Exception Testing Why test exception flows? Just like with all of your code, test coverage writes a contract between your code and the business functionality that the code is supposed to produce leaving you with a living documentation of the code along with the added ability to stress the functionality early and often. I won't go into the many benefits of testing instead I will focus on just Exception Testing. There are many ways to test an exception flow thrown from a piece of code. Lets say that you have a guarded method that requires an argument to be not null. How would you test that condition? How do you keep JUnit from reporting a failure when the exception is thrown? This blog covers a few different methods culminating with JUnit's ExpectedException implemented with JUnit's @Rule functionality. The "old" way In a not so distant past the process to test an exception required a dense amount of boilerplate code in which you would start a try/catch block, report a failure if your code did not produce the expected behavior and then catch the exception looking for the specific type. Here is an example: public class MyObjTest { @Test public void getNameWithNullValue() { try { MyObj obj = new MyObj(); myObj.setName(null); fail("This should have thrown an exception"); } catch (IllegalArgumentException e) { assertThat(e.getMessage().equals("Name must not be null")); } } } As you can see from this old example, many of the lines in the test case are just to support the lack of functionality present to specifically test exception handling. One good point to make for the try/catch method is the ability to test the specific message and any custom fields on the expected exception. We will explore this a bit further down with JUnit's ExpectedException and @Rule annotation. JUnit adds expected exceptions JUnit responded back to the users need for exception handling by adding a @Test annotation field "expected". The intention is that the entire test case will pass if the type of exception thrown matched the exception class present in the annotation. public class MyObjTest { @Test(expected = IllegalArgumentException.class) public void getNameWithNullValue() { MyObj obj = new MyObj(); myObj.setName(null); } } As you can see from the newer example, there is quite a bit less boiler plate code and the test is very concise, however, there are a few flaws. The main flaw is that the test condition is too broad. Suppose you have two variables in a signature and both cannot be null, then how do you know which variable the IllegalArgumentException was thrown for? What happens when you have extended a Throwable and need to check for the presence of a field? Keep these in mind as you read further, solutions will follow. JUnit @Rule and ExpectedException If you look at the previous example you might see that you are expecting an IllegalArgumentException to be thrown, but what if you have a custom exception? What if you want to make sure that the message contains a specific error code or message? This is where JUnit really excelled by providing a JUnit @Rule object specifically tailored to exception testing. If you are unfamiliar with JUnit @Rule, read the docs here. ExpectedException JUnit provides a JUnit class ExpectedException intended to be used as a @Rule. The ExpectedException allows for your test to declare that an exception is expected and gives you some basic built in functionality to clearly express the expected behavior. Unlike the @Test(expected) annotation feature, ExpectedException class allows you to test for specific error messages and custom fields via the Hamcrest matchers library. An example of JUnit's ExpectedException import org.junit.rules.ExpectedException; public class MyObjTest { @Rule public ExpectedException thrown = ExpectedException.none(); @Test public void getNameWithNullValue() { thrown.expect(IllegalArgumentException.class); thrown.expectMessage("Name must not be null"); MyObj obj = new MyObj(); obj.setName(null); } } As I eluded to above, the framework allows you to test for specific messages ensuring that the exception being thrown is the case that the test is specifically looking for. This is very helpful when the nullability of multiple arguments is in question. Custom Fields Arguably the most useful feature of the ExpectedException framework is the ability to use Hamcrest matchers to test your custom/extended exceptions. For example, you have a custom/extended exception that is to be thrown in a method and inside the exception has an "errorCode". How do you test that functionality without introducing the boiler plate code from the try/catch block listed above? How about a custom Matcher! This code is available at: https://github.com/mike-ensor/custom-exception-testing Solution: First the test case import org.junit.rules.ExpectedException; public class MyObjTest { @Rule public ExpectedException thrown = ExpectedException.none(); @Test public void someMethodThatThrowsCustomException() { thrown.expect(CustomException.class); thrown.expect(CustomMatcher.hasCode("110501")); MyObj obj = new MyObj(); obj.methodThatThrowsCustomException(); } } Solution: Custom matcher import com.thepixlounge.exceptions.CustomException; import org.hamcrest.Description; import org.hamcrest.TypeSafeMatcher; public class CustomMatcher extends TypeSafeMatcher { public static BusinessMatcher hasCode(String item) { return new BusinessMatcher(item); } private String foundErrorCode; private final String expectedErrorCode; private CustomMatcher(String expectedErrorCode) { this.expectedErrorCode = expectedErrorCode; } @Override protected boolean matchesSafely(final CustomException exception) { foundErrorCode = exception.getErrorCode(); return foundErrorCode.equalsIgnoreCase(expectedErrorCode); } @Override public void describeTo(Description description) { description.appendValue(foundErrorCode) .appendText(" was not found instead of ") .appendValue(expectedErrorCode); } } NOTE: Please visit https://github.com/mike-ensor/custom-exception-testing to get a copy of a working Hamcrest Matcher, JUnit @Rule and ExpectedException. And there you have it, a quick overview of different ways to test Exceptions thrown by your code along with the ability to test for specific messages and fields from within custom exception classes. Please be specific with your test cases and try to target the exact case you have setup for your test, remember, tests can save you from introducing side-effect bugs!

A couple of times recently I’ve needed to set up a .Net application to use Registration-Free COM, and each time I’ve had to hunt around to recall the details. Furthermore, just this week I needed to write some unit tests that involve instantiating these un-registered COM objects, and that wasn’t straightforward. So, as much for the benefit of my future self as for you, my loyal reader, I’m going to summarise my know-how in quick blog post before it becomes used-to-know-how. What is Registration-Free COM? If you’re still reading, I’ll assume you know all about COM, Microsoft’s ancient technology for enabling components written in different languages to talk to each other (I wrote a little about it here, with some links to introductory articles). You are probably also aware of DLL Hell. That isn’t a place where bad executables are sent when they are terminated. Rather, it was a pain inflicted on developers by the necessity of registering COM components (and other DLLs) in a central place in the OS. Since all components were dumped into the same pool, one application could cause all kinds of hell for others by registering different versions of shared DLLs. The OS doesn’t police this pool, and it certainly doesn’t enforce compatibility, so much unexpected weird and wonderful behaviour was the result. Starting with Windows XP, it has been possible to more-or-less escape this hell by not registering components in a central location, and instead using Registration-Free COM. This makes it much easier to deploy applications, because you can just copy a bunch of files – RegSvr32 is not involved, and there are no Registry keys to be written. You can be confident that your application will have no impact on others once installed. It is all done using manifests. Individual Manifest Files For each dll, or ocx file (or ax files in my case – I’m working with DirectShow filters) containing COM components you need to create a manifest. Suppose your dll is called MyCOMComponent.dll. Your manifest file should be called MyCOMComponent.sxs.manifest, and it should contain the following: Obviously you need to make sure that the clsid inside comClass is correct for your component. If you have more than one COM object in your dll you can add multiple comClass elements. For those not wanting to generate these manifests by hand, a StackOverflow answer lists some tools that might help. About Deployment When you deploy your application you should deploy both the dll/ocx/ax file and its manifest into the same directory as your .Net exe/dlls. When developing in Visual Studio, I customise the build process to make sure all these dlls get copied into the correct place for running and debugging the application. I stole the technique for doing this from the way ASP.Net MVC applications manage their dlls. Put all the dlls and manifests into a folder called _bin_deployableAssemblies alongside the rest of your source code. Then modify your csproj file and add the following Target at the end of it: To make sure that target is called when you build, update the AfterBuild target (uncomment it first if you’re not currently using it): The Application Manifest Now you need to make sure your application declares its dependencies. First add an app.manifest file to your project, if you haven’t already got one. To do this in Visual Studio, right click the project, select Add –> New Item … and then choose Application Manifest File. Having added the manifest, you need to ensure it is compiled into your executable. You do this by right-clicking the project, choosing Properties, then going to the Application tab. In the resources section you’ll see a Manifest textbox: make sure your app.manifest file is selected. Now you need to add a section to the app.manifest file for each dependency. By default your app.manifest file will probably already have a dependency for the Windows Common Controls. After that (so, nested directly inside the root element) you should add the following for each of the manifest files you created earlier: Notice that we drop the “.manifest” off the end of the manifest file name when we refer to it here. The other important thing is that the version number here and the one in the manifest file should exactly match, though I don’t think there’s any reason to change it from 1.0.0.0. Disabling the Visual Studio Hosting Process There’s just one more thing to do before you try running your application, and that is to turn off the Visual Studio hosting process. The hosting process apparently helps improve debugging performance, amongst other things (though I’ve not noticed greatly decreased performance with it disabled). The problem is that, when enabled, application executables are not loaded directly- rather, they are loaded by an intermediary executable with a name ending .vshost.exe. The upshot is that the manifest embedded in your exe is ignored, and COM components are not loaded. Disabling the hosting process is simple: go to the Debug tab of your project’s Properties and uncheck “Enable the Visual Studio hosting process” With everything set up, you’ll want to try running your application. If you got everything right first time, everything will go smoothly. If not you might see an error like this: If you do, check Windows’ Application event log for errors coming from SideBySide. These are usually pretty helpful in telling you which part of your configuration has a problem. Summary To re-cap briefly, here are the steps to enabling Registration-Free COM for you application: Create a manifest file for each COM dll Make sure both COM dlls and manifest files are deployed alongside your main executable Add a manifest file to your executable which references each individual manifest file Make sure you turn off the Visual Studio hosting process before debugging Unit Testing and Registration-Free COM And now, as promised, a word about running Unit Tests when Registration-Free COM is involved. If you have a Unit Test which tries to create a Registration-Free COM object you’ll probably get an exception like Retrieving the COM class factory for component with CLSID {1C123B56-3774-4EE4-A482-512B3AB7CABB} failed due to the following error: 80040154 Class not registered (Exception from HRESULT: 0x80040154 (REGDB_E_CLASSNOTREG)). If you don’t get this error, it’s probably because the component is still registered centrally on your machine. Running regsvr32 /u [Path_to_your_dll] will unregister it. Why do Unit Tests fail, when the application works? It is for the same reason that the Visual Studio hosting process breaks Registration-Free COM: your unit tests are actually being run in a different process (for example, the Resharper.TaskRunner), and the manifest file which you so carefully crafted for your exe is being ignored. Only the manifest on the entry executable is taken into account, and since that’s a generic unit test runner it says nothing about your COM dependencies. But there’s a workaround. Win32 has some APIs –the Activation Context APIs- which allow you to manually load up a manifest for each thread which needs to create COM components. Spike McLarty has written some code to make these easy to use from .Net, and I’ll show you a technique to incorporate this into your code so that it works correctly whether called from unit tests or not. Here’s Spike’s code, with a few minor modifications of my own: /// /// Code from http://www.atalasoft.com/blogs/spikemclarty/february-2012/dynamically-testing-an-activex-control-from-c-and /// class ActivationContext { static public void UsingManifestDo(string manifest, Action action) { UnsafeNativeMethods.ACTCTX context = new UnsafeNativeMethods.ACTCTX(); context.cbSize = Marshal.SizeOf(typeof(UnsafeNativeMethods.ACTCTX)); if (context.cbSize != 0x20) { throw new Exception("ACTCTX.cbSize is wrong"); } context.lpSource = manifest; IntPtr hActCtx = UnsafeNativeMethods.CreateActCtx(ref context); if (hActCtx == (IntPtr)(-1)) { throw new Win32Exception(Marshal.GetLastWin32Error()); } try // with valid hActCtx { IntPtr cookie = IntPtr.Zero; if (!UnsafeNativeMethods.ActivateActCtx(hActCtx, out cookie)) { throw new Win32Exception(Marshal.GetLastWin32Error()); } try // with activated context { action(); } finally { UnsafeNativeMethods.DeactivateActCtx(0, cookie); } } finally { UnsafeNativeMethods.ReleaseActCtx(hActCtx); } } [SuppressUnmanagedCodeSecurity] internal static class UnsafeNativeMethods { // Activation Context API Functions [DllImport("Kernel32.dll", SetLastError = true, EntryPoint = "CreateActCtxW")] internal extern static IntPtr CreateActCtx(ref ACTCTX actctx); [DllImport("Kernel32.dll", SetLastError = true)] [return: MarshalAs(UnmanagedType.Bool)] internal static extern bool ActivateActCtx(IntPtr hActCtx, out IntPtr lpCookie); [DllImport("kernel32.dll", SetLastError = true)] [return: MarshalAs(UnmanagedType.Bool)] internal static extern bool DeactivateActCtx(int dwFlags, IntPtr lpCookie); [DllImport("Kernel32.dll", SetLastError = true)] internal static extern void ReleaseActCtx(IntPtr hActCtx); // Activation context structure [StructLayout(LayoutKind.Sequential, Pack = 4, CharSet = CharSet.Unicode)] internal struct ACTCTX { public Int32 cbSize; public UInt32 dwFlags; public string lpSource; public UInt16 wProcessorArchitecture; public UInt16 wLangId; public string lpAssemblyDirectory; public string lpResourceName; public string lpApplicationName; public IntPtr hModule; } } } The method UsingManifestDo allows you to run any code of your choosing with an Activation Context loaded from a manifest file. Clearly we only need to invoke this when our code is being called from a Unit Test. But how do we structure code elegantly so that it uses the activation context when necessary, but not otherwise? Here’s my solution: public static class COMFactory { private static Func, object> _creationWrapper = function => function(); public static T CreateComObject() where T:new() { var instance = (T)_creationWrapper(() => new T()); return instance; } public static object CreateComObject(Guid guid) { Type type = Type.GetTypeFromCLSID(guid); var instance = _creationWrapper(() => Activator.CreateInstance(type)); return instance; } public static void UseManifestForCreation(string manifest) { _creationWrapper = function => { object result = null; ActivationContext.UsingManifestDo(manifest, () => result = function()); return result; }; } } Whenever I need to create a COM Object in my production code, I do it by calling COMFactory.CreateCOMObject. By default this will create the COM objects directly, relying on the manifest which is embedded in the executable. But in my Test project, before running any tests I call COMFactory.UseManifestForCreation and pass in the path to the manifest file. This ensures that the manifest gets loaded up before we try to create any COM objects in the tests. To avoid duplicating the manifest file, I share the same file between my Test project and main executable project. You can do this right clicking your test project, choosing Add->Existing Item… then app.manifest in your main project. Finally, click the down arrow on the Add split button, and choose Add as Link. If you’ve got any tips to share on using Registration-Free COM, whether in Unit Tests or just in applications, please do leave a comment.

People tell me A and B, They tell me how I have to see, Things that I have seen already clear, So they push me then from side to side (I Want Out - Helloween) Developing an application that uses RESTful web API may imply developing server and client side. Writing integration tests for the server side can be as easy as using Arquillian to start up server and REST-assured to test that the services works as expected. The problem is how to test the client side. In this post, we are going to see how to test the client side apart from using mocks. As a brief description, to test the client side, what we need is a local server which can return recorded JSON responses. The rest-client-driver is a library which simulates a RESTful service. You can set expectations on the HTTP requests you want to receive during a test. So it is exactly what we need for our java client side. Note that this project is really helpful to write tests when we are developing RESTful web clients for connecting to services developed by third parties like Flickr Rest API, Jira Rest API, Github ... First thing to do is adding rest-client-driver dependency: com.github.rest-driver rest-client-driver 1.1.27 test Next step we are going to create a very simple Jersey application which simply invokes a get method to required URI. public class GithubClient { private static final int HTTP_STATUS_CODE_OK = 200; private String githubBaseUri; public GithubClient(String githubBaseUri) { this.githubBaseUri = githubBaseUri; } public String invokeGetMethod(String resourceName) { Client client = Client.create(); WebResource webResource = client.resource(githubBaseUri+resourceName); ClientResponse response = webResource.type("application/json") .accept("application/json").get(ClientResponse.class); int statusCode = response.getStatus(); if(statusCode != HTTP_STATUS_CODE_OK) { throw new IllegalStateException("Error code "+statusCode); } return response.getEntity(String.class); } } And now we want to test that invokeGetMethod really gets the required resource. Let's suppose that this method in production code will be responsible of getting all issues name from a project registered on github. Now we can start to write the test: @Rule public ClientDriverRule driver = new ClientDriverRule(); @Test public void issues_from_project_should_be_retrieved() { driver.addExpectation( onRequestTo("/repos/lordofthejars/nosqlunit/issues"). withMethod(Method.GET), giveResponse(GET_RESPONSE)); GithubClient githubClient = new GithubClient(driver.getBaseUrl()); String issues = githubClient.invokeGetMethod("/repos/lordofthejars/nosqlunit/issues"); assertThat(issues, is(GET_RESPONSE)); } We use ClientDriverRule @Rule annotation to add the client-driver to a test. And then using methods provided by RestClientDriver class, expectations are recorded. See how we are setting the base URL using driver.getBaseUrl() With rest-client-driver we can also record http status response using giveEmptyResponse method: @Test(expected=IllegalStateException.class) public void http_errors_should_throw_an_exception() { driver.addExpectation( onRequestTo("/repos/lordofthejars/nosqlunit/issues") .withMethod(Method.GET), giveEmptyResponse().withStatus(401)); GithubClient githubClient = new GithubClient(driver.getBaseUrl()); githubClient.invokeGetMethod("/repos/lordofthejars/nosqlunit/issues"); } And obviously we can record a put action: Note that in this example, we are setting that our request should contain given message body to response a 204 status code. This is a very simple example, but keep in mind that also works with libraries like gson or jackson. Also rest-driver project comes with a module that can be used to assert server responses (like REST-assured project) but this topic will be addressed into another post. I wish you have found this post useful. We keep learning, Alex.

when writing testable code, your first port of call is often to abstract any dependencies and make them easy to mock. this is the same for any of your codebase that talks to ftp servers. testing the way your code behaves under real world conditions makes integration tests important regardless of abstraction, though. here’s a simple trick to test ftp code in the wild. a recent project of mine has involved writing code that talks to ftp servers with the goal of adding additional continuous integration automation to a project. although all of my main methods are easily abstracted and injectable, my project still needs to actually talk to ftp servers at the end of the day, and i need to test that these very methods do the right thing when they are met with different conditions; be they bad credentials, lack of read/write permissions etc. the challenge integration tests can be brittle at the best of times, so ensuring that they are repeatable and can be setup and torn down can often be almost as much of a challenge as writing your actual code itself. an ftp server is usually a static service that is installed on a server. you might think that running one and ensuring it stays up and doesn’t get hacked just so that all your integration tests work is a necessary evil, but there is an easier way. run local. run often. i was running an ftp server on my build server just so that it was “always around” for my tests until i stumbled across an interesting project over on github to do just this . the approach i'm about to show you doesn’t need you to go to the effort of running a dedicated server at all. all you need to do is add a single executable to your unit test project and wrap your unit test in a using statement. the ftp server executable is a single file ftp server called ftpdmin which offers a read/write ftp server that can be fired up from the command line with a minimum feature set and only a few command line parameters to make it all tick. by implementing idisposable the helper class that wraps around this command line exe allows you to take advantage of the using() pattern to take care of your executable’s lifetime and have it die when your code is done testing. steps to make it happen download ftpdmin from here . add the exe to the root of your test project (you can put this anywhere, but you’ll have to update the helper class below). now add the exe to your project (i.e “view all items” in your test project’s solution explorer, and add the exe). set the exe to “copy always” in it’s solution properties. add the following code to a helper class in your test project: public class ftptestserver: idisposable { private readonly process ftpprocess; public ftptestserver(string rootdirectory, int port = 21, bool allowuploads = true) { var psinfo = new processstartinfo { filename = appdomain.currentdomain.basedirectory + "\\ftpdmin.exe", arguments = string.format("-p {0} -ha 127.0.0.1 \"{1}\" {2}", port, rootdirectory, allowuploads ? string.empty : "-g"), windowstyle = processwindowstyle.hidden }; ftpprocess = process.start(psinfo); } public void dispose() { if (ftpprocess.hasexited) return; ftpprocess.kill(); ftpprocess.waitforexit(); } } now you can enjoy being able to write really clean integration testing code that starts and ftp server every time you run your tests and then tear it down when your test is done. an example integration test showing connecting to “127.0.0.1”: [testmethod] public void ftpcode_upload_canconnect() { try { // fire up a new ftp server instance using (new ftptestserver(rootdirectory: "./")) { // code that talks to an ftp server on 127.0.0.1 } } catch (webexception e) { assert.fail("failed to connect to our ftp server"); } } how awesome is that? the power of using ftpdmin is that it can be told to deny write permissions to simulate bad user permissions as well: [testmethod] public void ftpcode_upload_throwswebexception() { try { // fire up a new ftp server instance using (new ftptestserver(rootdirectory: "./", allowuploads: false)) { // code that talks to an ftp server on 127.0.0.1 } } catch (webexception e) { assert.fail("our code failed to upload a file because of invalid permissions"); } } all in all, the above has been a complete lifesaver when it comes to making my integration test projects portable – if a new developer joins my project, they instantly get access to my ftp test harness just by pulling down my project’s source code.

Everyone is thinking why in the world would anyone pick static, when you can be dynamic? Usually the thought process is, "what language am I most proficient in, that can do the job." Totally not a bad way to go about it. Now does this choice affect anything else? Testing? Speed of development? Robustness? Dynamic vs. Static Dynamic languages are languages that don’t necessarily need variables to be declared before they are used. Examples of dynamic languages are Python, Ruby, and PHP. So in dynamic languages the following is possible: num = 10 We have successfully assigned a value to variable without declaring it before hand. Simple enough, try doing this in Java (you can’t). This can *increase* development speed, without having to write boilerplate code. This can somewhat be a double edge sword, since dynamic languages types are checked during runtime, there is no way to tell if there is a bug in code until it is run. I know you can test, but you can’t test for everything. You can’t test for everything. Here is an example albeit trivial. def get_first_problem(problems): for problem in problems: problam = problem + 1 return problam Now if you are raging to some serious dubstep, its easy enough to miss that small typo, you go screw it and do it live, and deploy to production. Python will simply create the new variable and not a single thing will be said. Only you can stop bugs in production! Static languages are languages that variables need to be declared before use and type checking is done at compile time. Examples of static languages include Java, C, and C++. So in static languages the following is enforced static int awesomeNumber; awesomeNumber = 10; Many argue this increases robustness as well as decrease chances of Runtime Errors. Since the compiler will catch those horrible horrible mistakes you made throughout your code. Your methods contracts are tighter, downside to this is crap ton of boilerplate code. Weak and Strong Typing can be often be confused with dynamic and static languages. Weak typed languages can lead to philosophical questions like what does the number 2 added to the word ‘two’ give you? Things like this are possible with a weak typed language. a = 2 b = "2" concatenate(a, b) // Returns "22" add(a, b) // Returns 4 Traditionally languages may place restriction on what transaction may occur for example in a strong typed language adding a string and integer will result in a type error as shown below. >>> a = 10 >>> b = 'ten' >>> a + b Traceback (most recent call last): File "", line 1, in TypeError: unsupported operand type(s) for +: 'int' and 'str' >>> Conclusion Regardless of where you land on this discussion, claiming one is better than the other would lead to flame war, but there are places where each is strong. Dynamic languages are good for fast quick development cycles and prototyping, while static languages are better suited to longer development cycles where trivial bugs could be extremely costly (telecommunication systems, air traffic control). For example if some giant company called Moo Corp. spent millions of dollars on QA and Testing and a bug somehow gets into the field, to fix it would mean another round of testing. When sitting in that chair the choice is clear static languages FTW, its a hard job but someone has to milk the cows. Test, test, and test. Just a little food for thought, for when you are starting your next project. You never know what limitations you maybe placing on yourself and your team. What do you do consider when selecting a programming language for a project?

the test was run locally (no network involved ) on a lenovo w520 laptop with 8 cores & 8 gb ram with an ssd card. the storage engine we used was esent, safe transactions. default ravendb configuration, running in console, with logging disabled. we took the most obvious approach both in the code we wrote and the test approach. i am pretty sure that i’ll get a lot of helpful suggestions about the load testing. the code is available here, and you are more than welcome to take it for a spin and get your own results. what is important for me to note is that we have done exactly zero performance tuning . that is relevant to both the index we use, to the code that we wrote, everything. i just wrote things down, and didn’t worry about performance, even though this code is going to go through a load test. why don’t i worry about it? because ravendb is setup to do the right thing. it will self optimize itself without you need to take care of that. with that said, here are the test results: you can see that the red line is the number of users we have, and we have this worrying green line that seems to go crazy… except that this is actually the number of page served. the part that we care about is actually the avg. page time, and that is the blue line. this line , however, is basically flat no matter the load. here are the test results in details load test summary test run information load test name loadtest1 description start time 04/09/12 14:16:38 end time 04/09/12 14:21:38 warm-up duration 00:00:20 duration 00:05:00 controller local run number of agents 1 run settings used run settings1 overall results max user load 300 tests/sec 20.0 tests failed 0 avg. test time (sec) 12.5 transactions/sec 0 avg. transaction time (sec) 0 pages/sec 77.1 avg. page time (sec) 0.0062 requests/sec 77.1 requests failed 0 requests cached percentage 0 avg. response time (sec) 0.0062 avg. content length (bytes) 3,042 key statistic: top 5 slowest pages url (link to more details) 95% page time (sec) 0.018 0.018 0.014 0.014 0.014 key statistic: top 5 slowest tests name 95% test time (sec) browsing 19.3 browseandsearch 17.6 10.6 6 test results name scenario total tests failed tests (% of total) avg. test time (sec) browsing load 1,533 0 (0) 16.0 browseandsearch load 1,685 0 (0) 15.0 load 2,770 0 (0) 9.00 6 page results url (link to more details) scenario test avg. page time (sec) count load browsing 0.0072 1,629 load browseandsearch 0.0071 1,783 load browseandsearch 0.0064 3,443 load searching 0.0064 2,798 load browsing 0.0063 1,617 load browsing 0.0063 1,580 load browseandsearch 0.0063 1,760 load searching 0.0055 2,810 load searching 0.0055 2,839 load searching 0.0054 2,866 6 transaction results name scenario test response time (sec) elapsed time (sec) count 6 system under test resources machine name % processor time available memory at test completion (mb) 6 controller and agents resources machine name % processor time available memory at test completion (mb) 13.0 1,356 6 errors type subtype count last message you can dig in and look at the data, it is quite interesting. under the load of 300 users, the average page response time was… 0.0062 seconds. and ravendb was using just 13% of the cpu, and that include running the agents running the tests. after seeing how well ravendb does in perf testing, i decided to take it up a notch. starting from 10 users, with a step duration of 1 sec, add 50 users for each step, all the way to 3,000. start with a warm up period of 20 seconds, then run the test for 10 minutes. let us see what happens, okay? just to be clear, this is a ravendb application running with three thousands concurrent users , on an off the shelve laptop while i was busy doing other stuff. one word of warning before hand, because i run everything on a single machine, just running so many users on the machine significantly slowed down how ravendb is reacting. basically, the code for managing the perf test took so many resources that ravendb had to fight to get some to actually answer the queries. scared yet, because here are the results in graph form. now you can actually see that we have some fluctuations in the graphs, the number of users grows and grows until we get to 3,000 and we have 0.37 seconds response times. again, i remind you, we have done zero optimizations and this is idiomatic ravendb code. and we were able to serve requests at a frankly pretty amazing rate of speed. and here are they in their full details: load test summary test run information load test name loadtest1 description start time 04/09/12 15:28:48 end time 04/09/12 15:38:48 warm-up duration 00:00:20 duration 00:10:00 controller local run number of agents 1 run settings used load overall results max user load 3,000 tests/sec 196 tests failed 0 avg. test time (sec) 14.3 transactions/sec 0 avg. transaction time (sec) 0 pages/sec 741 avg. page time (sec) 0.37 requests/sec 741 requests failed 0 requests cached percentage 0 avg. response time (sec) 0.37 avg. content length (bytes) 3,080 key statistic: top 5 slowest pages url (link to more details) 95% page time (sec) 0.83 0.82 0.82 0.82 0.81 key statistic: top 5 slowest tests name 95% test time (sec) browsing 20.8 browseandsearch 19.8 12.9 6 test results name scenario total tests failed tests (% of total) avg. test time (sec) browsing load 31,843 0 (0) 17.4 browseandsearch load 33,989 0 (0) 16.8 load 51,650 0 (0) 10.8 6 page results url (link to more details) scenario test avg. page time (sec) count load browsing 0.40 32,338 load searching 0.39 52,597 load browsing 0.39 32,627 load browseandsearch 0.39 68,576 load browsing 0.38 32,803 load searching 0.38 52,283 load browseandsearch 0.37 34,766 load browseandsearch 0.36 34,982 load searching 0.35 51,991 load searching 0.33 51,846 6 transaction results name scenario test response time (sec) elapsed time (sec) count 6 system under test resources machine name % processor time available memory at test completion (mb) 6 controller and agents resources machine name % processor time available memory at test completion (mb) 85.4 1,203 6 errors type subtype count last message note that the reason fro the high cpu usage is that the tests and ravendb were running on the same machine. the code for the entire series can be found here: https://github.com/ayende/nuget.perf no, i’ll not do a similar sql version, if you want to, i would be very interested in seeing one, but that isn’t something that i intend to do. yes, it is a simple and trivial implementation, but that was pretty much the whole point. being able to get to that scale without actually doing anything special is what we strive for in ravendb.

Why create a mechanism to expect a test failure? There comes a time when one would want and expect a JUnit @Test case fail. Though this is pretty rare, it happens. I had the need to detect when a JUnit Test fails and then, if expected, to pass instead of fail. The specific case was that I was testing a piece of code that could throw an Assert error inside of a call of the object. The code was written to be an enhancement to the popular new Fest Assertions framework, so in order to test the functionality, one would expect test cases to fail on purpose. A Solution One possible solution is to utilize the functionality provided by a JUnit @Rule in conjunction with a custom marker in the form of an annotation. . Why use a @Rule? @Rule objects provide an AOP-like interface to a test class and each test cases. Rules are reset prior to each test case being run and they expose the workings of the test case in the style of an @Around AspectJ advice would. Required code elements @Rule object to check the status of each @Test case @ExpectedFailure custom marker annotation Test cases proving code works! Optional specific exception to be thrown if annotated test case does not fail NOTE: working code is available on my github page and will soon be in Maven Central. Feel free to Fork the project and submit a pull request Example Usage In this example, the "exception" object is a Fest assertion enhanced ExpectedException (look for my next post to expose this functionality). The expected exception will make assertions and in order to test those, the test case must be marked as @ExpectedFailure public class ExceptionAssertTest { @Rule public ExpectedException exception = ExpectedException.none(); @Rule public ExpectedTestFailureWatcher watcher = ExpectedTestFailureWatcher.instance(); @Test @ExpectedFailure("The matcher should fail becasue exception is not a SimpleException") public void assertSimpleExceptionAssert_exceptionIsOfType() { // expected exception will be of type "SimpleException" exception.instanceOf(SimpleException.class); // throw something other than SimpleException...expect failure throw new RuntimeException("this is an exception"); } } Implementation of Solution Reminder, the latest code is available on my github page. @Rule code (ExpectedTestFailureWatcher.java) import org.junit.rules.TestRule; import org.junit.runner.Description; import org.junit.runners.model.Statement; // YEAH Guava!! import static com.google.common.base.Strings.isNullOrEmpty; public class ExpectedTestFailureWatcher implements TestRule { /** * Static factory to an instance of this watcher * * @return New instance of this watcher */ public static ExpectedTestFailureWatcher instance() { return new ExpectedTestFailureWatcher(); } @Override public Statement apply(final Statement base, final Description description) { return new Statement() { @Override public void evaluate() throws Throwable { boolean expectedToFail = description.getAnnotation(ExpectedFailure.class) != null; boolean failed = false; try { // allow test case to execute base.evaluate(); } catch (Throwable exception) { failed = true; if (!expectedToFail) { throw exception; // did not expect to fail and failed...fail } } // placed outside of catch if (expectedToFail && !failed) { throw new ExpectedTestFailureException(getUnFulfilledFailedMessage(description)); } } /** * Extracts detailed message about why test failed * @param description * @return */ private String getUnFulfilledFailedMessage(Description description) { String reason = null; if (description.getAnnotation(ExpectedFailure.class) != null) { reason = description.getAnnotation(ExpectedFailure.class).reason(); } if (isNullOrEmpty(reason)) { reason = "Should have failed but didn't"; } return reason; } }; } } @ExpectedFailure custom annotation (ExpectedFailure.java) import java.lang.annotation.*; /** * Initially this is just a marker annotation to be used by a JUnit4 Test case in conjunction * with ExpectedTestFailure @Rule to indicate that a test is supposed to be failing */ @Documented @Retention(RetentionPolicy.RUNTIME) @Target(value = ElementType.METHOD) public @interface ExpectedFailure { // TODO: enhance by adding specific information about what type of failure expected //Class assertType() default Throwable.class; /** * Text based reason for marking test as ExpectedFailure * @return String */ String reason() default ""; } Custom Exception (Optional, you can easily just throw RuntimeException or existing custom exception) public class ExpectedTestFailureException extends Throwable { public ExpectedTestFailureException(String message) { super(message); } } Can't one exploit the ability to mark a failure as expected? With great power comes great responsibility, it is advised that you do not mark a test as being @ExpectedFailure if you do not understand exactly why the test if failing. It is recommended that this testing method be implemented with care. DO NOT use the @ExpectedFailure annotation as an alternative to @Ignore Possible future enhancements could include ways to specify the specific assertion or the specific message asserted during the test case execution. Known issues In this current state, the @ExpectedFailure annotation can cover up additional assertions and until the future enhancements have been put into place, it is advised to use this methodology wisely.

Fixing a problem usually starts with reproducing it – what Steve McConnell calls “stabilizing the error.” Technically speaking, you can’t be sure you are fixing the problem unless you can run through the same steps, see the problem happen yourself, fix it, and then run through the same steps and make sure that the problem went away. If you can’t reproduce it, then you are only guessing at what’s wrong, and that means you are only guessing that your fix is going to work. But let’s face it – it’s not always practical or even possible to reproduce a problem. Lots of bug reports don’t include enough information for you to understand what the hell the problem actually was, never mind what was going on when the problem occurred – especially bug reports from the field. Rahul Premraj and Thomas Zimmermann found in The Art of Collecting Bug Reports (from the book Making Software), that the two most important factors in determining whether a bug report will get fixed or not are: Is the description well-written, can the programmer understand what was wrong or why the customer thought something was wrong? Does it include steps to reproduce the problem, even basic information about what they were doing when the problem happened? It’s not a lot to ask – from a good tester at least. But you can’t reasonably expect this from customers. There are other cases where you have enough information, but don’t have the tools or expertise to reproduce a problem – for example, when a pen tester has found a security bug using specialist tools that you don’t have or don’t understand how to use. Sometimes you can fix a problem without being able to see it happen in front of you, come up with a theory on your own, trusting your gut – especially if this is code that you recently worked on. But reproducing the problem first gives you the confidence that you aren’t wasting your time and that you actually fixed the right issue. Trying to reproduce the problem should almost always be your first step. What’s involved in reproducing a bug? What you want to do is to find, as quickly as possible, a simple test that consistently shows the problem, so that you can then run a set of experiments, trace through the code, isolate what’s wrong, and prove that it went away after you fixed the code. The best explanation that I’ve found of how to reproduce a bug is in Debug It! where Paul Butcher patiently explains the pre-conditions (identifying the differences between your test environment and the customer’s environment, and trying to control as many of them as possible), and then how to walk backwards from the error to recreate the conditions required to make the problem happen again. Butcher is confident that if you take a methodical approach, you will (almost) always be able to reproduce the problem successfully. In Why Programs Fail: A guide to Systematic Debugging, Andreas Zeller, a German Comp Sci professor, explains that it’s not enough just to make the problem happen again. Your goal is to come up with the simplest set of circumstances that will trigger the problem – the smallest set of data and dependencies, the simplest and most efficient test(s) with the fewest variables, the shortest path to making the problem happen. You need to understand what is not relevant to the problem, what’s just noise that adds to the cost and time of debugging and testing – and get rid of it. You do this using binary techniques to slice up the input data set, narrowing in on the data and other variables that you actually need, repeating this until the problem starts to become clear. Code Complete’s chapter on Debugging is another good guide on how to reproduce a problem following a set of iterative steps, and how to narrow in on the simplest and most useful set of test conditions required to make the problem happen; as well as common places to look for bugs: checking for code that has been changed recently, code that has a history of other bugs, code that is difficult to understand (if you find it hard to understand, there’s a good chance that the programmers who worked on it before you did too). Replay Tools One of the most efficient ways to reproduce a problem, especially in server code, is by automatically replaying the events that led up to the problem. To do this you’ll need to capture a time-sequenced record of what happened, usually from an audit log, and a driver to read and play the events against the system. And for this to work properly, the behavior of the system needs to be deterministic – given the same set of inputs in the same sequence, the same results will occur each time. Otherwise you’ll have to replay the logs over and over and hope for the right set of circumstances to occur again. On one system that I worked on, the back-end engine was a deterministic state machine designed specifically to support replay. All of the data and events, including configuration and control data and timer events, were recorded in an inbound event log that we could replay. There were no random factors or unpredictable external events – the behavior of the system could always be recreated exactly by replaying the log, making it easy to reproduce bugs from the field. It was a beautiful thing, but most code isn’t designed to support replay in this way. Recent research in virtual machine technology has led to the development of replay tools to snapshot and replay events in a virtual machine. VMWare Workstation, for example, included a cool replay debugging facility for C/C++ programmers which was “guaranteed to have instruction-by-instruction identical behavior each time.” Unfortunately, this was an expensive thing to make work, and it was dropped in version 8, at the end of last year. Replay Solutions provides replay for Java programs, creating a virtual machine to record the complete stream of events (including database I/O, network I/O, system calls, interrupts) as the application is running, and then later letting you simulate and replay the same events against a copy of the running system, so that you can debug the application and observe its behavior. They also offer similar application record and replay technology for mobile HTML5 and JavaScript applications. This is exciting stuff, especially for complex systems where it is difficult to setup and reproduce problems in different environments. Fuzzing and Randomness If the problem is non-deterministic, or you can't come up with the right set of inputs, one approach to try is to simulate random data inputs and watch to see what happens - hoping to happen on a set of input variables that will trigger the problem. This is called fuzzing. Fuzzing is a brute force testing technique that is used to uncover data validation weaknesses that can cause reliability and security problems. It's effective at finding bugs, but it’s a terribly inefficient way to reproduce a specific problem. First you need to setup something to fuzz the inputs (this is easy if a program is reading from a file, or a web form – there are fuzzing tools to help with this – but a hassle if you need to write your own smart protocol fuzzer to test against internal APIs). Then you need time to run through all of the tests (with mutation fuzzing, you may need to run tens of thousands or hundreds of thousands of tests to get enough interesting combinations) and more time to sift through and review all of the test results and understand any problems that are found. Through fuzzing you will get new information about the system to help you identity problem areas in the code, and maybe find new bugs, but you may not end up any closer to fixing the problem that you started on. Reproducing problems, especially when you are working from a bad bug report (“the system was running fine all day, then it crashed… the error said something about a null pointer I think?”) can be a serious time sink. But what if you can’t reproduce the problem at all? Let’s look at that next…

Test-Driven Development is a code-level practice, based on running automated tests that are written before the production code they exercise. But practices can be applied only in the context where they were developed: when some premises are not present is difficult to apply TDD as-is. Automated specification For example, consider the premise of assertion automation: it is possible to write a (hopefully) small algorithm that is able to check the result of running production code and return true or false. In the case the problem is: Draw an antialiased circle on this blank canvas. -- Carlo Pescio it is not immediately clear how to define automated tests for this behavior. We could check that some pixels are still blank inside or outside the circle, or that there is a bound number of pixels of black color; or even that they are contiguous. An opinion I've heard (that I try not to misrepresent) is that we only need to write some looser tests in these cases, checking only a few pixels of the circle. This process will give us a little feedback on the API of our Canvas or Circle object, but not much on the algorithm we are implementing inside it. Are we going in the right direction? Have new test cases correctly been satisfied without a large intervention on the existing code? Are we painting some unrelated pixels due to an hidden bug? What I argument here is instead that we should change the nature of the feedback mechanism. Speaking in control theory terms, change the block that acquires the output and influences the input to our design process. Develop in the browser When I was developing a Couchapp, a kind of web application served directly from a CouchDB database, I was appaled by the difficulty of testing it. While the production code was composed of ~100 lines, it was a complex mix of technologies: HTML and CSS code, client-side JavaScript for managing user events and some server-side JavaScript for the "queries" (actually the server-side only consists of the database in Couchapps.) Some of this logic could be tested in automation, like the result of queries over views. Yet much of it was related to a user interface, and as such requiring a large time investment to automate. Instead of waking up my Selenium server and start to manipulate a browser with code, I noticed that this UI was almost read-only; there were a few cases where a new document would have to be inserted, but a manual test of them was short and did not even required to reload the page. The whole application state was observable. Summing it up, I performed a frequent manual test that took a few seconds instead of trying to define complex and brittle automation logic for testing the UI. Now that I've been introduced to a simple qualitative ROI model by Carlo Pescio's article, I would do the same for every context where: a large time investment is needed for automating tests. it is possible to perform manual tests quickly. as the only logic conclusion. A word of caution TDD has many benefits (including catching regressions early) so I'm not prepared to give it up just because it is difficult to test. These are technical scenarios where I have successfully followed TDD by the book: multithreaded and multiprocess code applications distributed over multiple machines computer vision (object recognition and tracking) image manipulation code (via comparison testing) development of browser bindings for Selenium And even in the case the big picture is not easy to test-first (like in the case of image manipulation), we can benefit from TDD the pieces of the solution. For example, in the computer vision case I wasn't able to write a test beforehand for tracking a car inside a movie. But I was able to TDD the objects that the algorithmic solution to the problem called for: Patch, Area, Cluster, Movement, and so on. End-to-end TDD is not always cheap but unit level TDD can often be, if it considers testability as a relevant property (while regression testing even at the end-to-end level is always possible, in the worst case with record and replay.) End-to-end specifications If we can't define automated assertions for our "big picture" problem, it doesn't mean that we cannot apply the TDD approach, by substituting a manual step. Going back to the circle problem, I would define manual test cases on an inspection page seen by a human. I've seen this done with layouts and multiple browsers to catch CSS rendering bugs, for example: It would be very difficult to check these screenshots automatically, as each browser renders pages a bit differently from the others. The iterative process becomes: Define a cheap manual test, automating the arrange and act phases but not the assertion. Write only the code necessary to make it pass. Refactor. As long as the number of tests does not increase without limit and the manual check can be performed quickly, this approach does not slow you down with respect to TDD by-the-book. You'll have to take care of regression with other means; but at least you define a set of manual test cases. Feedback! TDD is an instrument of feedback: if feedback cannot be gathered in an automated way, we have to resort to manual checking of the specifications. Here are other examples of manual tools for generating feedback: Read-Eval-Print Loops: you can experimenting with existing classes and functions, and easily repeat steps thanks to history. the browser refresh button: the fastest way to transform a PSD into an HTML and CSS template. MongoDB console for learning the database API; other kinds of consoles like Firebug and Chrome's, or Clojure's.

At the end of July 2012, Groovy 2.0 was released with support for static type checking and some performance improvements through the use of JDK7 invokedynamic and type inference as a result of type information now available through static typing. I was interested in seeing some estimate as to how significant the performance improvements in Groovy 2.0 have turned out and how Groovy 2.0 would now compare to Java in terms of performance. In case the performance gap had become minor, or at least acceptable, in the meantime, it would certainly be time to take a serious look at Groovy. Groovy has been ready for production for a long time. So, let's see whether it can compare with Java in terms of performance. The only performance measurement I could find on the Internet was this little benchmark measurment on jlabgroovy. The measurement only consists of calculating Fibonacci numbers with and without the @CompileStatic annotation. That's it; i.e., it's certainly not very meaningful in striving to get an overall impression. I was only interested in obtaining some rough estimate of how Groovy compares to Java as far as performance is concerned. Java performance measurement included Alas, no measurement was included in this little benchmark as to how much time Java takes to calculate Fibonacci numbers. So I "ported" the Groovy code to Java (here it is) and repeated the measurements. All measurements were done on an Intel Core2 Duo CPU E8400 3.00 GHz using JDK7u6 running on Windows 7 with Service Pack 1. I used Eclipse Juno with the Groovy plugin using the Groovy compiler version 2.0.0.xx-20120703-1400-e42-RELEASE. These are the figures I obtained without having a warm-up phase: Groovy 2.0 without @CompileStatic Groovy/Java performance factor Groovy 2.0 with @CompileStatic Groovy/Java performance factor Kotlin 0.1.2580 Java static ternary 4352ms 4.7 926ms 1.0 1005ms 924ms static if 4267ms 4.7 911ms 0.9 1828ms 917ms instance ternary 4577ms 2.7 1681ms 1.8 994ms 917ms instance if 4592ms 2.9 1604ms 1.7 1611ms 969ms I also did measurements with a warm-up phase of various length with the conclusion that there is no benefit for either language with or without the @CompileStatic. Since the Fibonacci algorithm is that recursive the warm-up phase seems to be "included" for any Fibonacci number that is not very small. We can see that the performance improvements due to static typing have made quite a difference. This little comparison does little justice, though. To me, the impression that static typing in Groovy has had in conjunction with type inference has led to significant performance improvements—and in the same way it has led to Groovy++ becoming very strong. With the @CompileStatic, the performance of Groovy is about 1-2 times slower than Java, and without Groovy, it's about 3-5 times slower. Unhappily, the measurements of "instance ternary" and "instance if" are the slowest. Unless we want to create masterpieces in programming with static functions, the measurements for "static ternary" and "static if" are not that relevant for most of the code with the ambition to be object-oriented (based on instances). Conclusion When Groovy was about 10-20 times slower than Java (see benchmark table almost at the end of this article) it is questionable whether the @CompileStatic was used or not. This means to me that Groovy is ready for applications where performance has to be somewhat comparable to Java. Earlier, Groovy (or Ruby, Closure, etc.) could only serve as a plus on your CV because of the performance impediment (at least here in Europe). New JVM kid on the block: Kotlin I added the figures for Kotlin as well (here is the code). Kotlin is a relatively new statically typed JVM-based Java-compatible programming language. Kotlin is more concise than Java by supporting variable type inferences, higher-order functions (closures), extension functions, mixins and first-class delegation, etc. Contrary to Groovy, it is more geared towards Scala, but also integrates well with Java. Kotlin is still under development and has yet to be officially released. So the figures have to be taken with caution as the guys at JetBrains are still working on the code optimization. Ideally, Kotlin should be as fast as Java. The measurements were done with the current "official" release 0.1.2580. And what about future performance improvements? At the time when JDK1.3 was the most recent JDK, I still earned my pay with Smalltalk development. At that time the performance of VisualWorks Smalltalk (now Cincom Smalltalk) and IBM VA for Smalltalk (now owned by Instantiations) was very good comparable to Java. And Smalltalk is a dynamically typed language, like pre-Goovy 2.0 and Ruby, where the compiler cannot make use of type inference to do optimizations. Because of this, it always appeared strange to me that Groovy, Ruby and other JVM-based dynamic languages had such a big performance penalty compared to Java when Smalltalk had not. From that point of view I think there's still room for Groovy performance improvements beyond @CompileStatic.

The more I use dependency injection (DI) in my code, the more it alters the way I see both my design and implementation. Injection is so convenient and powerful that you end up wanting to make sure you use it as often as you can. And as it turns out, you can use it in many, many places. Let’s cover briefly the most obvious scenarios where DI, and more specifically, Guice, are a good fit: objects created either at class loading time or very early in your application. These two aspects are covered by either direct injection or by providers, which allow you to start building some of your object graph before you can inject more objects. I won’t go too much in details about these two use cases since they are explained in pretty much any Guice tutorial you can find on the net. Once the injector has created your graph of objects, you are pretty much back to normal and instantiating your “runtime objects” (the objects you create during the life time of your application) the normal way, most likely with “new” or factories. However, you will quickly start noticing that you need some runtime information to create these objects, other parts of them could be injected. Let’s take the following example: we have a GeoService interface that provides various geolocation functions, such as telling you if two addresses are close to each other: public interface GeoService { /** * @return true if the two addresses are within @param{miles} * miles of each other. */ boolean isNear(Address address1, Address address2, int miles); } Then you have a Person class which uses this service and also needs a name and an address to be instantiated: public class Person { // Fields omitted public Person(String name, Address address, GeoService gs) { this.name = name; this.address = address; this.geoService = gs; } public boolean livesNear(Person otherPerson) { return geoService.isNear(address, otherPerson.getAddress(), 2 /* miles */); } } Something odd should jump at you right away with this class: while name and address are part of the identity of a Person object, the presence of the GeoService instance in it feels wrong. The service is a singleton that is created on start up, so a perfect candidate to be injected, but how can I achieve the creation of a Person object when some of its information is supplied by Guice and the other part by myself? Guice gives you a very elegant and flexible way to implement this scenario with “assisted injection”. The first step is to define a factory for our objects that represents exactly how we want to create them: public interface PersonFactory { Person create(String name, Address address); } Since only name and address participate in the identity of our Person objects, these are the only parameters we need to construct our objects. The other parameters should be supplied by Guice so we modify our Person constructor to let Guice know: @Inject public Person(@Assisted String name, @Assisted Address address, GeoService geoService) { this.name = name; this.address = address; this.geoService = geoService; } In this code, I have added an @Inject annotation on the constructor and an @Assisted annotation on each parameter that I will be providing. Guice will take care of injecting the rest. Finally, we connect the factory to its objects when creating the module: Module module1 = new FactoryModuleBuilder() .build(PersonFactory.class); The important part here is to realize that we will never instantiate PersonFactory: Guice will. From now on, all we need to do whenever we want to instantiate a Person object is to ask Guice to hand us a factory: @Inject private PersonFactory personFactory; // ... Person p = personFactory.create("Bob", new Address("1 Ocean st")); If you want to find out more, take a look at the main documentation for assisted injection, which explains how to support overloaded constructors and also how to create different kinds of objects within the same factory. Wrapping up Let’s take a look at what we did. First, we started with a suspicious looking constructor: public Person(String name, Address address, GeoService s) { This constructor is suspicious because it accepts parameters that do not participate in the identity of the object (you won’t use the GeoService parameter when calculating the hash code of a Person object). Instead, we replaced this constructor with a factory that only accepts identity fields: public interface PersonFactory { Person create(String name, Address address); } and we let Guice’s assisted injection take care of creating a fully formed object for us. This observation leads us to the Identity Constructor rule: If a constructor accepts parameters that are not used to define the identity of the objects, consider injecting these parameters. Once you start looking at your objects with this rule in mind, you will be surprised to find out how many of them can benefit from assisted injection.

DrupalCon Munich is next week, and I am lucky enough to be going. As part of preparing for the conference, I thought it would be worthwhile to see just how easy (or difficult) it would be to migrate an existing Drupal site to Windows Azure Web Sites. So, in this post, I’ll do just that. Fortunately, because Windows Azure Web Sites supports both PHP and MySQL, the migration process is relatively straightforward. And, because Drupal and PHP run on any platform, the process I’ll describe should work for moving Drupal to Windows Azure Web Sites regardless of what platform you are moving from. Of course, Drupal installations can vary widely, so YMMV. I tested the instructions below on relatively small (and simple) Drupal installation running on CentOS 5. (Unfortunately, I won’t be using Drush since it isn’t supported on Windows Azure Websites.) If you are considering moving a large and complex Drupal application, may want to consider moving to Windows Azure Cloud Services (more information about that here: Migrating a Drupal Site from LAMP to Windows Azure). Before getting started, it’s worth noting that Windows Azure Websites lets you run up to 10 Web Sites for free in a multitenant environment. And, you can seamlessly upgrade to private, reserved VM instances as your traffic grows. To sign up, try the Windows Azure 90-day free trial. 1. Create a Windows Azure Web Site and MySQL database There is a step-by-step tutorial on http://www.windowsazure.com that walks you through creating a new website and a MySQL database, so I’ll refer you there to get started: Create a PHP-MySQL Windows Azure web site and deploy using Git. If you intend to use Git to publish your Drupal site, then go ahead and follow the instructions for setting up a Git repository. Make sure to follow the instructions in the Get remote MySQL connection information section as you will need that information later. You can ignore the remainder of the tutorial for the purposes of deploying your Drupal site, but if you are new to Windows Azure Web Sites (and to Git), you might find the additional reading informative. Ok, now you have a new website with a MySQL database, your have your MySQL database connection information, and you have (optionally) created a remote Git repository and made note of the Git deployment instructions. Now you are ready to copy your database to MySQL in Windows Azure Web Sites. 2. Copy database to MySQL in Windows Azure Web Sites I’m sure there is more than one way to copy your Drupal database, but I found the mysqldump tool to be effective and easy to use. To copy from a local machine to Windows Azure Web Sites, here’s the command I used: mysqldump -u local_username --password=local_password drupal | mysql -h remote_host -u remote_username --password=remote_password remote_db_name You will, of course, have to provide the username and password for your existing Drupal database, and you will have to provide the hostname, username, password, and database name for the MySQL database you created in step 1. This information is available in the connection string information that you should have noted in step 1. i.e. You should have a connection string that looks something like this: Database=remote_db_name;Data Source=remote_host;User Id=remote_username;Password=remote_password Depending on the size of your database, the copying process could take several minutes. Now your Drupal database is live in Windows Azure Websites. Before you deploy your Drupal code, you need to modify it so it can connect to the new database. 3. Modify database connection info in settings.php Here, you will again need your new database connection information. Open the /drupal/sites/default/setting.php file in your favorite text editor, and replace the values of ‘database’, ‘username’, ‘password’, and ‘host’ in the $databases array with the correct values for your new database. When you are finished, you should have something similar to this: $databases = array ( 'default' => array ( 'default' => array ( 'database' => 'remote_db_name', 'username' => 'remote_username', 'password' => 'remote_password', 'host' => 'remote_host', 'port' => '', 'driver' => 'mysql', 'prefix' => '', ), ), ); Be sure to save the settings.phpfile, then you are ready to deploy. 4. Deploy Drupal code using Git or FTP The last step is to deploy your code to Windows Azure Web Sites using Git or FTP. If you are using FTP, you can get the FTP hostname and username from you website’s dashboard. Then, use your favorite FTP client to upload your Drupal files to the /site/wwwroot folder of the remote site. If you are using Git, you need to set up a Git repository in Windows Azure Web Sites (steps for this are in the tutorial mentioned earlier). And, you will need Git installed on your local machine. Then, just follow the instructions provided after you created the repository: One note about using Git here: depending on your Git settings, your .gitignore file (a hidden file and a sibling to the .git folder created in your local root directory after you executed git commit), some files in your Drupal application may be ignored. In my case, all the files in the sites directory were ignored. If this happens, you will want to edit the .gitignore file so that these files aren’t ignored and redeploy. After you have deployed Drupal to Windows Azure Web Sites, you can continue to deploy updates via Git or FTP. Related information If you are looking for more information about Windows Azure Web Sites, these posts might be helpful: Windows Azure Websites- A PHP Perspective Windows Azure Websites, Web Roles, and VMs- When to use which- Configuring PHP in Windows Azure Websites with .user.ini Files One last thing you might consider, depending on your site, is using the Windows Azure Integration Module to store and serve your site’s media files.

This is the second article of the series on Spring Integration. This article builds on top of the first article where we introduced Spring Integration. Context setting In the first article, we created a simple java application where A message was sent over a channel, It was intercepted by a service i.e. POJO and modified. It was then sent over a different channel The modified message was read from the channel and displayed. However, in doing this - keeping in mind that we were merely introducing the concepts there - we wrote some Spring specific code in our application i.e. the test classes. In this article we will take care of that and make our application code as insulated from Spring Integration api as possible. This is done by, what Spring Integration calls gateways. Gateways exist for the sole purpose of abstracting messaging related "plumbing" code away from "business" code. The business logic might really not care whether a functionality is being achieved be sending a message over a channel or by making a SOAP call. This abstraction - though logical and desirable - have not been very practical, till now. It is probably worth having a quick look at the Spring Integration Reference Manual at this point. However, if you are just getting started with Spring Integration, you are perhaps better off following this article for the moment. I would recommend you get your hands dirty before returning to reference manual, which is very good but also very exhaustive and hence could be overwhelming for a beginner. The gateway could be a POJO with annotations (which is convenient but in my mind beats the whole purpose) or with XML configurations (can very quickly turn into a nightmare in any decent sized application if unchecked). At the end of the day it is really your choice but I like to go the XML route. The configuration options for both styles are detailed out in this section of the reference implementation. Spring Integration with Gateways So, let's create another test with gateway throw in for our HelloWorld service (refer to the first article of this series for more context). Let's start with the Spring configuration for the test. File: src/test/resources/org/academy/integration/HelloWorld1Test-context.xml In this case, all that is different is that we have added a gateway. This is an interface called org.academy.integration.Greetings. It interacts with both "inputChannel" and "outputChannel", to send and read messages respectively. Let's write the interface. File: /src/main/java/org/academy/integration/Greetings.java package org.academy.integration; public interface Greetings { public void send(String message); public String receive(); } And then we add the implementation of this interface. Wait. There is no implementation. And we do not need any implementation. Spring uses something called GatewayProxyFactoryBean to inject some basic code to this gateway which allows it to read the simple string based message, without us needing to do anything at all. That's right. Nothing at all. Note - You will need to add more code for most of your production scenarios - assuming you are not using Spring Integration framework to just push around strings. So, don't get used to free lunches. But, while it is here, let's dig in. Now, lets write a new test class using the gateway (and not interact with the channels and messages at all). File: /src/test/java/org/academy/integration/HelloWorld1Test.java package org.academy.integration; import static org.junit.Assert.*; import org.junit.Test; import org.junit.runner.RunWith; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.test.context.ContextConfiguration; import org.springframework.test.context.junit4.SpringJUnit4ClassRunner; @RunWith(SpringJUnit4ClassRunner.class) @ContextConfiguration public class HelloWorld1Test { private final static Logger logger = LoggerFactory .getLogger(HelloWorld1Test.class); @Autowired Greetings greetings; @Test public void test() { greetings.send("World"); assertEquals(greetings.receive(), "Hello World"); logger.debug("Spring Integration with gateways."); } } Our test class is much cleaner now. It does not know about channels, or messages or anything related to Spring Integration at all. It only knows about a greetings instance - to which it gave some data by .send() method - and got modified data back by .receive() method. Hence, the business logic is oblivious of the plumbing logic, making for a much cleaner code. Now, simply type "mvn -e clean install" (or use m2e plugin) and you should be able to run the unit test and confirm that given string "World" the HelloWorld service indeed returns "Hello World" over the entire arrangement of channels and messages. Again, something optional but I highly recommend, is to run "mvn -e clean install site". This - assuming you have correctly configured some code coverage tool (cobertura in my case) will give you a nice HTML report showing the code coverage. In this case it would be 100%. I have blogged a series on code quality which deals this subject in more detail, but to cut long story short, it is very important for me to ensure that whatever coding practice / framework I use and recommend use, complies to some basic code quality standards. Being able to unit test and measure that is one such fundamental check that I do. Needless to say, Spring in general (including Spring integration) passes that check with flying colours. Conclusion That's it for this article. Happy coding.

It has been a while since I wanted to have CDI running with JavaFX2. Some people already blogged on how to proceed by getting Guice injection [1] to work with JavaFX & FXML. Well, now it's my turn to provide a way to empower JavaFX with CDI, using Weld as the implementation. My goal was just to have CDI working, no matter how I was using JavaFX, by directly coding in plain Java or using FXML. Ready? Let's go!!! Bootstrap JavaFX & Weld/CDI The launcher class will be the only place where we will have Weld-specific code—all the rest will be totally CDI compliant. The only trick here is to make the application parameters available as a CDI-compliant object so we can reuse them afterwards. Notice also that we use the CDI event mechanism to start up our real application code. public class WeldJavaFXLauncher extends Application { /** * Nothing special, we just use the JavaFX Application methods to boostrap * JavaFX */ public static void main(String[] args) { Application.launch(WeldJavaFXLauncher.class, args); } @SuppressWarnings("serial") @Override public void start(final Stage primaryStage) throws Exception { // Let's initialize CDI/Weld. WeldContainer weldContainer = new Weld().initialize(); // Make the application parameters injectable with a standard CDI // annotation weldContainer.instance().select(ApplicationParametersProvider.class).get().setParameters(getParameters()); // Now that JavaFX thread is ready // let's inform whoever cares using standard CDI notification mechanism: // CDI events weldContainer.event().select(Stage.class, new AnnotationLiteral() {}).fire(primaryStage); } } Start our real JavaFX application Here we start our real application code. We're just listening to the previously fired event (containing the Scene object to render into) so we can start showing our application. In the following example, we load an FXML GUI, but it might have been any node created in any way. public class LoginApplicationStarter { // Let's have a FXMLLoader injected automatically @Inject FXMLLoader fxmlLoader; // Our CDI entry point, we just listen to an event providing the startup scene public void launchJavaFXApplication(@Observes @StartupScene Stage s) { InputStream is = null; try { is = getClass().getResourceAsStream("login.fxml"); // we just load our FXML form (including controler and so on) Parent root = (Parent) fxmlLoader.load(is); s.setScene(new Scene(root, 300, 275)); s.show(); // let's show the scene } catch (IOException e) { throw new IllegalStateException("cannot load FXML login screen", e); } finally { // omitted is cleanup } } } But what about the FXML controller? First let's have a look at the controller we want to use inside our application. It is a pure POJO class annotated with both JavaFX & CDI annotations. // Simple application controller that uses injected fields // to delegate login process and to get default values from the command line using: --user=SomeUser public class LoginController implements Initializable { // Standard FXML injected fields @FXML TextField loginField; @FXML PasswordField passwordField; @FXML Text feedback; // CDI Injected service @Inject LoginService loginService; // Default application parameters retrieved using CDI @Inject Parameters applicationParameters; @FXML protected void handleSubmitButtonAction(ActionEvent event) { feedback.setText(loginService.login(loginField.getText(), passwordField.getText())); } @Override public void initialize(URL location, ResourceBundle resources) { loginField.setText(applicationParameters.getNamed().get("user")); } } In order to have injection working inside the FXML controller, we need to set up JavaFX so that controller objects are created by CDI. As we are in a CDI environment we can also have the FXMLLoader classes injected (that's exactly what we did in the previous LoginApplicationStarter class). How can we achieve this? We just have to provide a Producer class whose responsibility will be to create FXMLLoader instances that are able to load FXML GUIs and instantiate controllers using CDI. The only part that's a little tricky there is that the controller instantiation depends on the required class or interface (using fx:controller in your fxml file). In order to have such a runtime injection/resolution available we use a CDI Instance Object. public class FXMLLoaderProducer { @Inject Instance, Object>() { @Override public Object call(Class param) { return instance.select(param).get(); } }); return loader; } } I hope you found the article interesting and you do not hesitate to comment if you see some errors or possible enhancements. Finally, if you are interested you can find the full source code here. [1] http://andrewtill.blogspot.be/2012/07/creating-javafx-controllers-using-guice.htm

Autoscaling your webserver tier is typically straightforward. Image your apache server with source code or without, then sync down files from S3 upon spinup. Roll that image into the autoscale configuration and you’re all set. With the database tier though, things can be a bit tricky. The typical configuration we see is to have a single master database where your application writes. But scaling out or horizontally on Amazon EC2 should be as easy as adding more slaves, right? Why not automate that process? Below we’ve set out to answer some of the questions you’re likely to face when setting up slaves against your master. We’ve included instructions on building an AMI that automatically spins up as a slave. Fancy! How can I autoscale my database tier? Build an auto-starting MySQL slave against your master. Configure those to spinup. Amazon’s autoscaling loadbalancer is one option, another is to use a roll-your-own solution, monitoring thresholds on servers, and spinning up or dropping off slaves as necessary. Does an AWS snapshot capture subvolume data or just the SIZE of the attached volume? In fact, if you have an attached EBS volume and you create an new AMI off of that, you will capture the entire root volume, plus your attached volume data. In fact we find this a great way to create an auto-building slave in the cloud. How do I freeze MySQL during AWS snapshot? mysql> flush tables with read lock;mysql> system xfs_freeze -f /data At this point you can use the Amazon web console, ylastic, or ec2-create-image API call to do so from the command line. When the server you are imaging off of above restarts – as it will do by default – it will start with /data partition unfrozen and mysql’s tables unlocked again. Voila! If you’re not using xfs for your /data filesystem, you should be. It’s fast! The xfsprogs docs seem to indicate this may also work with foreign filesystems. Check the docs for details. How do I build an AMI mysql slave that autoconnects to master? Install mysql_serverid script below. Configure mysql to use your /data EBS mount. Set all your my.cnf settings including server_id Configure the instance as a slave in the normal way. When using GRANT to create the ‘rep’ user on master, specify the host with a subnet wildcard. For example ’10.20.%’. That will subsequently allow any 10.20.x.y servers to connect and replicate. Point the slave at the master. When all is running properly, edit the my.cnf file and remove server_id. Don’t restart mysql. Freeze the filesystem as described above. Use the Amazon console, ylastic or API call to create your new image. Test it of course, to make sure it spins up, sets server_id and connects to master. Make a change in the test schema, and verify that it propagates to all slaves. How do I set server_id uniquely? As you hopefully already know, in MySQL replication environment each node requires a unique server_id setting. In my Amazon Machine Images, I want the server to startup and if it doesn’t find the server_id in the /etc/my.cnf file, to add it there, correctly! Is that so much to ask? Here’s what I did. Fire up your editor of choice and drop in this bit of code: #!/bin/shif grep -q “server_id” /etc/my.cnf then : # do nothing – it’s already set else # extract numeric component from hostname – should be internet IP in Amazon environment export server_id=`echo $HOSTNAME | sed ‘s/[^0-9]*//g’` echo “server_id=$server_id” >> /etc/my.cnf # restart mysql /etc/init.d/mysql restart fi Save that snippet at /root/mysql_serverid. Also be sure to make it executable: $ chmod +x /root/mysql_serverid Then just append it to your /etc/rc.local file with an editor or echo: $ echo "/root/mysql_serverid" >> /etc/rc.local Assuming your my.cnf file does *NOT* contain the server_id setting when you re-image, then it’ll set this automagically each time you spinup a new server off of that AMI. Nice! Can you easily slave off of a slave? How? It’s not terribly different from slaving off of a normal master. A. First enable slave updates. The setting is not dynamic, so if you don’t already have it set, you’ll have to restart your slave. log_slave_updates=true B. Get an initial snapshot of your slave data. You can do that the locking way: mysql> flush tables with read lock;mysql> show master status\G; mysql> system mysqldump -A > full_slave_dump.mysql mysql> unlock tables; You may also choose to use Percona’s excellent xtrabackup utility to create hotbackups without locking any tables. We are very lucky to have an open-source tool like this at our disposal. MySQL Enterprise Backup from Oracle Corp can also do this. C. On the slave, seed the database with your dump created above. $ mysql < full_slave_dump.mysql D. Now point your slave to the original slave. mysql> change master to master_user='rep', master_password='rep', master_host='192.168.0.1', master_log_file='server-bin-log.000004', master_log_pos=399;mysql> start slave; mysql> show slave status\G; Slave master is set as an IP address. Is there another way? It’s possible to use hostnames in MySQL replication, however it’s not recommended. Why? Because of the wacky world of DNS. Suffice it to say MySQL has to do a lot of work to resolve those names into IP addresses. A hickup in DNS can interrupt all MySQL services potentially as sessions will fail to authenticate. To avoid this problem do two things: A. Set this parameter in my.cnf skip_name_resolve = true Remove entries in mysql.user table where hostname is not an IP address. Those entries will be invalid for authentication after setting the above parameter. Doesn’t RDS take care of all of this for me? RDS is Amazon’s Relational Database Service which is built on MySQL. Amazon’s RDS solution presents MySQL as a service which brings certain benefits to administrators and startups: Simpler administration. Nuts and bolts are handled for you. Push-button replication. No more struggling with the nuances and issues of MySQL’s replication management. Simplicity of administration of course has it’s downsides. Depending on your environment, these may or may not be dealbreakers. No access to the slow query log. This is huge. The single best tool for troubleshooting slow database response is this log file. Queries are a large part of keeping a relational database server healthy and happy, and without this facility, you are severely limited. Locked in downtime window When you signup for RDS, you must define a thirty minute maintenance window. This is a weekly window during which your instance *COULD* be unavailable. When you host yourself, you may not require as much downtime at all, especially if you’re using master-master mysql and zero-downtime configuration. Can’t use Percona Server to host your MySQL data. You won’t be able to do this in RDS. Percona server is a high performance distribution of MySQL which typically rolls in serious performance tweaks and updates before they make it to community addition. Well worth the effort to consider it. No access to filesystem, server metrics & command line. Again for troubleshooting problems, these are crucial. Gathering data about what’s really happening on the server is how you begin to diagnose and troubleshoot a server stall or pileup. You are beholden to Amazon’s support services if things go awry. That’s because you won’t have access to the raw iron to diagnose and troubleshoot things yourself. Want to call in an outside consultant to help you debug or troubleshoot? You’ll have your hands tied without access to the underlying server. You can’t replicate to a non-RDS database. Have your own datacenter connected to Amazon via VPC? Want to replication to a cloud server? RDS won’t fit the bill. You’ll have to roll your own – as we’ve described above. And if you want to replicate to an alternate cloud provider, again RDS won’t work for you. Related posts: Deploying MySQL on Amazon EC2 – 8 Best Practices Review: Host Your Web Site In The Cloud, Amazon Web Services Made Easy 5 Ways to Boost MySQL Scalability Top MySQL DBA interview questions (Part 2) MySQL Cluster In The Cloud – Managers Guide

When defining a Zabbix (1.8.2) trigger e.g. to inform you that there are errors in a log file, how do you verify that it is correct? As somebody recommended in a forum, you can use a Calculated Item with a similar expression (the syntax is little different from triggers). Contrary to triggers, the value of a calculated item is easy to see and the historical values are stored so you can check how it evolved. If your trigger expression is complex the you can create multiple calculated items, one for each subexpression. Example If we have a log item that sends us data whenever the text “ERROR” appears in a log line and the corresponding trigger expected to fire if we have got any data from the item in the last 600 sec (nodata() returns 1 if there indeed was no data): {hive.example.com:log["/tmp/ada/hive.log","ERROR",,20].nodata(600)}=0 Then we could test it with a calculated item with the expression nodata("hive.example.com:log[\"/tmp/ada/hive.log\",\"ERROR\",,20]", 600) (Notice that the function comes first, taking the host:item as its 0th parameter and that this is enclosed with “”, escaping any nested ” with \.) The value of the calculated item will be re-checked every (independent on whether the source item changed or not) and stored, in this case it will either thave the value of 0 or 1. We can also construct more complex expressions with &, + etc. similarly to trigger expressions.

the question everybody always asks when they learn about activiti, is as old as software development itself: “how does it perform?”. up till now, when you would ask me that same question, i would tell you about how activiti minimizes database access in every way possible, how we break down the process structure into an ‘execution tree’ which allows for fast queries or how we leverage ten years of workflow framework development knowledge. you know, trying to get around the question without answering it. we knew it is fast, because of the theoretical foundation upon which we have built it. but now we have proof: real numbers …. yes, it’s going to be a lengthy post. but trust me, it’ll be worth your time! disclaimer: performance benchmarks are hard. really hard. different machines, slight different test setup … very small things can change the results seriously. the numbers here are only to prove that the activiti engine has a very minimal overhead, while also integrating very easily into the java eco-system and offering bpmn 2.0 process execution. the activiti benchmark project to test process execution overhead of the activiti engine, i created a little side project on github: https://github.com/jbarrez/activiti-benchmark the project contains currently 9 test processes, which we’ll analyse below. the logic in the project is pretty straightforward: a process engine is created for each test run each of the processes are sequentially executed on this process engine, using a threadpool from 1 up to 10 threads. all the processes are thrown into a bag, of which a number of random executions are drawn. all the results are collected and a html report with some nice charts are generated to run the benchmark, simply follow the instructions on the github page to build and execute the jar. benchmark results the test machine i used for the results is my (fairly old) desktop machine: amd phenom ii x4 940 3.0ghz, 8 gb 800mhz ram and an old-skool 7200 rpm hd running ubuntu 11.10. the database used for the test runs on the same machine on which the tests also run. so keep in mind that in a ‘real’ server environment the results could even be better! the benchmark project i mentioned above, was executed on a default ubuntu mysql 5 database. i just switched to the ‘large.cnf’ setting (which throws more ram at the db and stuff like that) instead the default config. each of the test processes ran for 2500 times, using a threadpool going from one to ten threads . in simpleton language: 2500 process executions using just one thread, 2500 threads using two threads, 2500 process executions using three … yeah, you get it. each benchmark run was done using a ‘default’ activiti process engine. this basically means a ‘regular’ standalone activiti engine, created in plain java. each benchmark run was also done in a ‘spring’ config. here, the process engine was constructed by wrapping it in the factory bean, the datasource is a spring datasource and also the transactions and connection pool is managed by spring (i’m actually using a tweaked bonecp threadpool) each benchmark run was executed with history on the default history level (ie. ‘audit’) and without history enabled (ie. history level ‘none’) . the processes are in detail analyzed in the sections below, but here are the integral results of the test runs already: activiti 5.9 – mysql – default – history enabled activiti 5.9 – mysql – default – history disabled activiti 5.9 – mysql – spring – history enabled activiti 5.9 – mysql – spring – history disabled i ran all the tests using the latest public release of activiti, being activiti 5.9. however, my test runs brought some potential performance fixes to the surface (i also ran the benchmark project through a profiler). it was quickly clear that most of the process execution time was done actually cleaning up when a process ended. basically, more than often queries were fired which were not necessary if we would save some more state in our execution tree. i sat together with daniel meyer from camunda and my colleague frederik heremans, and they’ve managed to commit fixes for this! as such, the current trunk of activiti, being activiti 5.10-snapshot at the moment, is significantly faster than 5.9 . activiti 5.10 – mysql – default – history enabled activiti 5.10 – mysql – default – history disabled activiti 5.10 – mysql – spring – history enabled activiti 5.10 – mysql – spring – history disabled from a high-level perspective (scroll down for detailed analysis), there are a few things to note: i had expected some difference between the default and spring config, due to the more ‘professional’ connection pool being used. however, the results for both environments are quite alike. sometimes the default is faster, sometimes spring. it’s hard to really find a pattern. as such, i omitted the spring results in the detailed analyses below. the best average timings are most of the times found when using four threads to execute the processes . this is probably due to having a quad-core machine. the best throughput numbers are most of the times found when using eight threads to execute the processes. i can only assume that is also has something to do with having a quad-core machine. when the number of threads in the threadpool go up, the throughput (processes executed / second) goes up, both it has a negative effect on the average time. certainly with more than six or seven threads, you see this effect very clear. this basically means that while the processes on itself take a little longer to execute, but due to the multiple threads you can execute more of these ‘slower’ processes in the same amount of time. enabling history does have an impact. often, enabling history will double execution time. this is logical, given that many extra records are inserted when history is on the default level (ie. ‘audit’). there was one last test i ran, just out of curiosity: running the best performing setting on an oracle xe 11.2 database. the oracle xe is a free version of the ‘real’ oracle database. no matter how hard, i tried, i couldn’t get it decently running on ubuntu. as such, i used an old windows xp install on that same machine. however, the os is 32 bit, wich means the system only has 3.2 of the 8gb of ram available. here are the results: activiti 5.10 – oracle on windows – default – history disabled the results speak for itself. oracle blows away any of the (single-threaded) results on mysql (and they are already very fast!). however, when going multi-threaded it is far worse than any of the mysql results. my guess is that these are due to the limitations of the xe version : only one cpu is used, only 1 gb of ram, etc. i would really like to run these test on a real oracle-managed-by-a-real-dba … feel free to contact me if you are interested ! in the next sections, we will take a detailed look into the performance numbers of each of the test processes. an excel sheet containing all the the numbers and charts below can be downloaded for yourself . process 1: the bare micromum (one transaction) the first process is not a very interesting one, business-wise at least. after starting the process, the end is immediately reached. not very useful on itself, but its numbers learn us one essential thing: the bare overhead of the activiti engine. here are the average timings: this process runs in a single transaction, which means that nothing is saved to the database when the history is disabled due to activiti’s optimizations. with history enabled, you’ll basically get the cost for inserting one row into the historical process instance table, which is around 4.44 ms here. it is also clear that our fix for activiti 5.10 has an enormous impact here. in the previous version, 99% of the time was spent in the cleanup check of the process. take a look at the best result here: 0.47 ms when using 4 threads to execute 2500 runs of this process. that’s only half a millisecond ! it’s fair to say that the activiti engine overhead is extremely small. the throughput numbers are equally impressive: in the best case here, 8741 processes are executed. per second. by the time you arrive here reading the post, you could have executed a few millions of this process . you can also see that there is little difference between 4 or 8 threads here. most of the execution time here is cpu time, and no potential collisions such as waiting for a database lock happens here. in these numbers, you can also easily see that the oracle xe doesn’t scale well with multiple threads (which is explained above). you will see the same behavior in the following results. process 2: the same, but a bit longer (one transaction) this process is pretty similar to the previous one. we have again only one transaction. after the process is started, we pass through seven no-op passthrough activities before reaching the end. some things to note here: the best result (again 4 threads, with history disabled) is actually better than the simpler previous process. but also note that the single threaded execution is a tad slower. this means that the process on itself is a bit slower, which is logical as is has more activities. but using more threads and having more activities in the process does allow for more potential interleaving. in the previous case, the thread was barely born before it was killed again. the difference between history enabled/disabled is bigger than the previous process. this is logical, as more history is written here (for each activity one record in the database). again, activiti 5.10 is far more superior to activiti 5.9. the throughput numbers follow these observations: there is more opportunity to use threading here. the best result lingers around 12000 process execution per second . again, it demonstrates the very lightweight execution of the activiti engine. process 3: parallelism in one transaction this process executes a parallel gateway that forks and one that joins in the same transaction. you would expect something along the lines of the previous results, but you’d be surprised: comparing these numbers with the previous process, you see that execution is slower. so why is this process slower, even if it has less activities? the reason lies with how the parallel gateway is implemented, especially the join behavior. the hard part, implementation-wise, is that you need to cope with the situation when multiple executions arrive at the join. to make sure that the behavior is atomic, we internally do some locking and fetch all child executions in the execution tree to find out whether the join activates or not. so it is quite a ‘costly’ operation, compared to the ‘regular’ activities. do mind, we’re talking here about only 5 ms single threaded and 3.59 ms in the best case for mysql . given the functionality that is required for implementing the parallel gateway functionality, this is peanuts if you’d ask me. the throughput numbers: this is the first process which actually contains some ‘logic’. in the best case above, it means 1112 processes can be executed in a second. pretty impressive, if you’d ask me! . process 4: now we’re getting somewhere (one transaction) this process already looks like something you’d see when modeling real business processes. we’re still running it in one database transaction though, as all the activities are automatic passthroughs. here we also have two forks and two joins. take a look at the lowest number: 6.88 ms on oracle when running with one thread. that’s freaking fast , taking in account all that is happening here. the history numbers are at least doubled here (activiti 5.10), which makes sense because there is quite a bit of activity audit logging going on here. you can also see that this causes to have a higher average time for four threads here, which is probably due to the implementation of the joining. if you know a bit about activiti internals, you’ll understand this means there are quite a bit of executions in the execution tree. we have one big concurrent root, but also multiple children which are sometimes also concurrent roots. but while the average time rises, the throughput definitely benefits: running this process with eight threads, allows you to do 411 runs of this process in a single second. there is also something peculiar here: the oracle database performs better with more thread concurrency. this is completely contrary with all other measurements, where oracle is always slower in that environment (see above for explanation). i assume it has something to do with the internal locking and forced update we are applying when forking/joining, which is better handled by oracle it seems. process 5: adding some java logic (single transaction) i added this process to see the influence of adding a java service task in a process. in this process, the first activity generates a random value, stores it as a process variable and then goes up or down in the process depending on the random value. the chance is about 50/50 to go up or down. the average timings are very very good. actually, the results are in the same range as those of process 1 and 2 above (which had no activities or only automatic passthroughs). this means that the overhead of integrating java logic into your process is nearly non-existant (nothing is of course for free). of course, you can still write slow code in that logic, but you can’t blame the activiti engine for that throughput numbers are comparable to those of process 1 and 2: very, very high. in the best case here, more than 9000 processes are executed per second . that indeed also means 9000 invocations of your own java logic. process 6, 7 and 8: adding wait states and transactions the previous processes demonstrated us the bare overhead of the activiti engine. here, we’ll take a look at how wait states and multiple transactions have influence on performance. for this, i added three test processes which contain user tasks. for each user task, the engine commits the current transaction and returns the thread to the client. since the results are pretty much compatible for these processes, we’re grouping them here. these are the processes: here are the average timings results, in order of the processes above. for the first process, containing just one user task: it is clear that having wait states and multiple transaction does have influence on the performance. this is also logical: before, the engine could optimize by not inserting the runtime state into the database, because the process was finished in one transaction. now, the whole state, meaning the pointers to where you are currently, need to be saved into the database. the process could be ‘sleeping’ like this for many days, months, years now …. the activiti engine doesn’t hold it into memory now anymore, and it is freed to give its full attention to other processes. if you check the results of the process with only one user task, you can see that in the best case (oracle, single thread – the 4 threads on mysql is pretty close) this is done in 6.27ms . this is really fast, if you take in account we have a few inserts (the execution tree, the task), a few updates (the execution tree) and deletes (cleaning up) going on here. the second process here, with 7 user tasks: the second chart learns us that logically, more transactions means more time. in the best case here the process is done in 32.12 ms . that is for seven transactions, which gives 4.6 ms for each transactions. so it is clear that average time scales in a linearly way when adding wait states. this makes of course sense, because transactions aren’t free. also note that enabling history does add quite some overhead here. this is due to having the history level set to ‘audit’, which stores all the user task information in the history tables. this is also noticeable from the difference between activiti 5.9 with history disabled and activiti 5.10 with history enabled: this is a rare case where activiti 5.10 with history enabled is slower than 5.9 with history disabled. but it is logical, given the volume of history stored here. and the third process learns us how user tasks and parallel gateways interact: the third chart learns us not much new. we have two user tasks now, and the more ‘expensive’ fork/join (see above). the average timings are how we expected them. the throughput charts are as you would expect given the average timings. between 70 and 250 processes per second. aw yeah! to save some space, you’ll need to click them to enlarge: process 9: so what about scopes? for the last process, we’ll take a look at ‘scopes’. a ‘scope’ is how we call it internally in the engine, and it has to do with variable visibility, relationships between the pointers indicating process state, event catching, etc. bpmn 2.0 has quite some cases for those scopes, for example with embedded subprocesses as shown in the process here. basically, every subprocess can have boundary events (catching an error, a message, etc) that only are applied on its internal activities when it’s scope is active. without going into too much technical details: to get scopes implemented in the correct way, you need some not so trivial logic. the example process here has 4 subprocesses, nested in each other. the inner process is using concurrency, which is a scope on itself again for the activiti engine. there are also two user tasks here, so that means two transactions. so let’s see how it performs: you can clearly see the big difference between activiti 5.9 and 5.10. scopes are indeed an area where the fixes around the ‘process cleanup’ at the end have a huge benefit, as many execution objects are created and persisted to represent the many different scopes. single threaded performance is not so good on activiti 5.9. luckily, as you can see from the gap between the blue and the red bars, those scopes do allow for high concurrency. the numbers of oracle, combined with the multi-threaded results of the 5.10 tests, do prove that scopes are now efficiently handled by the engine. the throughput charts prove that the process nicely scales with more threads, as you can see by the big gap between the red and green line in the second last block. in the best case, 64 processes of this more complex process are handled by the engine. random execution if you have already clicked on the full reports at the beginning of the post, you probably have noticed also random execution is tested for each environment. in this setting, 2500 process executions were done, both the process was randomly chosen. as shown in those reports this meant that over 2500 executions, each process was executed almost the same number of times (normal distribution). this last chart shows the best setting (activiti 5.10, history disabled) and how the throughput of those random process executions goes when adding more threads: as we’ve seen in many of the test above, once passed four threads things don’t change that much anymore. the numbers (167 processes/second) prove that in a realistic situation (ie. multiple processes executing at the same time), the activiti engine nicely scales up. conclusion the average timing charts show two things clearly: the activiti engine is fast and overhead is minimal ! the difference between history enabled or disabled is definitely noticeably. sometimes it comes even down to half the time needed. all history tests were done using the ‘audit’ level, but there is a simpler history level (‘activity’) which might be good enough for the use case. activiti is very flexible in history configuration, and you can tweak the history level for each process specifically. so do think about the level your process needs to have, if it needs to have history at all ! the throughput charts prove that the engine scales very well when more threads are available (ie. any modern application server). activiti is well designed to be used in high-throughput and availability (clustered) architectures . as i said in the introduction, the numbers are what they are: just numbers. my main point which i want to conclude here, is that the activiti engine is extremely lightweight. the overhead of using activiti for automating your business processes is small. in general, if you need to automate your business processes or workflows, you want top-notch integration with any java system and you like all of that fast and scalable … look no further!

Here are the most important subjects for software engineering, with brief explanations: 1.Object oriented analysis & design: For better maintainability, reusability and faster development, the most well accepted approach, shortly OOAD and its SOLID principals are very important for software engineering. 2.Software quality factors: Software engineering depends on some very important quality factors. Understanding and applying them is crucial. 3.Data structures & algorithms: Basic data structures like array, list, stack, tree, map, set etc. and useful algorithms are vital for software development. Their logical structure should be known. 4. Big-O notation: Big-O notation indicates the performance of an algorithm/code section. Understanding it is very important for comparing performances. 5.UML notation: UML is the universal and complete language for software design & analysis. If there is lack of UML in a development process, it feels there is no engineering. 6.Software processes and metrics: Software enginnering is not a random process. It requires a high level of systematic and some numbers to monitor those techniques. So, processes and metrics are essential. 7.Design patterns: Design patterns are standard and most effective solutions for specific problems. If you don't want to reinvent the wheel, you should learn them. 8.Operating systems basics: Learning OS basics is very important because all applications runs on it. By learning it, we can have better vision, viewpoints and performance for our applications. 9.Computer organization basics: All applications including OS requires a hardware for physical interaction. So, learning computer organization basics is vital again for better vision, viewpoints and performance. 10.Network basics: Network is related with computer organization, OS and the whole information transfer process. In any case we will face it while software development. So, it is important to learn network basics. 11.Requirement analysis: Requirement analysis is the starting point and one of the most important parts of software engineering. Performing it correctly and practically needs experience but it is very essential. 12.Software testing: Testing is another important part of software engineering. Unit testing, its best practices and techniques like black box, white box, mocking, TDD, integration testing etc. are subjects which must be known. 13.Dependency management: Library (JAR, DLL etc.) management, and widely known tools (Maven, Ant, Ivy etc.) are essential for large projects. Otherwise, antipatterns like Jar Hell are inevitable. 14.Continuous integration: Continuous integration brings easiness and automaticity for testing large modules, components and also performs auto-versioning. Its aim and tools (like Hudson etc.) should be known. 15.ORM (Object relational mapping): ORM and its widely known implementation Hibernate framework is an important technique for mapping objects into database tables. It reduces code length and maintenance time. 16.DI (Dependency Injection): DI or IoC (Inversion of Control) and its widely known implementation Spring framework makes life easy for object creation and lifetime management on big enterprise applications. 17.Version controlling systems: VCS tools (SVN, TFS, CVS etc.) are very important by saving so much time for collaborative works and versioning. Their logical viewpoint and standard cammands should be known. 18.Internationalization (i18n): i18n by extracting strings into external files is the best way of supporting multiple languages in our applications. Its practices on different IDEs and technologies must be known. 19.Architectural patterns: Understanding architectural design patterns (like MVC, MVP, MVVM etc.) is essential for producing a maintainable, clean, extendable and testable source code. 20.Writing clean code: Working code is not enough, it must be readable and maintainable also. So, code formatting and readable code development techniques are needed to be known and applied.

Envers is a Hibernate module that can be configured to automatically audit changes made to your entities. Each audited entity are thus associated with a list of revisions, each revision capturing the state of the entity when a change occurs. There is however an obstacle I came across while I was "unit testing" my DAO, and that's what I want to share to avoid others to fall in the same pit. First, let's have an overview of the couple of steps needed to use Envers: Annotate your entity with the @Audited annotation: @Entity @Audited public class Person { // Properties } Register the Envers AuditEventListener in your Hibernate SessionFactory through Spring: org.hibernate.dialect.H2Dialect Configure the Hibernate transaction manager as your transaction manager. Note auditing won't be triggered if you use another transaction manager (DataSourceTransactionManager comes to mind): Now is the time to create your test class: @ContextConfiguration("classpath:spring-persistence.xml") @TransactionConfiguration(defaultRollback = false) public class PersonDaoImplTest extends AbstractTransactionalTestNGSpringContextTests { @Autowired private PersonDao personDao; @BeforeMethod protected void setUp() { // Populate database } @Test public void personShouldBeAudited() { Person person = personDao.get(1L); person.setFirstName("Jane"); List history = personDao.getPersonHistory(1L); assertNotNull(history); assertFalse(history.isEmpty()); assertEquals(history.size(), 1); } } Strangely, when you execute the previous test class, the test method fails when checking the list is not empty: it is, meaning there's no revision associated with the entity. Morevoer, nothing shows up in the log. However, the revision shows up in the audited table at the end of the test (provide you didn't clear the table after its execution). Comes the dreaded question: why? Well, it seems Hibernate post-event listeners are only called when the transaction is commited. In our case, it matches: the transaction is commited by Spring after method completion, and our test trie to assert inside the method. In order for our test to pass, we have to manually manage a transaction inside our method, to commit the update to the database. @Test public void personShouldBeAuditedWhenUpdatedWithManualTransaction() { PlatformTransactionManager txMgr = applicationContext.getBean(PlatformTransactionManager.class); // A new transaction is required, the wrapping transaction is for Envers TransactionStatus status = txMgr.getTransaction(new DefaultTransactionDefinition(PROPAGATION_REQUIRES_NEW)); Person person = personDao.get(1L); person.setFirstName("Jane"); txMgr.commit(status); List history = personDao.getPersonHistory(1L); assertNotNull(history); assertFalse(history.isEmpty()); assertEquals(history.size(), 1); } On one hand, the test passes and the log shows the SQL commands accordingly. On the other hand, the cost is the additional boilerplate code needed to make it pass. Of course, one could (should?) question the need to test the feature in the first place. Since it's a functionality brought by a library, the reasoning behind could be that if you don't trust the library, don't use it at all. In my case, it was the first time I used Envers, so there's no denying I had to build the trust between me and the library. Yet, even with trusted libraries, I do test specific cases: for example, when using Hibernate, I create test classes to verify that complex queries get me the right results. As such, auditing qualifies as a complex use-case whose misbehaviors I want to be aware of as soon as possible. You'll find the sources for this article here, in Maven/Eclipse format.

One of the advantages of code generation is the ability to see how a specific language feature or framework is used. As I discussed in the post NetBeans 7.2 beta: Faster and More Helpful, NetBeans 7.2 beta provides TestNG integration. I did not elaborate further in that post other than a single reference to that feature because I wanted to devote this post to the subject. I use this post to demonstrate how NetBeans 7.2 can be used to help a developer new to TestNG start using this alternative (to JUnit) test framework. NetBeans 7.2's New File wizard makes it easier to create an empty TestNG test case. This is demonstrated in the following screen snapshots that are kicked off by using New File | Unit Tests (note that "New File" is available under the "File" drop-down menu or by right-clicking in the Projects window). Running the TestNG test case creation as shown above leads to the following generated test code. TestNGDemo.java (Generated by NetBeans 7.2) package dustin.examples; import org.testng.annotations.AfterMethod; import org.testng.annotations.AfterClass; import org.testng.annotations.BeforeMethod; import org.testng.annotations.BeforeClass; import org.testng.annotations.Test; import org.testng.Assert; /** * * @author Dustin */ public class TestNGDemo { public TestNGDemo() { } @BeforeClass public void setUpClass() { } @AfterClass public void tearDownClass() { } @BeforeMethod public void setUp() { } @AfterMethod public void tearDown() { } // TODO add test methods here. // The methods must be annotated with annotation @Test. For example: // // @Test // public void hello() {} } The test generated by NetBeans 7.2 includes comments indicate how test methods are added and annotated (similar to modern versions of JUnit). The generated code also shows some annotations for overall test case set up and tear down and for per-test set up and tear down (annotations are similar to JUnit's). NetBeans identifies import statements that are not yet used at this point (import org.testng.annotations.Test; and import org.testng.Assert;), but are likely to be used and so have been included in the generated code. I can add a test method easily to this generated test case. The following code snippet is a test method using TestNG. testIntegerArithmeticMultiplyIntegers() @Test public void testIntegerArithmeticMultiplyIntegers() { final IntegerArithmetic instance = new IntegerArithmetic(); final int[] integers = {4, 5, 6}; final int expectedProduct = 2 * 3 * 4 * 5 * 6; final int product = instance.multiplyIntegers(2, 3, integers); assertEquals(product, expectedProduct); } This, of course, looks very similar to the JUnit equivalent I used against the same IntegerArithmetic class that I used for testing illustrations in the posts Improving On assertEquals with JUnit and Hamcrest and JUnit's Built-in Hamcrest Core Matcher Support. The following screen snapshot shows the output in NetBeans 7.2 beta from right-clicking on the test case class and selecting "Run File" (Shift+F6). The text output of the TestNG run provided in the NetBeans 7.2 beta is reproduced next. [TestNG] Running: Command line suite [VerboseTestNG] RUNNING: Suite: "Command line test" containing "1" Tests (config: null) [VerboseTestNG] INVOKING CONFIGURATION: "Command line test" - @BeforeClass dustin.examples.TestNGDemo.setUpClass() [VerboseTestNG] PASSED CONFIGURATION: "Command line test" - @BeforeClass dustin.examples.TestNGDemo.setUpClass() finished in 33 ms [VerboseTestNG] INVOKING CONFIGURATION: "Command line test" - @BeforeMethod dustin.examples.TestNGDemo.setUp() [VerboseTestNG] PASSED CONFIGURATION: "Command line test" - @BeforeMethod dustin.examples.TestNGDemo.setUp() finished in 2 ms [VerboseTestNG] INVOKING: "Command line test" - dustin.examples.TestNGDemo.testIntegerArithmeticMultiplyIntegers() [VerboseTestNG] PASSED: "Command line test" - dustin.examples.TestNGDemo.testIntegerArithmeticMultiplyIntegers() finished in 12 ms [VerboseTestNG] INVOKING CONFIGURATION: "Command line test" - @AfterMethod dustin.examples.TestNGDemo.tearDown() [VerboseTestNG] PASSED CONFIGURATION: "Command line test" - @AfterMethod dustin.examples.TestNGDemo.tearDown() finished in 1 ms [VerboseTestNG] INVOKING CONFIGURATION: "Command line test" - @AfterClass dustin.examples.TestNGDemo.tearDownClass() [VerboseTestNG] PASSED CONFIGURATION: "Command line test" - @AfterClass dustin.examples.TestNGDemo.tearDownClass() finished in 1 ms [VerboseTestNG] [VerboseTestNG] =============================================== [VerboseTestNG] Command line test [VerboseTestNG] Tests run: 1, Failures: 0, Skips: 0 [VerboseTestNG] =============================================== =============================================== Command line suite Total tests run: 1, Failures: 0, Skips: 0 =============================================== Deleting directory C:\Users\Dustin\AppData\Local\Temp\dustin.examples.TestNGDemo test: BUILD SUCCESSFUL (total time: 2 seconds) The above example shows how easy it is to start using TestNG, especially if one is moving to TestNG from JUnit and is using NetBeans 7.2 beta. Of course, there is much more to TestNG than this, but learning a new framework is typically most difficult at the very beginning and NetBeans 7.2 gets one off to a fast start.