Category: tagged

  • “Unable to load stream. Please make sure port 7446 is open on your NVR.”

    In case you are seeing the error message “Unable to load stream. Please make sure port 7446 is open on your NVR.” when trying to watch the live camera stream on your Unifi Video NVR system from Ubiquiti, then the solution is rather simple.

    The error is caused by the self-issued certificate that comes by default with the Unifi NVR. As Mozilla Firefox has strict SSL (HTTPS) rules making sure that one does not connect to the wrong or fake website as could happen in e.g. a phishing attempt. I have not tried myself but from what I have read apparently Google Chrome does not enforce this in the same manner making the stream work out of the box in Chrome.

    The solution is to open the URL to the video stream directly in a separate tab or window so that you can accept the self-signed certificate for the session:

    https://NVR_IP_address:7446

    The page itself will return an error after the certificate is accepted but that is fine, it is simply to accept the certificate. After that simply reload the NVR interface and the HTTPS video stream will work fine

    Please note that if you are accessing the NVR from an external network you may have a firewall instance blocking the actual port. This instead would require port forwarding to be set up and that is a different story.

  • Running several Firefox versions simultaneously

    Ever been in the need of running more than one instance of Mozilla Firefox? Well, I have. Multiple times.

    I use Firefox for my browsing and development. Combining both in the same profile has been a pain many times especially when ones development is intensive leading to high memory and CPU consumption unfortunately ending with a browser crash.

    Some time ago I starting using Firefox Beta since it included the Sync feature built-in. At some point I wanted to revert back to the release version again as the add-ons needed to be ready for Beta as well and it could take the add-on authors some time to update leaving them disabled till then.

    The easiest thing is to run two or more versions simultaneously of Firefox which provides you with increased flexibility, especially if you do web development of any type.

    To proceed simply download another version of Firefox and install selecting the custom choice. Make sure you install it in a different path than your current Firefox installation.

    The key to get more than one instance of Firefox running is to use multiple profiles. To create an additional profile for your next version open the command prompt (if on Windows) and type the following:

    "C:\Program Files\Mozilla Firefox" -profilemanager -no-remote

    Obviously change the path to match your own system if needed.

     

     

    Now the profile manager should be open and you can simply create a new profile and call it something else like e.g. “BETA”. Then click “Exit” afterwards to quit the profile manager.

     

     

    Next you need to edit the launcher of your newest installation of Firefox, by right-clicking its icon in the Start menu and selecting Properties. At the first text box add the following to the path:

    -P yournewprofilename -no-remote

    So it looks something like this:

    "C:\Program Files\Mozilla Firefox Beta\firefox.exe" -P BETA -no-remote

     

    Save and launch your second instance of Firefox by clicking on the same icon. You should now see two different “Firefox.exe” processes in your task manager.

     

     

    In that same screen you were before you can also change the title and icon itself so that it is easier to distinguish which version is which.

     

  • Google+ integration

    I wanted a tighter integration with my blog and my Google+ stream to avoid having to visit two places.

    I have added a little widget which will display the latest posts.

  • Lightbox 2 and XHTML 1.1 validation

    I recently started using the Lightbox 2 WordPress plugin on my Marbella news site to automatically add the zooming overlay effect to certain images. It is really plug & play and works out of the box.

    I did encounter however that my pages stopped validating as XHTML 1.1 which is the standard doctype on all sites I develop.

    The error in question is due to the square brackets (“[” and “]”) being used to link individual images in one single post together as a gallery under the “rel” attribute. The square brackets are not permitted characters and therefore the W3 Validator spits out the following error message:

    character "[" is not allowed in the value of attribute "rel"
    It is possible that you violated the naming convention for this attribute. For example, id and name attributes must begin with a letter, not a digit.

    To avoid this XHTML 1.1 validation problem simply edit the following line in the Lightbox 2 plugin file named lightbox2.php under the “autoexpand_rel_wlightbox” function:

    from:

    $replacement    = '$1 rel="lightbox['.$post->ID.']">';

    to:

    $replacement    = '$1 rel="lightbox_'.$post->ID.'">';
    

    You should now be able to validate that plugin’s output as XHTML 1.1 with no issues.

  • WordPress + CDATA nightmare

    Do you know what is worse than not knowing the solution or even the reason to a problem?

    Well, that must be, knowing that you have experienced the same problem in the past and resolved it, but no matter how much you try, you can not remember at all what the problem was due and what you did to fix it.

    So this evening several hours went by due to a silly and stupid WordPress issue I already fixed once in the past but lost during an upgrade. This time I wanted to make sure I got the solution written down as I would probably forget it again.

    On my Marbella news site I was working on embedding an external flash video inside an iframe, which itself was invoked via Javascript to avoid the W3 Validator complaining about XHTML 1.1 and iframes being taboo.

    Basically it was WordPress’ over-zealous filters that caught a greater-than sign (>) and converted it to the equivalent HTML entity (>). The most annoying thing is that I already have the Raw HTML plugin installed which disables WordPress’ text filtering. And it also worked fine for the rest of the Javascript and HTML tags in the page text. It was just on one single character, namely in the CDATA closing tag with // ]]> converted to // ]]>.

    After searching many different combinations I finally hit it on the nail. I was not alone either.

    The annoying culprit is this little line in /wp-includes/post-template.php, which I once more commented out:

    $content = str_replace(']]>', ']]>', $content);

    For some reason the WordPress developers have not yet fixed this issue after 3 years of it being reported.

    As a big fan of open source I appreciate the time and money spent by volunteers in many similar projects, however this is not the first time I see quick to fix issues hanging around for ages, simply due to people preferring to have an academic discussion on who can provide the most elegant solution, instead of simply fixing the issue first and talk later.

    Sometimes I wonder how projects ever become real projects.

    Anyhow there was a happy ending as you can see for yourself.