----- Original Message ----- > > Felix/Daniel, > I have a question on your solution. > To mitigate the problem of handing out broken connections > > the pool often > > has functionality to check a connection, before it hands it > > to the > > client. When using the tomcat database connection pool, that > > feature can > > be configured by setting validationQuery to a valid > > SQL-query (e.g > > "select 1 from dual" for oracle) and testOnBorrow to true. > > Suppose the client tries to get a connection from the pool, and if > the connection is broken , the validation fails. If the validation > fails, does the client gets a new connection from the pool.
When you define your connection pool in Tomcat (i.e. using a <Resource/> tag), if you have "testOnBorrow" set to true and a "validationQuery" specified then the connection pool will test connections before it distributes them to your application. If a test fails the connection that failed the test will be removed from the connection pool. The pool will then attempt to borrow another connection from the pool. This does not mean it will be a "new" connection though. It will just go back to the pool and try again. >From the DBCP documentation... "The indication of whether objects will be validated before being borrowed from the pool. **If the object fails to validate, it will be dropped from the pool, and we will attempt to borrow another.** NOTE - for a true value to have any effect, the validationQuery parameter must be set to a non-null string." https://commons.apache.org/dbcp/configuration.html Dan > > > > > --- On Sat, 21/7/12, Felix Schumacher > <felix.schumac...@internetallee.de> wrote: > > > From: Felix Schumacher <felix.schumac...@internetallee.de> > > Subject: Re: java.sql.SQLException: No more data to read from > > socket > > To: users@tomcat.apache.org > > Date: Saturday, 21 July, 2012, 6:41 PM > > Am Samstag, den 21.07.2012, 08:44 > > +0800 schrieb vijay mathew: > > > For the last 2 failures, this issue appeared on > > Monday and DB gets restarted every Saturday. So I assume > > that this issue has to do something with the DB restart for > > the last 2 failures. > > > > > > However 1 months back, the same issue appeared on a > > Wednesday. > > > > > > So I was not able find a real pattern for this issue > > when I compare the last 3 failures. > > You are using a pool of connections. The pool maintains at > > least three > > different types of connections: > > > > * alive connections, which are borrowed to a client > > * alive connections, which are idling in the pool > > * closed connections, which wait to be opened by the pool > > > > When you restart the database, all alive connections will > > (probably) be > > broken. If you ask the pool for a new connection, it can > > give you a > > previously closed one and open it, or a previously alive > > (and probably > > broken) one. > > > > So you might be lucky by getting a newly opened one, or have > > bad luck > > and get a broken connection. The pool should prefer to give > > one of the > > "alive" connections to you, though. > > > > To mitigate the problem of handing out broken connections > > the pool often > > has functionality to check a connection, before it hands it > > to the > > client. When using the tomcat database connection pool, that > > feature can > > be configured by setting validationQuery to a valid > > SQL-query (e.g > > "select 1 from dual" for oracle) and testOnBorrow to true. > > > > On the other hand, your code could store the connection > > somewhere and > > use it after a long period of time again. If the database > > was restarted > > in the meantime, you would get that error, but the > > validation query in > > the pool would not be able to help you. > > > > Regards > > Felix > > > > > > > > > --- On Fri, 20/7/12, Christopher Schultz > > > <ch...@christopherschultz.net> > > wrote: > > > > > > > From: Christopher Schultz <ch...@christopherschultz.net> > > > > Subject: Re: java.sql.SQLException: No more data > > to read from socket > > > > To: "Tomcat Users List" <users@tomcat.apache.org> > > > > Date: Friday, 20 July, 2012, 10:21 PM > > > > -----BEGIN PGP SIGNED MESSAGE----- > > > > Hash: SHA1 > > > > > > > > - -----BEGIN PGP SIGNED MESSAGE----- > > > > Hash: SHA1 > > > > > > > > Vijay, > > > > > > > > On 7/19/12 8:25 PM, vijay mathew wrote: > > > > > > > > > java.sql.SQLException: No more data to read > > from socket > > > > at > > > > > > > > > > > oracle.jdbc.driver.SQLStateMapping.newSQLException(SQLStateMapping.java:74) > > > > > > > > > > > > > > > > > > > > > > > > at > > > > > > oracle.jdbc.driver.DatabaseError.newSQLException(DatabaseError.java:110) > > > > > at > > > > > > > > > > > oracle.jdbc.driver.DatabaseError.throwSqlException(DatabaseError.java:171) > > > > > > > > > > > > > > > > > > > > > > > > at > > > > > > oracle.jdbc.driver.DatabaseError.throwSqlException(DatabaseError.java:227) > > > > > at > > > > > > > > > > > oracle.jdbc.driver.DatabaseError.throwSqlException(DatabaseError.java:439) > > > > > > > > > > > > > > > > > > > > > > > > at > > > > > > oracle.jdbc.driver.T4CMAREngine.unmarshalUB1(T4CMAREngine.java:1042) > > > > > at > > > > > > > > > > > oracle.jdbc.driver.T4CMAREngine.unmarshalSB1(T4CMAREngine.java:999) > > > > > > > > > > > > > > > > > > > > > > > > at > > oracle.jdbc.driver.T4C8Oall.receive(T4C8Oall.java:584) > > > > > at > > > > > > oracle.jdbc.driver.T4CStatement.doOall8(T4CStatement.java:183) > > > > > > > > > at > > > > > > > > > > > oracle.jdbc.driver.T4CStatement.executeForDescribe(T4CStatement.java:774) > > > > > > > > > > > > > > > > > > > > > > > > at > > > > > > oracle.jdbc.driver.T4CStatement.executeMaybeDescribe(T4CStatement.java:849) > > > > > at > > > > > > > > > > > oracle.jdbc.driver.OracleStatement.doExecuteWithTimeout(OracleStatement.java:1186) > > > > > > > > > > > > > > > > > > > > > > > > at > > > > > > oracle.jdbc.driver.OracleStatement.executeQuery(OracleStatement.java:1377) > > > > > at > > > > > > > > > > > oracle.jdbc.driver.OracleStatementWrapper.executeQuery(OracleStatementWrapper.java:386) > > > > > > > > > > > > > > > > > > > > > > > > at > > > > > > com.merck.mrl.pcisrr.mrlsos.loginservlet.service(loginservlet.java:124) > > > > > at > > > > > > javax.servlet.http.HttpServlet.service(HttpServlet.java:802) > > > > > > > > So, when this happens, it always happens alongside > > an Oracle > > > > restart, > > > > but an Oracle restart does not always mean you'll > > get one of > > > > these errors? > > > > > > > > - - -chris > > > > - -----BEGIN PGP SIGNATURE----- > > > > Version: GnuPG/MacGPG2 v2.0.17 (Darwin) > > > > Comment: GPGTools - http://gpgtools.org > > > > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ > > > > > > > > > > iEUEARECAAYFAlAJixAACgkQ9CaO5/Lv0PDr8wCY2P8stZkV5AmvW28eYVf2wAYQ > > > > SACeJBVjBTrLzWTTtkH1Hcnsh5IPKLI= > > > > =pXsE > > > > - -----END PGP SIGNATURE----- > > > > -----BEGIN PGP SIGNATURE----- > > > > Version: GnuPG/MacGPG2 v2.0.17 (Darwin) > > > > Comment: GPGTools - http://gpgtools.org > > > > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ > > > > > > > > > > iEYEARECAAYFAlAJjJ8ACgkQ9CaO5/Lv0PBKQgCcCQPPTZaSCmug5EBYrXhONKXv > > > > PskAn2hQ4zG+gQGK+9t7IyRqWO/0z2AM > > > > =ckQ+ > > > > -----END PGP SIGNATURE----- > > > > > > > > > > --------------------------------------------------------------------- > > > > To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org > > > > For additional commands, e-mail: users-h...@tomcat.apache.org > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org > > > For additional commands, e-mail: users-h...@tomcat.apache.org > > > > > > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org > > For additional commands, e-mail: users-h...@tomcat.apache.org > > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org > For additional commands, e-mail: users-h...@tomcat.apache.org > > --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
