I'm facing problems with "too many connections" for PHPUnit tests for ZF3 and Doctrine, because I'm executing ~200 tests per PHPUnit execution.
I've already found some questions and answers on stack overflow but non of these work.
My setup:
ZF2/ZF3, Doctrine 2 and PHPUnit.
I have a base test class for all tests and the setUp and tearDown function look like this:
public function setUp()
{
$this->setApplicationConfig(Bootstrap::getConfig());
Bootstrap::loadAllFixtures();
if (!static::$em) {
echo "init em";
static::$em = Bootstrap::getEntityManager();
}
parent::setUp();
....
}
public function tearDown()
{
parent::tearDown();
static::$em->flush();
static::$em->clear();
static::$em->getConnection()->close();
$refl = new \ReflectionObject($this);
foreach ($refl->getProperties() as $prop) {
if (!$prop->isStatic() && 0 !== strpos($prop->getDeclaringClass()->getName(), 'PHPUnit_')) {
$prop->setAccessible(true);
$prop->setValue($this, null);
}
}
gc_collect_cycles();
}
public static function (Bootstrap::)loadAllFixtures()
{
static::$em->getConnection()->executeUpdate("SET foreign_key_checks = 0;");
$loader = new Loader();
foreach (self::$config['data-fixture'] as $fixtureDir) {
$loader->loadFromDirectory($fixtureDir);
}
$purger = new ORMPurger(static::$em);
$executor = new ORMExecutor(static::$em, $purger);
$executor->execute($loader->getFixtures());
$executor = null;
$purger = null;
static::$em->getConnection()->executeUpdate("SET foreign_key_checks = 1;");
static::$em->flush();
static::$em->clear();
}
I'm monitoring my local MySQL server with innotop and the number of connections is increasing.
Do you have any ideas what I'm missing?
Thank you,
Alexander
Update 14.02.2017:
I've changed functions to use static::$em and added Bootstrap::loadAllFixtures method.
If I add static::$em->close() to tearDown method, all following test fail with message like "EntityManager already closed". echo "init em"; is only call once and shown for the first test.
Is there a possibility to check if my Application opens connections without closing them? My test cases are based on AbstractHttpControllerTestCase
I came across this problem too. Following the advice in PHPUnit's documentation I had done the following:
final public function getConnection()
{
if ($this->conn === null) {
if (self::$pdo == null) {
//We get the EM from dependency injection container
$container = $this->getContainer();
self::$pdo = $container->get('Doctrine.EntityManager')->getConnection()->getWrappedConnection();
}
$this->conn = $this->createDefaultDBConnection(self::$pdo, 'spark_api_docker');
}
return $this->conn;
}
While self:$pdo was being shared, the number of 'threads_connected', when I observed show status like '%onn%'; on my database, crept up until it reached the limit.
I found two solutions to this:
1) Close the connection after each test
public function tearDown()
{
parent::tearDown();
//You'll probably need to get hold of our entity manager another way
$this->getContainer()->get('Doctrine.EntityManager')->getConnection()->close();
}
importantly, do not set self::$pdo to null. I had seen this as a recommendation elsewhere, but there's no point setting it as a static property and then resetting it after each test.
This works my closing connections that are no longer needed. When a testcase is finished, unless you have closed the connection it will remain open until the script ends (i.e. PHPUnit finishes running your test). Since you're creating a new connection for each test case, the number of connections goes up.
2) Run each test in a seperate PHP thread
This is the sledgehammer approach. It will likely impact the speed of your tests to one degree or another. In your phpunit.xml`:
<?xml version="1.0" encoding="UTF-8"?>
<phpunit
...
processIsolation = "true"
>
...
</phpunit>
Returning to PHPUnit's advice, storing the connection and PDO helps with not creating new connections for each test but does not help you when you have many test cases. Each test case gets instantianted in the same thread, and each will create a new connection.
Your tearDown method looks like it should do the trick. I just do this and have never experienced this issue
protected function tearDown()
{
parent::tearDown();
$this->em->close();
$this->em = null;
}
What does Bootstrap::loadAllFixtures method do? Is there any db connection in there that might be being overlooked?
Related
I followed this guide: https://symfony.com/doc/current/console/lockable_trait.html and implemented the command lock feature for my one of my commands to see how it works. It worked as described and then I was going to implement it for all of my commands. But the issue is that I have about 50 commands and:
I do not want spent time adding the necessary code to each command
I want to have the centralized management of commands locking. I mean, adding extra option to regular commands so that they will be used by my future management center. For now I will need a pretty simple option protected function isLocked() for a regular command which will help me to manage if a command should have lockable feature.
So, I went to the source of \Symfony\Component\Console\Command\LockableTrait and after some time created the following listener to the event console.command:
use Symfony\Component\Console\Event\ConsoleCommandEvent;
use Symfony\Component\Console\Exception\LogicException;
use Symfony\Component\Lock\Lock;
use Symfony\Component\Lock\LockFactory;
use Symfony\Component\Lock\LockInterface;
use Symfony\Component\Lock\Store\FlockStore;
use Symfony\Component\Lock\Store\SemaphoreStore;
class LockCommandsListener
{
/**
* #var array<string, Lock>
*/
private $commandLocks = [];
private static function init()
{
if (!class_exists(SemaphoreStore::class)) {
throw new LogicException('To enable the locking feature you must install the symfony/lock component.');
}
}
public function onConsoleCommand(ConsoleCommandEvent $event)
{
static::init();
$name = $event->getCommand()->getName();
$this->ensureLockNotPlaced($name);
$lock = $this->createLock($name);
$this->commandLocks[$name] = $lock;
if (!$lock->acquire()) {
$this->disableCommand($event, $name);
}
}
private function disableCommand(ConsoleCommandEvent $event, string $name)
{
unset($this->commandLocks[$name]);
$event->getOutput()->writeln('The command ' . $name . ' is already running');
$event->disableCommand();
$event->getCommand()->setCode()
}
private function createLock(string $name): LockInterface
{
if (SemaphoreStore::isSupported()) {
$store = new SemaphoreStore();
} else {
$store = new FlockStore();
}
return (new LockFactory($store))->createLock($name);
}
private function ensureLockNotPlaced(string $name)
{
if (isset($this->commandLocks[$name])) {
throw new LogicException('A lock is already in place.');
}
}
}
I made some tests and it kind of worked. But I am not sure this is the right way of doing things.
Another problem is that I can not find the proper exit code when I disabled a command. Should I just disable it? But it seems that exit code would be a great feature here. Specially when it comes to this listener testing (PHPUnit testing).
And I also have with testing itself. How can I run commands in parallel in my test class. For now I have this:
class LockCommandTest extends CommandTest
{
public function testOneCommandCanBeRun()
{
$commandTester = new ApplicationTester($this->application);
$commandTester->run([
'command' => 'app:dummy-command'
]);
$output = $commandTester->getDisplay();
dd($output);
}
}
It will allow only to run my commands one by one. But I would like to run them both so after running the first one, the second will fail (with some exit code).
As for me the best way to make background task is doing it via supervisor, create config file, like:
[program:your_service]
command=/usr/local/bin/php /srv/www/bin/console <your:app:command>
priority=1
numprocs=1
# Each 5 min.
startsecs=300
autostart=true
autorestart=true
process_name=%(program_name)s_%(process_num)02d
user=root
this is the best way to be sure that your command will be ran only in one process
After years of searching for answers in other people's posts, I finally have to ask.
I'm currently trying to implement a WatchDog / Shepherd pattern to monitor long processes (from 2-3 minutes to several hours) running in my app.
The app is fairly complicated, and I started working on it mid-project, so I don't grasp it entirely.
Two servers are running : The one we'll call FACADE, with an Apache/2.4.10 (Debian), and CALCULUS, with an Apache/2.4.6 (Red Hat Enterprise Linux).
I have designed things like this :
The Shepherd initializes on FACADE, when the user click on a button that triggers a long process
The process, running on CALCULUS, initialises a WatchDog that connects to the Shepherd using a TCP Socket.
At some key points of the process, the WatchDog 'barks', i.e sends a string to the Shepherd to tell him at which step the Process is (the string is like "M(essage)#S(tarting)#Indexing", "M#C(ompleted)#Indexing", "M#S#Transfert", "M#C#Transfert"...) and if there's been an error (in which case it sends "E#Indexing" => Error happened during Indexing)
When he gets a Message from the dog, the Shepherd does what he has to (process, displaying to the user, filling a bit the progress bar...)
That my friends, was purely theory. Now comes the Implementation :
/**
*Implements the Singleton Design Pattern
*And of course, the Watchdog Pattern, using a Socket to communicate with the Shepherd.
*/
class Watchdog{
/** Singleton Design pattern. Use $watchdog = Watchdog::getInstance() to use the doggy */
private static $instance = null;
private function __construct(){
$this->init();
}
public static function getInstance(){
if (Watchdog::$instance === null)
Watchdog::$instance = new Watchdog();
return Watchdog::$instance;
}
private $socket;
private $connected = false;
private function init(){
$this->socket = fsockopen("tcp://A.B.C.D", 4242, $errno, $errstr);
if($this->socket === false){
echo "$errno : $errstr";
}else{
$this->connected = true;
}
}
public function kill(){
fclose($this->socket);
Watchdog::$instance = null;
}
public function bark($message){
if($this->connected === true){
fwrite($this->socket, "M#".$message."\n");
}
}
public function alert($err){
if($this->connected === true){
fwrite($this->socket, "E#".$err."\n");
}
}
}
?>
And the Shepeherd :
/**Implements the Singleton Design Pattern,
* And the Shepherd / Watchdog Patter, using a Socket to communicate with the Watchdog.
*/
class Shepherd{
/** Singleton Design pattern. Use $shepherd = Shepherd::GetInstance() to use the shepherd */
private static $instance;
private function __construct()
{
$this->init();
}
public static function GetInstance(){
if (self::$instance === null)
self::$instance = new Shepherd();
return self::$instance;
}
/**Instance variables **/
private $socket;
public $initialised = false;
public $doggyConnected = false;
private $dogsocket;
/**Init : Initializes the shephrd by binding it to the watchdog on the 4242 port**/
private function init(){
ob_implicit_flush();
$this->socket = stream_socket_server("tcp://0.0.0.0:4242", $errno, $errstr);
$this->initialised = true;
}
/**Sit And Wait : Waits for Dog connection**/
public function sitandwait(){
//Waiting for doggy connection
do {
if (($this->dogsocket = stream_socket_accept($this->socket)) === false) {
echo "socket_accept() failed: reason: " . socket_strerror(socket_last_error($this->dogsocket)) . "\n";
break;
}else{
$this->doggyConnected = true;
}
}while($this->doggyConnected === false);
}
/**Listen to Dog : Waits for Dog to send a message and echoes it **/
public function listentodog(){
if($this->dogsocket !== false) {
$buf = fgets($this->dogsocket);
return $buf;
}
}
/**Kill : Kills the shepherd and closes connections **/
private function kill(){
stream_socket_shutdown($this->socket, STREAM_SHUT_RDWR);
stream_socket_shutdown($this->dogsocket, STREAM_SHUT_RDWR);
fclose ($this->dogsocket);
fclose ($this->socket);
Shepherd::$instance = null;
}
/**Run : Runs the Shepherd till he got a message from dog **/
public function run(){
if($this->doggyConnected === false)
$this->sitandwait();
return $this->listentodog();
}
}
After trying those without success on the real website, I decided to try and see what was happening on both servers, thanks to netcat command. There's the catch : When I fake the Shepherd thanks to netcat, the dog can connect and send accurate data about the process. When I fake the dog thanks to netcat --send-only, the Shepherd gets the data and does the right things with them.
But when I run both from the application, I get a "Connexion refused" at the dog->init() (fsockopen : Connexion refusée), and of course, the Shepherd dies from Timeout.
But wait there's more !! You might think that the problem comes from the connection, and with netcat It doesn't come up because I don't do the connection from PHP (or I don't connect to a PHP opened socket).
I thought that too; I wrote two scripts, test_dog.php and test_shepherd.php, that are used EXACTLY in the same way than in my real live application. When I try to make them communicate, It works ! It even works with a real dog(monitoring a real application process) and test_shepherd.php, or with a real Shepherd (from my app), and test_watchdog.php
I decided to come here to ask you guys from help, because I'm utterly lost. i don't understand why It doesn't work with the real code, but does with the test_ scripts. In those, I made sure to use the objects exactly the same way than in the real application.
To show you everything, here are my test_ scripts :
test_watchdog.php
require "Watchdog.php";
$dog = Watchdog::getInstance();
$dog->bark("Try try try");
$dog->bark("Trying hard !!");
sleep(5);
$dog->bark("Trying harder to see...");
sleep(2);
$dog->bark("END");
$dog->kill();
test_shepherd.php
require "Shepherd.php";
$shep = new _Shepherd();
echo $shep->run();
... i think that's All. Please answer if you have the faintest idea that might help me, you're my last hope, I'm lost and desperate...
Thank you in advance :)
EDIT : On CALCULUS, the Watchdog is called by a Thousand-lines-long class, called Process (that runs the main process). The point is to be able to call Watchdog nearly everywhere in the code, where the user might have to wait.
Here is for instance the __construct of Process, intializing the Watchdog, and one of the methods that calls the $doggy->bark();
public function __construct($photoId = 0) {
$this->params = array();
$this->doggy = Watchdog::GetInstance();
$this->params['photoid'] = $photoId ;
date_default_timezone_set ('Europe/Paris');
}
public function transfertProject() {
try {
$this->doggy->bark('s#transfert');
//Traitement long
set_time_limit (0);
ini_set('post_max_size', 0);
ini_set('upload_max_filesize', 0);
$response = false;
if (!isset($_FILES['file'])) {
$post_max_size = ini_get('post_max_size');
$upload_max_size = ini_get('upload_max_filesize');
return "Le fichier ne semble pas avoir été posté, vérifier la taille maximal d'upload";
}
$name = $_FILES['file']['name'];
$filename = "../Workspace/projects/".$name;
$tmp = $_FILES['file']['tmp_name'];
if (move_uploaded_file($_FILES['file']['tmp_name'], $filename)) {
$response = $this->unzipProjectArchive();
unlink($filename);
}
$this->doggy->bark('c#transfert');
return $response;
} // END TRY
catch (Exception $ex) {
//Watchdog telling shepherd
$this->doggy->alert('transfert');
}
}
during unit testing i'm always get confused about what to test.
Do i need to test the API and only the API or also the method result values.
class SomeEventHandler
{
public function onDispatch (Event $event)
{
if ($event->hasFoo)
{
$model = $this->createResponseModel('foo');
}
else
{
$model = $this->createResponseModel('bar');
}
// End.
return $model;
}
private function createResponseModel ($foo)
{
$vars = array(
'someVare' => true,
'foo' => $foo
);
// End.
return new ResponseModel($vars);
}
}
So should i test if the method onDispatch returns a instance of ResponseModel or should i also test if the variable foo is set properly?
Or is the test below just fine?
class SomeEventHandlerTest
{
// assume that a instance of SomeEventHandler is created
private $someEventHandler;
public function testOnDispatch_EventHasFoo_ReturnsResponseModel ()
{
$e = new Event();
$e->hasFoo = true;
$result = $someEventHandler->onDispatch($e);
$this->assertInstanceOf('ResponseModel', $result);
}
public function testOnDispatch_EventHasNoFoo_ReturnsResponseModel ()
{
$e = new Event();
$e->hasFoo = false;
$result = $someEventHandler->onDispatch($e);
$this->assertInstanceOf('ResponseModel', $result);
}
}
If you were checking the code by hand what is it that you would check? Just that a ResponseModel was returned or that it also had the proper values?
If you weren't writing tests and executed the code what would you look for to ensure that the code was doing what it was supposed to. You would check that the values in the returned object were correct. I would do that by using the public API of the object and verify that the values are right.
One idea is to have the tests such that if the code were deleted, you would be able to recreate all the functionality via only having the tests. Only checking the returned object could result in a function that just has return new ResponseModel();. This would pass the test but would not be what you want.
In short, what you decide to test is subjective, however you should at the minimum test all your public methods.
Many people limit their tests to public methods and simply ensure code coverage on the protected/private methods is adequate. However, feel free to test anything you think warrants a test. Generally speaking, the more tests the better.
In my opinion you should certainly test for your response data, not just the return type.
I rely on Unit Tests to let me make code changes in the future and be satisfied my changes have not created any breaks, just by running the tests.
So in your case, if the "foo" or "bar" response data is important, you should test it.
That way if you later change the response strings by accident, your tests will tell you.
I'm trying to run a job queue to create a PDF file using SlmQueueBeanstalkd and DOMPDFModule in ZF".
Here's what I'm doing in my controller:
public function reporteAction()
{
$job = new TareaReporte();
$queueManager = $this->serviceLocator->get('SlmQueue\Queue\QueuePluginManager');
$queue = $queueManager->get('myQueue');
$queue->push($job);
...
}
This is the job:
namespace Application\Job;
use SlmQueue\Job\AbstractJob;
use SlmQueue\Queue\QueueAwareInterface;
use SlmQueue\Queue\QueueInterface;
use DOMPDFModule\View\Model\PdfModel;
class TareaReporte extends AbstractJob implements QueueAwareInterface
{
protected $queue;
public function getQueue()
{
return $this->queue;
}
public function setQueue(QueueInterface $queue)
{
$this->queue = $queue;
}
public function execute()
{
$sm = $this->getQueue()->getJobPluginManager()->getServiceLocator();
$empresaTable = $sm->get('Application\Model\EmpresaTable');
$registros = $empresaTable->listadoCompleto();
$model = new PdfModel(array('registros' => $registros));
$model->setOption('paperSize', 'letter');
$model->setOption('paperOrientation', 'portrait');
$model->setTemplate('empresa/reporte-pdf');
$output = $sm->get('viewPdfrenderer')->render($model);
$filename = "/path/to/pdf/file.pdf";
file_put_contents($filename, $output);
}
}
The first time you run it, the file is created and the work is successful, however, if you run a second time, the task is buried and the file is not created.
It seems that stays in an endless cycle when trying to render the model a second time.
I've had a similar issue and it turned out it was because of the way ZendPdf\PdfDocument reuses it's object factory. Are you using ZendPdf\PdfDocument?
You might need to correctly close factory.
class MyDocument extends PdfDocument
{
public function __destruct()
{
$this->_objFactory->close();
}
}
Try to add this or something similar to the PdfDocument class...
update : it seem you are not using PdfDocument, however I suspect this is the issue is the same. Are you able to regenerate a second PDF in a normal http request? It is your job to make sure the environment is equal on each run.
If you are unable to overcome this problem a short-term quick solution would be to set max_runs configuration for SlmQueue to 1. That way the worker is stopped after each job and this reset to a vanilla state...
How do I get started with mocking a web service in PHP? I'm currently directly querying the web API's in my unit testing class but it takes too long. Someone told me that you should just mock the service. But how do I go about that? I'm currently using PHPUnit.
What I have in mind is to simply save a static result (json or xml file) somewhere in the file system and write a class which reads from that file. Is that how mocking works? Can you point me out to resources which could help me with this. Is PHPUnit enough or do I need other tools? If PHPUnit is enough what part of PHPUnit do I need to check out? Thanks in advance!
You would mock the web service and then test what is returned. The hard coded data you are expecting back is correct, you set the Mock to return it, so then additional methods of your class may continue to work with the results. You may need Dependency Injection as well to help with the testing.
class WebService {
private $svc;
// Constructor Injection, pass the WebService object here
public function __construct($Service = NULL)
{
if(! is_null($Service) )
{
if($Service instanceof WebService)
{
$this->SetIWebService($Service);
}
}
}
function SetWebService(WebService $Service)
{
$this->svc = $Service
}
function DoWeb($Request)
{
$svc = $this->svc;
$Result = $svc->getResult($Request);
if ($Result->success == false)
$Result->Error = $this->GetErrorCode($Result->errorCode);
}
function GetErrorCode($errorCode) {
// do stuff
}
}
Test:
class WebServiceTest extends PHPUnit_Framework_TestCase
{
// Simple test for GetErrorCode to work Properly
public function testGetErrorCode()
{
$TestClass = new WebService();
$this->assertEquals('One', $TestClass->GetErrorCode(1));
$this->assertEquals('Two', $TestClass->GetErrorCode(2));
}
// Could also use dataProvider to send different returnValues, and then check with Asserts.
public function testDoWebSericeCall()
{
// Create a mock for the WebService class,
// only mock the getResult() method.
$MockService = $this->getMock('WebService', array('getResult'));
// Set up the expectation for the getResult() method
$MockService->expects($this->any())
->method('getResult')
->will($this->returnValue(1)); // Change returnValue to your hard coded results
// Create Test Object - Pass our Mock as the service
$TestClass = new WebService($MockService);
// Or
// $TestClass = new WebService();
// $TestClass->SetWebServices($MockService);
// Test DoWeb
$WebString = 'Some String since we did not specify it to the Mock'; // Could be checked with the Mock functions
$this->assertEquals('One', $TestClass->DoWeb($WebString));
}
}
This mock may then be used in the other functions since the return is hard coded, your normal code would process the results and perform what work the code should (Format for display, etc...). This could also then have tests written for it.