Tag Archives: apache

MythTV Install and Export to iPhone

My MythTV backend is now working the way I’d like it to.  I bought a HDHomerun off newegg for the 2 tuners that both handle ATSC and QAM and I like that it’s on the network.  I have an Ubuntu machine running Karmic that I wanted to put the Myth backend on.

The most awesome thing is that there is a MythTV package in apt.  So the install was simple.  Since I couldn’t remember the mysql root password (because how often do you add databases/tables?) I had to override password and set it to something I knew and clear and reinstall the MythTV database package.  With that done, setup was pretty straight forward:  finding the tuners, scanning channels, adding a schedulesdirect.com account ($20 per year for listings since zap2it won’t do that for free anymore).

The next part was exporting commercialless recordings into iPhone format.  There is a package (again through apt) called mythexport that claims to do this.  It handles jobs started from the Myth frontend pretty well, but required a lot of tweaking.  Using a web browser, go to localhost/mythexport and set up some initial settings for what you’re trying to do.  Any jobs created probably fail now and you’ll have to make some changes.  First add the medibuntu repository to apt and update the codecs to regain AAC audio if you’re running Karmic (apparently not a problem before).  The command to reencode the video is in /etc/mythtv/mythexport/mythexport_settings.cfg.  It’s the long line, you can’t miss it.  So change the mp3 library to libfaac.  Try exporting something, /var/log/mythtv/mythexport.log will not show you the exact errors, but will give you the command to run to try again if it did fail (it’s the command starting with “nice”).  Copy it out and try running it yourself with different arguments until it works.

My settings are currently (the  “-ac 2 -ar 48000” was important and I had to add it):

ffmpegArgs=-y -acodec libfaac -ab 128kb -vcodec mpeg4 -b 600kb -mbd 2
-flags +4mv+aic -trellis 2 -cmp 2 -subcmp 2 -s 480x320 -aspect 16:9
-ac 2 -ar 48000

And now it works!

Also note that mythexport adds an Apache2 directory to your configuration.  I had the Apache I got from apt configured as a public facing webserver so I had to lock down those directories with .htaccess files (iTunes will ask for the password when downloading videos in the podcast/rss feed so it’s really not too limiting to do BasicAuth).

WordPress Online

EDIT 8/10/09: I’ve been told on the ale.org list that ubuntu’s wordpress package has an older version of wordpress and has not been updated recently. So for now I’ll recommend NOT yet using apt to manage your wordpress install.


otherroute.net is back online.  And after doing some work getting a lot of different settings right, I realized that it could have been much easier.

sudo apt-get install wordpress

Yup, that’s all it takes.

My last webserver was an Xbox that ran Debian.  To set that up, I found all the source files I needed, configured and patched things, compiled and installed.  It was a familiar exercise of trying to compile only to find what’s missing, downloading more compressed tar files, and trying to compile everything again.  An afternoon or two later and I had a moderately working Apache webserver to host my personal website, projects for school, and other things that wanted to run on Linux.  And then I learned about apt, which makes life easier by doing all those things for you.  A single command and you’ve got Apache running.

I wanted to avoid that unnecessary work this time around when getting WordPress running on my new dedicated Ubuntu machine.  I was able to make good use of apt.  It found Apache2 for me and installed it.  It found PHP and MySQL and installed them too.  And then I went off configuring things.  I made a database user and databases.  I worked on the /var/www folder tree where I had put WordPress, setting permissions so that Apache could do everything it needed but nothing more.  I turned on Apache’s mod-rewrite when the usual WordPress URL formatting wouldn’t work, and then turned on an Apache permission for the site when mod-rewrite still wasn’t working.  Before calling everything complete, I wrote a script that would download WordPress’ latest.tar.gz and update the site.  And then I found that apt could have done all that and the updates would have been integrated in my usual system update process.

I had again wasted a couple afternoons configuring things that I could have just had the system do for me.  I hadn’t thought about a different way to do things at the time.  I knew how to do get the job done, so I started working the way I knew how.  Almost exactly like last time, after I was done I found that someone had provided me with a much easier way to get things started and in the future keep them up to date.  This time around, I thought I knew the tools available to me, but my knowledge was dated and the end result was the same as last.

But now I’ve got a new strategy that I learned from this exercise that I didn’t learn before.  Before doing things that I know how to do, I need to reevaluate that my known way is the best way.  Things change too quick, and nothing from a few years ago should be applied to technology today.  Sure, knowing the hard way by hand, like configuring WordPress or compiling Apache, helps in understanding what’s going on, but there’s better, popular and well-known ways to get that done.  And check apt for everything, even webapps.