from ....";
await _context.Database.ExecuteSqlRawAsync(deleteQuery);
}
var insertQuery = "insert into ...";
await _context.Database.ExecuteSqlRawAsync(insertQuery);
// there we tell DB to finish the transaction,
// mark all changes as permanent and release all locks
transaction.Commit();
}
catch (Exception ex)
{
// there we tell DB to discard all changes
// made by this transaction that may be discarded
transaction.Rollback();
// log error
}
}
But I never used BeginTransaction*Async* personally before.
This method doesn't start transaction on it's own. If you need to execute queries in transaction you need to first call
BeginTransaction(DatabaseFacade, IsolationLevel) or UseTransaction.
Reference
learn.microsoft.com
So in your case it will execute queries in a transaction and roll back all the queries if any of the query failed

Connectjboss 4 with Firebird 3 via Jaybird 2.214-jdk1.6 [duplicate]

As soon as my code gets to my while(rs.next()) loop it produces the ResultSet is closed exception. What causes this exception and how can I correct for it?
EDIT: I notice in my code that I am nesting while(rs.next()) loop with another (rs2.next()), both result sets coming from the same DB, is this an issue?
Sounds like you executed another statement in the same connection before traversing the result set from the first statement. If you're nesting the processing of two result sets from the same database, you're doing something wrong. The combination of those sets should be done on the database side.
This could be caused by a number of reasons, including the driver you are using.
a) Some drivers do not allow nested statements. Depending if your driver supports JDBC 3.0 you should check the third parameter when creating the Statement object. For instance, I had the same problem with the JayBird driver to Firebird, but the code worked fine with the postgres driver. Then I added the third parameter to the createStatement method call and set it to ResultSet.HOLD_CURSORS_OVER_COMMIT, and the code started working fine for Firebird too.
static void testNestedRS() throws SQLException {
Connection con =null;
try {
// GET A CONNECTION
con = ConexionDesdeArchivo.obtenerConexion("examen-dest");
String sql1 = "select * from reportes_clasificacion";
Statement st1 = con.createStatement(
ResultSet.TYPE_SCROLL_INSENSITIVE,
ResultSet.CONCUR_READ_ONLY,
ResultSet.HOLD_CURSORS_OVER_COMMIT);
ResultSet rs1 = null;
try {
// EXECUTE THE FIRST QRY
rs1 = st1.executeQuery(sql1);
while (rs1.next()) {
// THIS LINE WILL BE PRINTED JUST ONCE ON
// SOME DRIVERS UNLESS YOU CREATE THE STATEMENT
// WITH 3 PARAMETERS USING
// ResultSet.HOLD_CURSORS_OVER_COMMIT
System.out.println("ST1 Row #: " + rs1.getRow());
String sql2 = "select * from reportes";
Statement st2 = con.createStatement(
ResultSet.TYPE_SCROLL_INSENSITIVE,
ResultSet.CONCUR_READ_ONLY);
// EXECUTE THE SECOND QRY. THIS CLOSES THE FIRST
// ResultSet ON SOME DRIVERS WITHOUT USING
// ResultSet.HOLD_CURSORS_OVER_COMMIT
st2.executeQuery(sql2);
st2.close();
}
} catch (SQLException e) {
e.printStackTrace();
} finally {
rs1.close();
st1.close();
}
} catch (SQLException e) {
} finally {
con.close();
}
}
b) There could be a bug in your code. Remember that you cannot reuse the Statement object, once you re-execute a query on the same statement object, all the opened resultsets associated with the statement are closed. Make sure you are not closing the statement.
Also, you can only have one result set open from each statement. So if you are iterating through two result sets at the same time, make sure they are executed on different statements. Opening a second result set on one statement will implicitly close the first.
http://java.sun.com/javase/6/docs/api/java/sql/Statement.html
The exception states that your result is closed. You should examine your code and look for all location where you issue a ResultSet.close() call. Also look for Statement.close() and Connection.close(). For sure, one of them gets called before rs.next() is called.
You may have closed either the Connection or Statement that made the ResultSet, which would lead to the ResultSet being closed as well.
Proper jdbc call should look something like:
try {
Connection conn;
Statement stmt;
ResultSet rs;
try {
conn = DriverManager.getConnection(myUrl,"","");
stmt = conn.createStatement();
rs = stmt.executeQuery(myQuery);
while ( rs.next() ) {
// process results
}
} catch (SqlException e) {
System.err.println("Got an exception! ");
System.err.println(e.getMessage());
} finally {
// you should release your resources here
if (rs != null) {
rs.close();
}
if (stmt != null) {
stmt.close();
}
if (conn != null) {
conn.close();
}
}
} catch (SqlException e) {
System.err.println("Got an exception! ");
System.err.println(e.getMessage());
}
you can close connection (or statement) only after you get result from result set. Safest way is to do it in finally block. However close() could also throe SqlException, hence the other try-catch block.
I got same error everything was correct only i was using same statement interface object to execute and update the database.
After separating i.e. using different objects of statement interface for updating and executing query i resolved this error. i.e. do get rid from this do not use same statement object for both updating and executing the query.
Check whether you have declared the method where this code is executing as static. If it is static there may be some other thread resetting the ResultSet.
make sure you have closed all your statments and resultsets before running rs.next. Finaly guarantees this
public boolean flowExists( Integer idStatusPrevious, Integer idStatus, Connection connection ) {
LogUtil.logRequestMethod();
PreparedStatement ps = null;
ResultSet rs = null;
try {
ps = connection.prepareStatement( Constants.SCRIPT_SELECT_FIND_FLOW_STATUS_BY_STATUS );
ps.setInt( 1, idStatusPrevious );
ps.setInt( 2, idStatus );
rs = ps.executeQuery();
Long count = 0L;
if ( rs != null ) {
while ( rs.next() ) {
count = rs.getLong( 1 );
break;
}
}
LogUtil.logSuccessMethod();
return count > 0L;
} catch ( Exception e ) {
String errorMsg = String
.format( Constants.ERROR_FINALIZED_METHOD, ( e.getMessage() != null ? e.getMessage() : "" ) );
LogUtil.logError( errorMsg, e );
throw new FatalException( errorMsg );
} finally {
rs.close();
ps.close();
}
A ResultSetClosedException could be thrown for two reasons.
1.) You have opened another connection to the database without closing all other connections.
2.) Your ResultSet may be returning no values. So when you try to access data from the ResultSet java will throw a ResultSetClosedException.
It happens also when using a ResultSet without being in a #Transactional method.
ScrollableResults results = getScrollableResults("select e from MyEntity e");
while (results.next()) {
...
}
results.close();
if MyEntity has eager relationships with other entities. the second time results.next() is invoked the ResultSet is closed exception is raised.
so if you use ScrollableResults on entities with eager relationships make sure your method is run transactionally.
"result set is closed" happened to me when using tag <collection> in MyBatis nested (one-to-many) xml <select> statement
A Spring solution could be to have a (Java) Spring #Service layer, where class/methods calling MyBatis select-collection statements are annotated with
#Transactional(propagation = Propagation.REQUIRED)
annotations being:
import org.springframework.transaction.annotation.Propagation;
import org.springframework.transaction.annotation.Transactional;
this solution does not require to set the following datasource properties (i.e., in JBoss EAP standalone*.xml):
<xa-datasource-property name="downgradeHoldCursorsUnderXa">**true**\</xa-datasource-property>
<xa-datasource-property name="resultSetHoldability">**1**</xa-datasource-property>

should I set T-SQL command timeout back to default?

I hope this question is not too stupid. I have a long process t-sql command in my ADO.Net. I would like to increase the command timeout (please see below).
cmd.CommandTimeout = 600; // default is 30 sec. increase to 10 mins
try
{
cmd.ExecuteNonQuery();
cmd.CommandTimeout = 30; // set it back
}
catch (Exception ex)
{
string debug = ex.Message;
throw ex;
}
Should I set the timeout back to default after the long process is done? I am looking for the best practice. Thank you :-)
If you re-use the command object, you can add a finally to the try-catch, so it will be reset regardless of the result of the query. (success or exception).
But when you dispose the SqlCommand after you've used it. There's no need to reset the CommandTimeout
The best practice is to not reuse the command because you should not be reusing the connection. Holding a connection is not good for scale.
static private dbConnectionString = "CONNECTION DETAILS";
using (SQLConnection connection = new SQLConnection(dbConnectionString))
{
try
{
connection.Open();
using (SQLCommand command = connection.CreateCommand())
{
command.CommandTimeout = 600;
......
using(SQLDataReader rdr = command.ExecuteReader())
{
}
}
}
catch(SQLException ex)
{
}
finally
{
connection.Close();
}
}

POSTGRESQL: Batch entry 0 SELECT

Hi I have the following method which calls a stored function in postgresql. The call works when I use a standard executequery() method but does not work when I start using batchs. Any help will be appreciated.
public void addstuff3() throws Exception {
Statement statement = null;
ResultSet resultSet = null;
Connection conn = null;
try {
// this will load the MySQL driver, each DB has its own driver
Class.forName("org.postgresql.Driver");
// setup the connection with the DB.
conn = DriverManager
.getConnection("jdbc:postgresql://localhost/newmydb?"
+ "user=new_user&password=password");
// statements allow to issue SQL queries to the database
statement = conn.createStatement();
conn.setAutoCommit(false);
statement.addBatch("SELECT ADDSTUFF('comp1', 'mdel1','power','PROPERTY','STRING','ON', '1396983600000', 'testing');");
statement.addBatch("SELECT ADDSTUFF('comp2', 'mdel2','power','PROPERTY','STRING','ON', '1396983600000', 'testing');");
conn.commit();
statement.executeBatch();
} catch (ClassNotFoundException | SQLException e) {
throw e;
} finally {
conn.close();
// resultSet.close();
statement.close();
}
This is the Error I get:
Batch entry 0 SELECT ADDSTUFF('comp1', 'mdel1','power','PROPERTY','STRING','ON', '1396983600000', 'testing') was aborted. Call getNextException to see the cause.
at org.postgresql.jdbc2.AbstractJdbc2Statement$BatchResultHandler.handleError(AbstractJdbc2Sta tement.java:2743)
at org.postgresql.core.v3.QueryExecutorImpl$1.handleError(QueryExecutorImpl.java:461)
at org.postgresql.core.v3.QueryExecutorImpl.processResults(QueryExecutorImpl.java:1928)
at org.postgresql.core.v3.QueryExecutorImpl.execute(QueryExecutorImpl.java:405)
at org.postgresql.jdbc2.AbstractJdbc2Statement.executeBatch(AbstractJdbc2Statement.java:2892)
at jdbc.testing.MySQLAccess.addIndicators3(MySQLAccess.java:125)
at jdbc.testing.JDBCTesting.main(JDBCTesting.java:21)
Any help? I am using jdbc and postgresql.
ok thanks to #Dave I found that
e.getNextException()
Prints:
A result was returned when none was expected
I should not return a value
Works!

how to try execute repeatly if command.ExecuteNonQuery() fails

how to try execute repeatly if command.ExecuteNonQuery() fails?
You can try
bool executed = false;
while (!executed)
{
try
{
command.ExecuteNonQuery();
executed = true;
}
catch
{
}
}
You can add some more conditions like a timer or a counter but this does not seem to be a good idea. You should probably come up with a better recovery scenario.
The simplest way I can think is:
while(true) {
try {
command.ExecuteNonQuery();
break;
} catch(SqlException ex) { }
}
You should anyway put some extra control code in the catch block to prevent an infinite loop and/or to log the error.