[mythtv-users] Moving from Fedora to Ubuntu
painethom at gmail.com
Wed Mar 14 22:40:12 UTC 2007
> Honestly, long term stability has a lot to do with not mucking around
> with something once it's working. If it ain't broke, don't fix it.
This is what I have been doing since getting my install running in
Sept 2005. I only updated it once and that was to get Myth .20
installed. It runs fairly reliably and I seldom touch it because I am
well aware of the if it ain't broke addage. :)
> Have you considered fixing your glitches? I know Linux can be a fickle
> friend, but when I see messages that say something like "I've already
> installed $LINUX_DISTRO six times trying to get mythtv to work", it
> makes me wonder which is more effort - fixing the problem or OS
I tried a few things to fix my problems, but currently both my FE and
BE are on FC4. That is now unsupported and I want to get to something
that is supported.
The only annoying glitch on the FE is that it hangs loading X the
first time going in. It gets part way through the X loading screen
where KDE is coming up and it freezes. To fix this I just have to ssh
into the box, init 3, modprobe lirc_serial, service lircd restart, and
init 5. I don't figure this glitch is wirth the effort of risking a
working machine for a broken one since my wife is home on maternity
> If you want to keep the backend light, you can leave off X and
> windowmanagers completely. I wouldn't necessarily say that *MOST*
> popular distros are significantly more or less bloated than any
> But nowadays, especially after seeing the typical hardware
> requirements for HDTV mythtv boxes, one has to wonder whether how much
> time and effort to invest in a lightweight setup vs. "thowing hardware
> at the problem".
I have no problems throwing hardware at problems like this, but I
don't think I have hardware issues at this point.
> I mean, seriously, how many people would do a linuxfromscratch build
> of mythtv just to get a few % better performance or save less disk
> space than an hour of HDTV takes? (NOTE: Linuxfromscratch.org is very
> cool and I think everyone interested in Linux should build their own
> system *ONCE* just for the sake of learning way more than you'll ever
> learn from doing apt-get install foo)
I'm not so concerned with disk space, I have 1.8TB currently. I was
thinking more of a performance increase on my machine for transcoding
commercials and stuff. I've noticed my machine is recording more and
more tv everyday. When I first installed it, it recorded a couple of
shows a day. Now it is doing over 20, auto-deleting some like the kids
cartoons and recording new episodes and stuff.
> As far as eliminating computer noise from the TV room... well there's
> an area where it's probably worth the effort keep things "light" so
> that you reduce background noise.
> If you treat your mythtv box like a production server, then you
> probably want to minimize the number of changes done at once and make
> sure that your systems boots, functions properly after a given
> changeset before you go off changing more things. This way, if there's
> an error or problems, you'll know it's related to the ONE thing you
> just changed or upgraded.
> Servers are so cheap and powerful these days that I'm starting to
> think it's prudent to have a test/dev mythtv box and a production
> mythtv box. Do your new OS install on the test/dev box, get it
> running, get hardware detected, get lirc configured, get mythtv
> working, import your old database, etc. THEN... with a working
> template/fallback system and that experience under your belt, you're
> prepared for planning to upgrade your production mythtv system
I try to treat my Mythtv hardware like an appliance and not muck with
it if I can help it. I think that is part of the reason that I have
such good success wuith mine.
More information about the mythtv-users