Author: ilmarkerm
- Written by: ilmarkerm
- Category: Blog entry
- Published: June 20, 2012
Back-story: A developer came to me and wanted explanation for a weird behavior in MySQL. They inserted a record (to InnoDB table), committed, and after receiving a message (on another application) tried to read that inserted record immediately, but the newly inserted record was not found. Problem only happened in production, but not always (quite frequently).
After comparing the MySQL parameter files between production and development environments I discovered that in production autocommit was disabled to make MySQL behave more like Oracle. This setting was removed from development after we rebuilt the environment (to use multiple MySQL instances with Oracle Clusterware, instead of one large MySQL instance), but the rebuild was not yet done in production.
The default transaction level for MySQL InnoDB is REPEATABLE READ (unlike Oracle, that has READ COMMITTED as default), that means that the SELECT query always returns the data at the time point when the transaction was started. If autocommit is off, then the first issued select statement will open the transaction and any subsequent select statement will return the data at the time point when the first select was issued, until transaction is ended with COMMIT/ROLLBACK. If autocommit is enabled, SELECT statement is run in a self-contained transaction, ending with COMMIT, so the end result is like READ COMMITTED isolation level in Oracle.
Here is an example what you’d expect to see as a result:
mysql session 1$ create table test (id integer unsigned primary key) engine=innodb; Query OK, 0 rows affected (0.01 sec) mysql session 1$ set autocommit=1; Query OK, 0 rows affected (0.00 sec) mysql session 1$ select * from test; Empty set (0.01 sec) mysql session 2$ begin; Query OK, 0 rows affected (0.00 sec) mysql session 2$ insert into test values (1); Query OK, 1 row affected (0.05 sec) mysql session 2$ commit; Query OK, 0 rows affected (0.00 sec) mysql session 1$ select * from test; +----+ | id | +----+ | 1 | +----+ 1 row in set (0.00 sec)
The same example of having autocommit off and transaction isolation level set as default:
mysql session 1$ set autocommit=0; Query OK, 0 rows affected (0.00 sec) mysql session 1$ select * from test; +----+ | id | +----+ | 1 | +----+ 1 row in set (0.00 sec) mysql session 2$ begin; Query OK, 0 rows affected (0.00 sec) mysql session 2$ insert into test values (2); Query OK, 1 row affected (0.05 sec) mysql session 2$ commit; Query OK, 0 rows affected (0.00 sec) mysql session 1$ select * from test; +----+ | id | +----+ | 1 | +----+ 1 row in set (0.00 sec) mysql session 1$ commit; Query OK, 0 rows affected (0.00 sec) mysql session 1$ select * from test; +----+ | id | +----+ | 1 | | 2 | +----+ 2 rows in set (0.00 sec)
Now, autocommit off and transaction isolation level set to READ COMMITTED:
mysql session 1$ set autocommit=0; Query OK, 0 rows affected (0.00 sec) mysql session 1$ set session transaction isolation level read committed; Query OK, 0 rows affected (0.03 sec) mysql session 1$ select * from test; +----+ | id | +----+ | 1 | | 2 | +----+ 2 rows in set (0.01 sec) mysql session 2$ begin; Query OK, 0 rows affected (0.00 sec) mysql session 2$ insert into test values (3); Query OK, 1 row affected (0.05 sec) mysql session 2$ commit; Query OK, 0 rows affected (0.00 sec) mysql session 1$ select * from test; +----+ | id | +----+ | 1 | | 2 | | 3 | +----+ 3 rows in set (0.00 sec)
You can read more about how autocommit is handled by InnoDB in documentation.
Another key component in our puzzle – application server used connection pools for managing MySQL connections, so after Java session requested a connection from pool, it was only a matter of chance whether the connection already had a transaction left open by the previous user of that connection, did the previous user commit, or was it a brand new connection.
- Written by: ilmarkerm
- Category: Blog entry
- Published: May 17, 2012
Oracle Database has had the possibility to run Java code inside the database for a long time. It’s a very rare occasion when you need to use it but still. Here is one example I used to download content from HTTPS website that required user certificates for authentication. Please take the code below more as an example how to put simple Java code inside the database, not as a solution for user certificates authentication, because UTL_HTTP can do the same thing (although I wasn’t successful in implementing it under 11.2.0.2).
First, load the Java source into database. The code below shows:
- How to return simple datatype (int) from Java function – makeConnection
- How to return Oracle CLOB datatype from Java – makeConnectionClob
- How to execute SQL from Java, in the same calling session
Note that method main is just added for testing from command line.
Then you need to create a wrapper package in database. This declares the PL/SQL wrapper function names and input/output parameters.
Download the source: java_source.java and PL/SQL wrapper.sql.
When you first execute the code, you will most likely get some privilege errors, but the error message will tell you how to grant the needed privileges. For example, for this code the following grants were needed:
exec dbms_java.grant_permission( 'OWNER', 'SYS:java.util.PropertyPermission', 'javax.net.ssl.keyStore', 'write' ); exec dbms_java.grant_permission( 'OWNER', 'SYS:java.util.PropertyPermission', 'javax.net.ssl.trustStore', 'write' ); exec dbms_java.grant_permission( 'OWNER', 'SYS:java.util.PropertyPermission', 'javax.net.ssl.keyStorePassword', 'write' ); exec dbms_java.grant_permission( 'OWNER', 'SYS:java.net.SocketPermission', 'site.that.requires.user.cert', 'resolve' ); exec dbms_java.grant_permission( 'OWNER', 'SYS:java.net.SocketPermission', '1.2.3.4:443', 'connect,resolve' );
- Written by: ilmarkerm
- Category: Blog entry
- Published: May 17, 2012
Oracle Wallet Manager and orapki do not let you extract the private key associated with user certificate located in Oracle Wallet. If you need it for some reason, for example testing with external tools like wget, then its possible to extract the private key using openssl, since Orale Wallet (ewallet.p12 file) is just PKCS#12 file.
[oracle@jfadboc1n01 wallet]$ openssl pkcs12 -in /home/oracle/wallet/ewallet.p12 -nocerts -out private_key.pem Enter Import Password: MAC verified OK Warning unsupported bag type: secretBag Enter PEM pass phrase: Verifying - Enter PEM pass phrase: [oracle@jfadboc1n01 wallet]$ ls -l total 16 -rw-r--r-- 1 oracle oinstall 11629 May 15 11:38 ewallet.p12 -rw-r--r-- 1 oracle oinstall 1879 May 17 08:53 private_key.pem
Here private_key.pem contains the private key extracted from Oracle Wallet.
The first password that openssl asks (Enter Import Password) is the wallet password, the other password (Enter PEM pass phrase) is used to protect the exported key.
- Written by: ilmarkerm
- Category: Blog entry
- Published: May 10, 2012
When instance fails in Oracle RAC, the services that were using this instance as a preferred instance are automatically relocated to instances marked as available for this service. But after the failed instance recovers and starts up again, the relocated services are not moved back and need manual srvctl relocate service command from administrator to move them back.
Here is a little Bash script to automate this process. Oracle Clusterware (Grid Infrastructure) can execute user callout scripts on FAN events, like INSTANCE up/down. Place this script under $GRID_HOME/racg/usrco/ and set the execute bits on the file. Then clusterware will execute that script for all FAN events, but the script will start processing only for instance up event.
Why is it needed? We just switched over to 4-node RAC consisting of many different applications, almost each of them connecting to its own schema. We created each application its own service restricting it to 1 (or max 2) nodes (1 or 2 nodes as preferred, all other nodes listed as available). After the first rolling patching, I noticed that the connection count and load on each node was very unbalanced, vast majority of the connections were connected to node1 and the last patched node had almost none and it did not get better over a few hours. This was because most of the services ended up on node1 and I had to manually look over each service and relocate it back where it belongs. This script attempts to automate this process.
Tested on Oracle Linux 5.8 with Oracle Grid Infrastructure 11.2.0.3 and Oracle Database 11.2.0.2 and 11.2.0.3.
- Written by: ilmarkerm
- Category: Blog entry
- Published: November 9, 2011
I’ve been doing lately quite many database clustering implementations; Oracle RAC and since we have many MySQL instances in production, had to find a good way to make MySQL instances highly available also.
One good solution for this is managing MySQL instances with clusterware and since we are planning to use Oracle RAC on Oracle Enterprise Linux anyway, then Oracle Clusterware is an excellent candidate for this task. Also… Oracle Clusterware is included with Oracle Enterprise Linux at no additional charge.
Requirements I had:
- Multiple MySQL instances running in the same cluster, in case of node failure affected MySQL instances are moved to any other surviving node (least loaded)
- Different MySQL instances may run different MySQL RDBMS software versions
- Each MySQL instance is listening to its own dedicated and clusterware managed IP address (so MySQL can always respond on default 3306 port and can move independently around nodes without any dependencies to other MySQL instances)
- Clusterware monitors each MySQL instance and in case of trouble automatically moves IP and starts instance on another cluster node
For this task I wrote some Bash scripts to help managing MySQL instances in Oracle Clusterware environment. All these scripts are available here: Google Code project page, also with documentation of the whole setup and how to use scripts: Documentation
All comments welcome!