Real-world needs demand real-world "what happens if I poke here?" solutions.
One thing I love about my job at Krengeltech is doing discovery work to determine if a collection of technologies can work together to accomplish a business need. Such was the case today. A business had the need to communicate from Node.js on CentOS to DB2 for i. There are many options available to accomplish this task and I've listed some of them below. I've included commentary concerning my thoughts about each.
- Use DB2 Connect from IBM. This requires the purchase of licenses from IBM for each developer and server. I've used this before for another IBM i customer that had Node.js on Ubuntu, so I know it is a solid solution.
- Communicate via the Node.js iToolkit from IBM. Under the covers, this communicates via HTTP to XMLSERVICE on the IBM i and allows for interaction with DB2 and RPG. It isn't as fast as what one would probably desire when compared to 1-tier database access.
- Utilize the jdbc npm. I've always considered the JDBC driver to be a solid solution for connecting Java to DB2 for i. This does require a Java Virtual Machine (JVM) to be loaded on the machine. Not a huge issue, but it is another technology in the stack (the fewer the better in my opinion).
- Create Node.js micro services on IBM i that can be invoked from Node.js on CentOS. These "micro services" could be thick or thin. What I mean by that is you could have a thin Node.js program that simply received in an SQL string, executed it against DB2 for i, and returned the result set(s). Or you could have a thicker Node.js program that invoked business functions (e.g., getCustomerList) instead of allowing raw SQL to be run.
After a brief customer call, we decided to pursue the jdbc npm. The install instructions said to run the following command. Note I am running the below command on my Mac, not on the IBM i.
$ cd git && mkdir node_jdbc && cd node_jdbc
$ npm i --save jdbc
npm WARN deprecated node-uuid@1.4.7: use uuid module instead
-
> java@0.8.0 install /Users/aaronbartell/s3/git/node_jdbc/node_modules/jdbc/node_modules/java
> node-gyp rebuild
gyp WARN download NVM_NODEJS_ORG_MIRROR is deprecated and will be removed in node-gyp v4, please use NODEJS_ORG_MIRROR
gyp WARN download NVM_NODEJS_ORG_MIRROR is deprecated and will be removed in node-gyp v4, please use NODEJS_ORG_MIRROR
gyp WARN download NVM_NODEJS_ORG_MIRROR is deprecated and will be removed in node-gyp v4, please use NODEJS_ORG_MIRROR
CXX(target) Release/obj.target/nodejavabridge_bindings/src/java.o
CXX(target) Release/obj.target/nodejavabridge_bindings/src/javaObject.o
CXX(target) Release/obj.target/nodejavabridge_bindings/src/javaScope.o
CXX(target) Release/obj.target/nodejavabridge_bindings/src/methodCallBaton.o
CXX(target) Release/obj.target/nodejavabridge_bindings/src/nodeJavaBridge.o
CXX(target) Release/obj.target/nodejavabridge_bindings/src/utils.o
SOLINK_MODULE(target) Release/nodejavabridge_bindings.node
clang: warning: libstdc++ is deprecated; move to libc++ with a minimum deployment target of OS X 10.9
> java@0.8.0 postinstall /Users/aaronbartell/s3/git/node_jdbc/node_modules/jdbc/node_modules/java
> node postInstall.js
jdbc@0.4.0 node_modules/jdbc
??? node-uuid@1.4.7
??? winston@2.3.1 (cycle@1.0.3, eyes@0.1.8, stack-trace@0.0.9, isstream@0.1.2, async@1.0.0, colors@1.0.3)
??? async@2.1.4
???
??? java@0.8.0 (nan@2.4.0, find-java-home@0.1.3, async@2.0.1, glob@7.1.1,
Success! At this point, we have the jdbc npm installed on the laptop. If you're on Windows, I'd highly recommend you pursue "Git for Windows" from Github. It gives you a Linux-like shell that will allow you to run things like the above. You can see a fuller tutorial for getting set up with Node.js on Windows 10 at this tutorial.
Having used JDBC for many things in the past, I knew I'd need the jt400.jar file to use as the database adapter. You can obtain the jt400.jar file from the JTOpen project. Specifically, go here to download the jtopen_9_1.zip file and unzip it at the location of your choosing.
Next, run the following commands to get the jt400.jar file into your project.
$ cd node_jdbc
$ mkdir drivers
$ cp path/to/jtopen_9_1/lib/jt400.jar drivers
At this point, I believe I have everything I need and go looking for example code in the Github repo for the jdbc npm project. I do a search for "as400" in the repo and come up empty handed. I searched for "as400" because of prior experience knowing that's what the URL connection string contains in it.
At this point, I start piecing together code from other examples in the jdbc npm Github repo. I'll often look for MySQL examples and hack them to work for DB2 for i because they can sometimes be more similar than others.
Below is the full working example.
----app.js----
//
// node app.js <schema> <user> <password>
//
var JDBC = require('jdbc');
var jinst = require('jdbc/lib/jinst');
var asyncjs = require('async');
if (!jinst.isJvmCreated()) {
jinst.addOption("-Xrs");
jinst.setupClasspath(['./drivers/jt400.jar']);
}
var server = process.argv[2];
var schema = process.argv[3];
var user = process.argv[4];
var password = process.argv[5];
var config = {
url: 'jdbc:as400://' + server + '/' + schema,
drivername: 'com.ibm.as400.access.AS400JDBCDriver',
minpoolsize: 10,
maxpoolsize: 100,
properties: {
user: user,
password: password
}
};
var ibmi = new JDBC(config);
ibmi.initialize(function(err) {
if (err) {
console.log(err);
}
});
ibmi.reserve(function(err, connObj) {
// The connection returned from the pool is an object with two fields
// {uuid: <uuid>, conn: <Connection>}
if (connObj) {
console.log("Using connection: " + connObj.uuid);
// Grab the Connection for use.
var conn = connObj.conn;
// Adjust some connection options. See connection.js for a full set of
// supported methods.
asyncjs.series([
function(callback) {
conn.setAutoCommit(false, function(err) {
if (err) {
callback(err);
} else {
callback(null);
}
});
},
function(callback) {
conn.setSchema(schema, function(err) {
if (err) {
callback(err);
} else {
callback(null);
}
});
}
], function(err, results) {
// Process result
});
// Query the database.
asyncjs.series([
function(callback) {
// CREATE SQL.
conn.createStatement(function(err, statement) {
if (err) {
callback(err);
} else {
statement.executeUpdate("CREATE TABLE blah "
+ "(id int, name varchar(10), date DATE, "
+ " time TIME, timestamp TIMESTAMP)",
function(err, count) {
if (err) {
callback(err);
} else {
console.log('SUCCESS: create table')
callback(null, count);
}
});
}
});
},
function(callback) {
conn.createStatement(function(err, statement) {
if (err) {
callback(err);
} else {
statement.executeUpdate("INSERT INTO blah "
+ "VALUES (1, 'Jason', CURRENT_DATE, "
+ "CURRENT_TIME, CURRENT_TIMESTAMP)",
function(err, count) {
if (err) {
callback(err);
} else {
console.log('SUCCESS: insert into')
callback(null, count);
}
});
}
});
},
function(callback) {
// Update statement.
conn.createStatement(function(err, statement) {
if (err) {
callback(err);
} else {
statement.executeUpdate("UPDATE blah "
+ "SET id = 2 "
+ "WHERE name = 'Jason'",
function(err, count) {
if (err) {
callback(err);
} else {
console.log('SUCCESS: update')
callback(null, count);
}
});
}
});
},
function(callback) {
// Select statement example.
conn.createStatement(function(err, statement) {
if (err) {
callback(err);
} else {
// Adjust some statement options before use. See statement.js for
// a full listing of supported options.
statement.setFetchSize(100, function(err) {
if (err) {
callback(err);
} else {
statement.executeQuery("SELECT * FROM blah",
function(err, resultset) {
if (err) {
callback(err)
} else {
console.log('SUCCESS: select')
resultset.toObjArray(function(err, results) {
if (results.length > 0) {
console.log("ID: " + results[0].ID);
}
callback(null, resultset);
});
}
});
}
});
}
});
},
function(callback) {
conn.createStatement(function(err, statement) {
if (err) {
callback(err);
} else {
statement.executeUpdate("DELETE FROM blah WHERE id = 2", function(err, count) {
if (err) {
callback(err);
} else {
console.log("SUCCESS: DELETE FROM blah WHERE id = 2")
callback(null, count);
}
});
}
});
},
function(callback) {
conn.createStatement(function(err, statement) {
if (err) {
callback(err);
} else {
statement.executeUpdate("DROP TABLE blah", function(err, count) {
if (err) {
callback(err);
} else {
console.log("SUCCESS: DROP TABLE blah")
callback(null, count);
}
});
}
});
}
], function(err, results) {
console.log('results2:' + results);
ibmi.release(connObj, function(err) {
if (err) {
console.log(err.message);
}
});
});
}
});
As you can see, there are a lot of things going on with this code. I will not go into a number of things like how Node.js communicates with Java or how the asyncjs npm is used to synchronize the calling of SQL statements one after another. Instead, I will call out a few things that are unique to IBM i.
The first thing to notice is the config variable's url and drivername property values. They are both IBM i-specific, and I usually need to Google for them to remember the makeup of the strings. The other thing I had to do to the example code was remove the trailing semicolons from the SQL statements so DB2 would not error out.
Now it's time to give the program a try, as shown below.
$ node app.js 192.168.0.10 SCHEMA USER PASSWORD
module.js:327
throw err;
^
Error: Cannot find module 'async'
at Function.Module._resolveFilename (module.js:325:15)
at Function.Module._load (module.js:276:25)
at Module.require (module.js:353:17)
at require (internal/module.js:12:17)
at Object.<anonymous> (/Users/aaronbartell/s3/git/sofn_node_jdbc/app.js:32:15)
at Module._compile (module.js:409:26)
at Object.Module._extensions..js (module.js:416:10)
at Module.load (module.js:343:32)
at Function.Module._load (module.js:300:12)
at Function.Module.runMain (module.js:441:10)
Shoot! I forgot to install the async npm. Let's do that now.
$ npm install async
async@2.1.4 node_modules/async
???
OK, now to run the program again.
$ node app.js 192.168.0.10 SCHEMA USER PASSWORD
Using connection: 33dda6bd-xxxx-9999-xxxx-4d616777f555
SUCCESS: create table
SUCCESS: insert into
SUCCESS: update
SUCCESS: select
ID: 2
SUCCESS: DELETE FROM blah WHERE id = 2
SUCCESS: DROP TABLE blah
results2:0,1,1,[object Object],1,0
Rock. On. As you can see, it worked and we are now communicating from Node.js on Mac/Linux to DB2 for i using JDBC. #winning
The above code accomplished a variety of database interactions for the sake of proof. I should note that I spent the better part of 10 minutes trying to figure out why I couldn't connect, only to learn my profile was disabled. I convey this so you know the error reporting might not be what you expect.
I hope it is obvious that you need to have Java installed on the machine where this is being run. On that note, IBM i has a good JVM that could allow Node.js on IBM i to communicate to other databases. I have not tried that. Maybe one of you will give that a try and add your results to the comments section below?
If you have any questions or comments, then please comment below or email me at
LATEST COMMENTS
MC Press Online