I just found this very inspiring and interesting talk about marketing, presented for developers by Seth Godin (held at Business of Software 2008).
Today I solved a problem I had with check_apachestatus_auto.pl executed via NRPE.
Running the command as root worked fine but, running as user “nagios” just gave no output – which translated to the error message “NRPE: Unable to read output error” on the Nagios NRPE Client.
After checking file permissions, any debug modes and logs i found… i finally got to /tmp and there was it: “127.0.0.1……”, a temporary file created by that particular check plugin. Since i tested the plugin first as root, the file had root-only permissions, and the plugin executed as nagios just stopped somewhere while not being able to use/overwrite that file.
I deleted that particular file, and everything works fine now…
Zend Studio 7.0 brings a host of new features and enhancements that will help you develop faster, resolve defects more quickly, and take advantage of the latest PHP technologies directly from your development environment.
With full support for PHP 5.3, greatly enhanced source code editing, easy debugging through integration with Zend Server, code generation through integration with Zend Framework, and improved performance, Zend Studio maintains its position as the leading solution for professional PHP developers.
New features in Zend Framework 1.9:
- Complete support for PHP 5.3 as well as 5.2 means developers can use the latest PHP language features in their Zend Framework-based apps
- RESTful web services: now made easier through automated routing/detection
- Message queues: useful for offload processing (credit card transactions, media uploads), cross-platform communication, user messaging features, and more.
- LDAP: Microsoft ActiveDirectory & Novell, plus searching, filtering, and tree features
- RSS & Atom: consume these popular feed formats using a common API and higher performance cached HTTP
- DBUnit support: DBUnit’s test data setup and teardown make unit testing Zend Framework applications much easier
Downloads are running, Testsuites will show if ZF 1.9 breaks anything, and I will take the opportunity to fix some plugin issues with Zend Studio while upgrading
Complex Liquid Admin template that contains a login page as well as a modular content page that includes all the elements you could possibly need in your admin. From these elements you should be able to generate any required page.
Please open the preview to check out all the elements.
Liquid as oposed to a fixed size means all the screen real-estate is used in the most efficient manner possible and also eliminates compatibility issues with non-standard, very low or very high screen resolutions.
Tableless CSS coding, cross-browser compatible and very stylish. Impress your customers today !
The template includes all page .psd files as well as slices and should be Easy to customise with medium knowledge of Photoshop and CSS .
Later edit : This template also contains rectangular notification boxes (ex. positive, negative, site message etc.) like our FlexiAdmin, however the shots don’t show this. So if you need these no worries, you will find them in the download package.
I’m impressed with the details and “amount of work” you get for that small price. I’m definitely going to use this in some projects – at least in my own
My newest toy arrived today: OCZ Apex SSD 120Gb – a very fast Solid State Drive. The manufacturer tags it with “230mb/s read, 160mb/s write, <0.3ms access time”.
After my six hour hassle of (trying to…) moving the boot partition from my old harddrive to the new one (see the full report/steps below), I was curious about the first performance test, which I used HD Tune for.
This is the result for my old harddrive, a Samsung HD642JJ:
HD Tune: SAMSUNG HD642JJ Benchmark
Transfer Rate Minimum : 55.0 MB/sec
Transfer Rate Maximum : 102.8 MB/sec
Transfer Rate Average : 83.0 MB/sec
Access Time : 13.2 ms
Burst Rate : 135.3 MB/sec
CPU Usage : 4.5%
Temperature : 25°C
And this is the result for my new SSD (OZC Apex SSD 120GB):
HD Tune: OCZ APEX_SSD Benchmark
Transfer Rate Minimum : 1.4 MB/sec
Transfer Rate Maximum : 172.7 MB/sec
Transfer Rate Average : 138.1 MB/sec
Access Time : 0.2 ms
Burst Rate : 129.5 MB/sec
CPU Usage : 5.5%
Temperature : 44°C
As I said before, I moved my bootpartition to the new drive. These are my subjective feelings about the new system:
- The Windows bootup isn’t that much faster, but I didn’t measure the startup time of the old harddrive. (Still too long…)
- Applications start much faster now. I think this really is the point where the ridiculous low access time of the SSD kicks in. Zend Studio i.e. is down from ~30 secs to 12. (And I haven’t switched the projects/workspace to a new SSD partition yet!)
The application startup speed, and working with big filecount-intensive projects were the main reasons for me to try out a SSD. So far (this still is my impression after only 2-3 hours of usage) I am very satisfied with the results.
The temperature of the SSD is ~20° warmer, but I think even 44°C is not high enough to care about.
In about one month I will provide you with even more subjective feedback
The steps for the Vista bootpartition-move, using Acronis True Image, should be these:
- Boot from the Acronis True Image CD
- Backup your boot partition from the old drive
- Restore the backup to the new drive, mark the new partition active (Cloning didn’t work, because the SSD is much smaller, and the UI doesn’t let me clone just one partition)
- Reboot from your Vista Installation disc
- Repair the installation on your new drive (the log should show something about “missing bootmanager – fixed”)
- Boot from your new SSD
- If your bootpartition is not C, fix it this way: How to restore the system/boot drive letter in Windows
- Done. Finished. Hopefully.. (At least, you should have 2 layers of backups to restart if something screws up…)
Just two small things OCZ could improve:
- Inform me as a customer, that there is no SATA cable in the retail box. Every other retail (SATA) harddrive I bought up to this day had at least one packed.
- As most of the customers will be “switchers” imho, ship something to help with the partition migration process. I will happily pay some more bucks, if it just works.
I got this errormessage on “some” places while trying to install an AIR application package, while other places just worked fine.
The application could not be installed. Try installing it again. If the problem persists, contact the application author. Error #0
The first step debugging this was to find out where the AIR application installer logs errors – which by default he does not. To activate installer logging, you have to put two files in your home directory (C:\Users\youraccountname\ on Vista or C:\Documents and Settings\youraccoutname\ on XP) called .airappinstall.log and .airinstall.log.
If those two files are present, the logging writes to them automagically.
The error I found there looked like this:
Installing C:\Dokumente und Einstellungen\***\Lokale Einstellungen\Temp\fla12.tmp\setup.msi Error occurred; beginning rollback: [ErrorEvent type="error" bubbles=false cancelable=false eventPhase=2 text="1603" errorID=0]
From here on I was very sure that it has to do with privileges… (No, there is no context menu entry on AIR packages to “install as Administrator” or “run as”…)
Finally, I checked the windows event log, and found this:
Product: *** -- Error 1925. You do not have sufficient privileges to complete this installation for all users of the machine. Log on as administrator and then retry this installation.
Why can’t they show this error at the installer instead of error #0?…
Yes, installing as Administrator solved the problem!
Over the years, I had to do html scraping sometimes. At my very first coding attempts some years ago, I found the explode() function to be pretty nice for this purpose. Not being satisfied with its (nonexistant) fault-tolerance, I learned regular expressions and used them. But regex only gets you so far in the DOM-tree. Finally, I arrived at the DOMDocument class, the builtin PHP answer to DOM parsing/manipulation. Looking into this is really worth it, because it a) solves the problem (99.99%) and has a nice OOP way to achieve it.
I will show you two very simple examples with all three methods – see for yourself why one should use DOMDocument
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="de" lang="de"> <head> <title>example</title> </head> <body> <div id="fetchme"> foobar </div> </body> </html>
// explode $dataFirst = explode('<div id="fetchme">', $sample); $dataSecond = explode('</div>', $dataFirst); $result = trim($dataSecond); echo $result; // regex preg_match('#<div id="fetchme">(.*)</div>#is', $sample, $matches); $result = trim($matches); echo $result; // domdocument $doc = new DOMDocument(); $doc->loadHTML($sample); $element = $doc->getElementById('fetchme'); $result = trim($element->nodeValue); echo $result;
All three methods get this very simple example right ($result will be ‘foobar’).
Adding complexity, two of them fail badly, because they do not know which closing tag belongs to which opening tag.
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="de" lang="de"> <head> <title>example</title> </head> <body> <div> <div id="fetchme"> foo <div> bar </div> </div> </div> </body> </html>
The scraping methods are the same as for the first example, but with different output:
The explode method returns: ‘foo <div> bar’
The regex method returns: ‘foo <div> bar </div> </div>’ (the non-greedy regex would return the same as the explode version).
The DOMDocument method returns ‘foo bar’, because nodeValue strips htmltags. But the DOMElement $element holds just the content of the div #fetchme as we wanted it. It gets it right again.
You should know, some sites do not follow the (x)html standards correctly and therefore DOMDocument can fail. It also likes to throw exceptions then, but besides complaining you can still use > 90% of the document. If you want to disable the exceptions, you could either do it with @$doc->loadHTML() or $doc->strictErrorChecking = false.
Besides DOMDocument solving the tag-matching issues, it also provides a much cleaner way to do HTML scraping jobs imho. It also is a lot more fault tolerant, one could use xpath-queries against the document, etc.
# add the backport mirror for java and subversion echo "deb http://www.backports.org/debian etch-backports main contrib non-free" >> /etc/apt/sources.list # add the backport mirror key gpg --keyserver hkp://pgp.mit.edu --recv-keys 16BA136C gpg --armor --export 16BA136C | apt-key add - apt-get update # install backports (java and subversion) apt-get -t etch-backports install sun-java6-bin sun-java6-jre apt-get -t etch-backports install subversion # install apache (mainly for phpmyadmin) and php5 + cli apt-get install apache2-mpm-prefork apt-get install php5 php5-cli php5-dev apt-get install php-pear make # install xdebug (needed for phpunit) pecl install xdebug echo "zend_extension=/usr/lib/php5/20060613+lfs/xdebug.so" >> /etc/php5/cli/php.ini # install phpunit and phpundercontrol via pear pear upgrade --force pear pear channel-discover pear.phpunit.de pear channel-discover components.ez.no pear install phpunit/phpunit pear install --force --alldeps channel://components.ez.no/Graph pear install --force --alldeps channel://pear.phpunit.de/phpundercontrol-0.4.7 # get and extract cruisecontrol apt-get install unzip wget cd ~ wget http://freefr.dl.sourceforge.net/sourceforge/cruisecontrol/cruisecontrol-bin-2.8.2.zip unzip cruisecontrol-bin-2.8.2.zip -d /opt cd /opt ln -s cruisecontrol-bin-2.8.2 cruisecontrol # run phpundercontrol modifications against cruisecontrol phpuc install /opt/cruisecontrol # first testrun cd /opt/cruisecontrol ./cruisecontrol.sh
Since the first time I have seen Balsamiq Mockups, it rocked my world. It is now a pleasure to create mockups. With Mockups, this is a fast and intuitive process, for everyone. Yes, I like this tool as a developer, and I am sure that any non-developer will also like it as much!
There are a lot of shapes one can choose like a browser window, iphone window, any other window, webcontrols, etc. The result can be exported as PNG or saved as XML. With XML files one can even diff/merge mockups
- Create software mockups in minutes
- Collaborate with your team
- Focus on creating your product
The guy/company behind Mockups, Giacomo ‘Peldi’ Guilizzoni, is even more impressive. There is a very good interview over at 37 signals (see Sources):
“I never really understood the concept of building a company with the goal to sell it, or why one should have an ‘exit strategy.’ I just don’t get it. If you’re doing what you love, why would you want to ‘exit’? Maybe it’s because I am Italian, but I see nothing wrong with a business staying small in the long run. As long as we do great work, are happy to do it, and make people happy with it, I see no reason to change anything.”
Thank you again for the kind response to my email, Balsamiq Team, and the chance to review this very nice (and therefore rare) piece of software.
I did this to “outsource” my feed traffic and improve statistics for analyzing what you like to read
Yes, I also moved the subscribe button/links to a more prominent position (top right, larger icon).
Update: Automagic redirection in place. You shouldn’t need to update anything