I had several:

Generating code coverage report, this may take a moment.PHP Fatal error: Allowed memory size of 268435456 bytes exhausted (tried to allocate 21503527 bytes) in /usr/share/php/PHPUnit/Util/Report/Node/File.php on line 406

And here is how it started:

At work we are using phpunit to test our framework, we flew from SimpleTest as there was no real integration with our tools and no real followup of the the product.

So to automate the test, as there is no “build” in php (interpreted vs compiled remember?) we ran it after each commit …

Until it tooks 30 minutes to run throught the whole set of tests…

Then we decided to run it locally before comitting… until it tooks up to 2 hours to run tests (yes we have slow laptops I know, at least 4 times slower than the servers).

Actually we use phpunit for recipe test, including database change, third parties, providers… everything from A to Z… and that needs a lot of time and CPU power…

So we solved the problem by running test in a cronjob, hourly on the integration server.

This was done running phpunit in command line


> phpunit --include-path ../code/ MyTestSuite.php


Of course as we had several tests suites and we had to put them all in one bigger testsuite… but that was not easy so we moved to the Phpunit config file, added code coverage reporting and send a mail everytime sometimes goes wrong

PhpunitConfig.xml

<?xml version="1.0" encoding="UTF-8"?>
<phpunit
         colors="true"
         >
         <testsuites>
         	<testsuite name="RTBF_Service">
         		<file>RTBF/Service/Rating/Adapter/AbstractPhpunittest.php</file>
         		<file>RTBF/Service/Rating/Adapter/MediaPhpunittest.php</file>
         	<--! ... -->
         	</testsuite>
         	<testsuite name="RTBF">
         		<file>RTBF/AclPhpunittest.php</file>
         	</testsuite>
         	<testsuite name="RTBF_Utility">
         		<file>RTBF/Utility/AjaxConnectorPhpunittest.php</file>
	         	<--! ... -->
         	</testsuite>
         	<--! ... -->
	</testsuites>
	<listeners>
  		<listener class="SimpleTestListener" file="PhpunitTestListener.php">
  		</listener>
  	</listeners>
  	<!--
	<logging>
		<log type="coverage-html" target="/reporting/phpunit/library/" charset="UTF-8" />
	</logging>
	-->
	<php>
		<ini name="memory_limit" value="2047M" />
	</php>
	<filter>
	  <blacklist>
	    <directory suffix=".php">library/code/Zend</directory>
	  </blacklist>
	</filter>
</phpunit>

PhpunitTestListener.php

class SimpleTestListener implements PHPUnit_Framework_TestListener
{
	private $emailContent=array(self::Error=>"",self::Failure=>"",self::Incomplete=>"",self::Skipped=>"");
	private $assertions=0;
	private $itemCount=array(self::Error=>0,self::Failure=>0,self::Incomplete=>0,self::Skipped=>0);
	private $totalItems=0;

	const Error="Error";
	const Failure="Failure";
	const Incomplete="Incomplete";
	const Skipped="Skipped";

	private function addEmailContent($severity,$test,$e){
		$result = $test->getResult();
		$traces = $e->getTrace();
		$continue=true;
		$tracingMessage="";
		for($i=1;$i<count($traces)&&$continue;$i++){
			if(isset($traces[$i]['file'])){
				$tracingMessage.=$traces[$i]['file'].":(".$traces[$i]['line'].")\n\t";
			}else{
				$continue=false;
			}
		}
		$this->itemCount[$severity]++;
		$this->emailContent[$severity].=$this->itemCount[$severity].")\t".get_class($test).'::'.$test->getName(TRUE)."\n\t".str_replace("\n","\n\t",PHPUnit_Framework_TestFailure::exceptionToString($e))."".$tracingMessage."\n\n";
		$this->totalItems++;
	}

	public function addError(PHPUnit_Framework_Test $test,Exception $e,$time){
		$this->addEmailContent(self::Error,$test,$e);
	}

	public function addFailure(PHPUnit_Framework_Test $test,PHPUnit_Framework_AssertionFailedError $e,$time){
		$this->addEmailContent(self::Failure,$test,$e);
	}

	public function addIncompleteTest(PHPUnit_Framework_Test $test,Exception $e,$time){
		$this->addEmailContent(self::Incomplete,$test,$e);
	}

	public function addSkippedTest(PHPUnit_Framework_Test $test,Exception $e,$time){
		$this->addEmailContent(self::Skipped,$test,$e);
	}

	public function startTest(PHPUnit_Framework_Test $test){
	}

	public function endTest(PHPUnit_Framework_Test $test, $time){
	}

	public function startTestSuite(PHPUnit_Framework_TestSuite $suite){
	}

	public function endTestSuite(PHPUnit_Framework_TestSuite $suite){
	}

	public function __destruct(){
		if($this->totalItems>0){
			// HIDDEN but concat everything from $this->emailContent nicely and send the mail
			//print("\n\nMail sent: ".$text."\n");
		}
	}
}

That’s it for the big picture…

And as you can see there are several point in the XML that are quite important:

memory limit 2047MBthat’s huge… but this is the price to pay to have a fully working code-coverage HTML report. As we are using suhosin we had to change the memory limit in suhosin too… and yes when the limit is 2048 MB don’t try to set the memory to 2048 MB it really means this is a limit you have to be under that limit (here 2047 MB).

2 Gigs!!! I know this is huge but believe me I went through several tries and 128 MB was not enought, nor 256 MB, nor 512 MB, not even 1 GB … I didn’t try to see where is my limit … I put 2 gigs and then it worked… we have about 30% of lines covered so I think that if we shift to 100% we will have to give more memory… I’m pretty sure, there is something leaking somewhere… but according to phpunit.de site… I have the answer remains: Give more memory

/etc/php5/conf.d/suhosin.ini

logging is commented out because logging needs xdebug and while we stay with ZendServer CE we do not have xdebug… (see my previous post for more info).

But the server which is not running Zend Server CE does!

So I commented out the line in the config to let developers run the unit test and pasted it as command line argument on the server’s cronjob.

phpunit --include-path ../code/ --coverage-html /reporting/phpunit/library/ --configuration PhpunitConfig.xml

the listener This is a simple easy listener. It’s always useful to have one sample so I pasted here above, feel free to copy, enhance… The only ‘hack’ we use it to call the send email on listener destruction, so the email will be sent at the end. But there is a drawback… if there is a php error somewhere, nothing is sent. And as nothing is sent when everythign goes well… we dont know if everythings goes well or if there is a real bug…
This problem will be addressed later. Stay tuned!

user@laptop:~$ sudo bash

1st be sre to have the pecl, pear available as root in command line

root@laptop:~# PATH=$PATH:/usr/local/zend/bin

trying to install

root@laptop:~# pecl install xdebug
downloading xdebug-2.1.0.tgz ...
Starting to download xdebug-2.1.0.tgz (301,354 bytes)
.....................................................done: 301,354 bytes
66 source files, building
running: phpize
Configuring for:
PHP Api Version: 20041225
Zend Module Api No: 20060613
Zend Extension Api No: 220060519
Cannot find autoconf. Please check your autoconf installation and the
$PHP_AUTOCONF environment variable. Then, rerun this script.

ERROR: `phpize' failed

Argl it seems to be failing…

looks like autoconf is missing

root@laptop:~# autoconf
The program 'autoconf' can be found in the following packages:
* autoconf
* autoconf2.13
Try: apt-get install <selected package>


ok let’s add it:


root@laptop:~# apt-get install autoconf


and now let’s try again


root@laptop:~# pecl install xdebug
downloading xdebug-2.1.0.tgz ...
Starting to download xdebug-2.1.0.tgz (301,354 bytes)
.............................................................done: 301,354 bytes
66 source files, building
running: phpize
Configuring for:
PHP Api Version: 20041225
Zend Module Api No: 20060613
Zend Extension Api No: 220060519
configure.in:3: warning: prefer named diversions
configure.in:3: warning: prefer named diversions
building in /var/tmp/pear-build-root/xdebug-2.1.0
running: /tmp/pear/temp/xdebug/configure
checking for grep that handles long lines and -e... /bin/grep
checking for egrep... /bin/grep -E
checking for a sed that does not truncate output... /bin/sed
checking for cc... cc
...
Build process completed successfully
Installing '/usr/local/zend/lib/php_extensions/xdebug.so'
install ok: channel://pecl.php.net/xdebug-2.1.0
configuration option "php_ini" is not set to php.ini location
You should add "extension=xdebug.so" to php.ini


Great!

Don’t forget to add the librairy in the php.ini

create the xdebug.ini in the ext.d directory and the symlink in the conf.d directory


root@laptop:/usr/local/zend/etc/ext.d# more xdebug.ini
zend_extension=/usr/local/zend/lib/php_extensions/xdebug.so
; You may put any extension-specific directives here

root@laptop:/usr/local/zend/etc/conf.d# ln -s ../ext.d/xdebug.ini xdebug.ini


Now let’s try PHP:

scips@laptop:~$ php
Segmentation fault


Looks like I need to investigate deeper…

Ok problem solved:

Instructions

Warning: You seem to be using Zend Server, which is known to cause issues with Xdebug. It might work, but you’re on your own.

Download xdebug-2.1.0.tgz
Unpack the downloaded file with tar -xvzf xdebug-2.1.0.tgz
Run: cd xdebug-2.1.0
Run: phpize

As part of its output it should show:

Configuring for:
…
Zend Module Api No: 20060613
Zend Extension Api No: 220060519
If it does not, you are using the wrong phpize. Please follow this FAQ entry and skip the next step.

Run: ./configure
Run: make
Run: cp modules/xdebug.so /usr/local/zend/lib/php_extensions
Open /usr/local/zend/etc/conf.d/debugger.ini and put a ; in front of the line that says zend_extension_manager.dir.debugger= so that it says ;zend_extension_manager.dir.debugger=
Edit /usr/local/zend/etc/php.ini and add at the begining of the file the line
zend_extension = /usr/local/zend/lib/php_extensions/xdebug.so

I did just paste in has requested my php -i output on http://xdebug.org/find-binary.php then compiled and everything ran fine.

The problem is that you have to use one or the other not the both at the same time… still not a real option for me.

first of all you need the best script ever made to backup subversion: the one provided with!

Get it here: http://svn.apache.org/repos/asf/subversion/trunk/tools/backup/hot-backup.py.in

This script needs to be modified

replace:


# Path to svnlook utility
svnlook = r"@SVN_BINDIR@/svnlook"
# Path to svnadmin utility
svnadmin = r"@SVN_BINDIR@/svnadmin"


with:


# Path to svnlook utility
svnlook = r"/usr/bin/svnlook"
# Path to svnadmin utility
svnadmin = r"/usr/bin/svnadmin"


or your appropriate svnlook and svnadmin path

Then you need to check that your user can modify files

once it’s verified let’s create a repository to store backup ‘/backup’ for instance and create your directory structure there

Test it

test the whole stuff

./hot-backup.py --archive-type=bz2 --num-backups=45 /svn/TEST_REPO/ /backup/subversion/test_repo/


you should get something like TEST_REPO-2.tar.bz2 (here minus 2 means revision n°2)

Schedule it

If you want to run it daily, put it in a cron such as a daily job

As I have multiple repositories here is my script


#!/bin/bash
NUMBACKUP=15
for i in `ls /svn/`
do
mkdir -p /backup/svn/$i
./hot-backup.py --archive-type=bz2 --num-backups=$NUMBACKUP /svn/$i /backup/svn/$i
done


Restore it

Ok that’s nice, you got a backup, but unless you can restore it, it’s useless!

So here is how to restore it:

1. tar -xjvf /backup/svn/TEST/test-2.tar.bz2
2. cp -R /backup/svn/TEST/TEST-2/ /svn/TEST/

and… that’s it

of course at the end of your backup script I recommend to move everything to another computer, because you will probably need it when disk crashes ;-)

post commit auto deploy

In 4 steps

1. svn commit to the repository
2. the pre-commit hook will check the code with svnlook and will validate the commit message (based on regexp such has bug# …, text minimal size…)
3. the post-commit hook will try to connect to the ITT (Integration Team Testing) if you have the good acronym, I’ll update
4. once connected a script should be launched (and forgot &;) This script should:
   • Do a svn checkout locally so the code could be tested (in /var/www …). This method has several drawback: instability of the site while exporting, destruction of ITT generated file (if any)…
   • Do a svn checkout in a temporary directory (/export/<projectname>) and then use rsync to do a fast compare and copy changes only without deleting the existing, excluding the .svn folders
   • Do a svn export in a temporary directory (/tmp/<projectname>) and then use the same rsync than above but this time you don’t have to exclude the .svn

   The fastest is the second one (svn checkout -> /export/<projectname> + rsync – .svn folders)

Option 2

There is another alternative if you have space and sufficient right on your svn repository: the svn export/svn checkout is done locally and you use rsync to deploy the site on the ITT server.

Option 3

There is a third alternative if you have one more server (a synchro server) that will be used to store the svn export (at least the last one and maybe 2-3 version before) and that will take care of the rsync between one of its local checkout/export and the ITT server.

Conclusion

The last option has the big advantage of being a intermediate repository that can deploy any revision to any server … but that’s another paper.

I used the 3 options, and the first one (the simplest/cheapest) is very easy to maintain and to set up, so it’s a must have for team < 15 … bigger team may think about the other options but will have to put additional resources to design the whole deployment system.

Dr. Wave ready to listen

While muting a wave and moving it to the trash folder :-)

I can’t wait to wave :P

And unfortunately, I will not be retired at that time… I’m pretty sure that it will be like in the 80s with every developer thinking that his code will last only few months (years)… when it will actually last much longer.

Most of the liunuix (linux and unix) systems use timestamp stored as an unisgned integer 32 bit starting Thu, 01 Jan 1970 01:00:00 +0100.

The idea was to be able to measure time for a “long” period of time, from now (in the 70s) and in the past.
So they decided to use half of the 2^32 for the past (prior to 1/1/1970) and the other half for the future.

So be it…

Now we are about 40 years later, and we still use the same representation. We already had a party for the unix timestamp 1234567890 (which is quite stupid because it’s in base 10).

And now javascript, Java, PHP and C … use the unix timestamp… which means ajax, web 2.0, comet … which means: most of the online banking system, internet sites and all the smalls scripts that in few line check your credential and your billing to allow you to connect to the Internet… All the logging system, all the monitoring systems… well everything that you don’t see, just under the internet sea you surf on.

So what will be the next party thema?

• the end of the internet as we knew it
• the end of the unix system
• the 0×80000000 seconds since epoch?

Anyway, because of the ongoing ecological breakdown, the human kind will probably collapse before… so after the Y2K38 bug, the world will just be their kingdom. The kingdom of bugs.

Unless we spend time (which is running) in each of our application to think about that right now (and if we spend just about the same amount of time solving the earth ecology problem it will not be useless).

Cindy Foster – CindyFoster1533@live.com
to quechupxxx@xxx.sbw.be

H! I find your post interesting. I am pretty chick. I am moving to your area from russia
I am looking for a good man to show me around and may be more ;)
Could you please kindly reply to me at jclapptakomapark@gmail.com
I will send you some of my pictures back in reply.
Could you reply to my email account I am supplying in the body of this email as I am writing from my work email which i don’t use for personal correspondence

Windows Live™: Keep your life in sync. Check it out.

–> Don’t use quechup