[MythDora] MythDora 4 Final

[Jarod Wilson]

On Jun 09, 2007, at 21:36, gchris wrote:

> Dennis Hand wrote:
>> On 6/9/07, gchris <gchris at bellsouth.net> wrote:
>>> Guys, I'm gradually recovering from knee surgery and haven't yet  
>>> been
>>> able to get much computer "seat time", but I have gotten just  
>>> enough to
>>> grab a copy of MythDora 4 Final and get it installed.  This is  
>>> without a
>>> doubt the very best Mythtv package I've ever seen or ever  
>>> expected to
>>> see!  It is incredibly smooth and slick in appearance, the menus are
>>> clear and straightforward, and a total install took me 35 minutes
>>> including filling the database.  To my mind this is what Mythtv  
>>> always
>>> should have been but never was until now!  Congratulations on a  
>>> job well
>>> done and and an outstanding collaboration!
>> Well Thank You very much Chris. Jarod and I put in alot of hours
>> getting things just right. A few choice words may have also been said
>> in the process ;-) We're also proud of the way things turned out. I
>> myself think it's the best and most solid version of MythDora.
>>
> I'd go a step further and call it the best and most solid of any  
> Mythtv
> distro I've ever seen and believe me I've seen a bunch!

Woo!

>>> On three occasions the box booted with no audio.  Twice it started
>>> working after I escaped out to the desktop and then restarted  
>>> Mythtv.
>>> Once it started working after I cycled through the three tuner  
>>> cards.
>>> Once working it seems to continue to work.
>> This has been posted on the site since version 3.2. I noticed it
>> myself when Myth went to version 0.20. The fix is (at least for  
>> me) to
>> change the audio output device from /dev/dsp to ALSA:default. Also
>> make sure you change /dev/mixer to "default" without the quotes. This
>> works for me everytime now.
>>
> Where is the script that defines that?  I haven't had much occasion to
> tune alsa stuff previously.

No script, its in the frontend setup somewhere or other.

>>> On two occasions the box warned that it was unable to contact the
>>> mythbackend after it booted.  If I selected something like "display
>>> system status" I found that if I hit the Enter key several times it
>>> would occasionally get status from the backend successfully, but  
>>> then
>>> fail again the next time with a "Unable to contact backend"  
>>> message the
>>> next time Enter was pressed.  Clearly, backend was running but  
>>> frontend
>>> was not able to communicate with it reliably.  Once I cured this
>>> behavior by running Mythtv setup without changing anything and then
>>> restarting Mythtv.  The second time it resolved itself when the  
>>> machine
>>> was powered down and later restarted.
>> I myself have not experienced this but I have heard of just a few
>> saying something vaguely similiar. Anytime you start the frontend  
>> from
>> the desktop shortcut, the backend gets checked then started if it's
>> stopped.
>>
> I just experienced something even wilder.  I just booted the box  
> and had
> no audio (but normal video) in Watch TV.  I escaped out and ran  
> dmesg in
> a terminal window to see if I could spot anything.  Nothing amiss.   
> Went
> back into Mythtv and now I had normal audio and video.  Tried changing
> tuners - I had one and two but no three.  Escaped out of Watchtv and
> went to System Status to view Tuner Status and got the Mythbackend not
> accessible error!  Hit enter several times and could occasionally see
> Tuner Status appear interspersed with unable to access Mythbackend
> errors.  Could no longer Watchtv, it was now blocked by unable to  
> access
> backend errors.  Clearly the backend had been up and running normally
> after boot but my ability to access it went south when I went from
> watching TV to displaying System Status!!!

Hrm... That's a new one to me... Wondering if it makes a difference  
which user you're running the backend as...

>>> A third glitch was that my PVR-350 (third tuner) couldn't be  
>>> selected in
>>> WatchTV for the first several days after MythDora was installed.   
>>> System
>>> Status showed that the tuner was there and not busy but cycling  
>>> through
>>> the tuners with the "Y" key would give me Tuner 1, Tuner 2, Tuner 1,
>>> simply skipping Tuner 3 as though it didn't exist!  Today I  
>>> booted the
>>> machine and all three tuners are working normally.  I haven't  
>>> changed
>>> anything and dmesg and /dev looked perfectly normal when I couldn't
>>> access the third tuner.
>> Can you confirm that this happened during your mythbackend snaffu  
>> that
>> you just mentioned? And has it happened since and how long has it  
>> been
>> since. Sounds like your backend may have just been in a funky state.
>>
> No, I was able to view and record and playback programming from the
> HD3000s at the same time I was unable to access the third tuner so I
> wouldn't tie the backend access problem together with the third tuner
> problem.
>
>>> I have yet to see anything in logs or messages to explain any of  
>>> this
>>> weirdness and the system seems to be perfectly stable once it is  
>>> up and
>>> running so I haven't a clue other than a suspicion that something
>>> strange may be occurring during boot.  Like I said, not complaining,
>>> just observing.
>> So you do get bad symptoms still on a reboot then?
>>
> Yes, randomly.  After any cold boot, any one or more of these problems
> may be present.  Just now I went into setup using the desktop icon (to
> stop mythbackend) and then escaped out and went back into mythtv (to
> restart mythbackend) and this time it did not fix the connection  
> problem
> so I'm completely at a loss as to what is going on here.  BTW, this is
> the same hardware I've been using all along and I've never seen any of
> these problems before this final release so I'm thinking this has  
> to be
> something in the final release software.

Very little changed outside of the 'mythdora' package and the artwork  
for the last two weeks or so, iirc...

>>> One other observation related to the kernel.  I'd mentioned a  
>>> while back
>>> that I'd noted an intermittent failure related to anachron during  
>>> power
>>> down.  That is still happening.  On power down, Init going down  
>>> stops
>>> anachron.  Sometimes but not always, when killall starts toward  
>>> the end
>>> of power down it sometimes also tries to bring down anachron.   
>>> When that
>>> occurs, anachron shows FAIL, probably because it is already  
>>> down.  When
>>> killall starts without mentioning anachron, there is no failure
>>> indication.  This may be a "don't care" thing because it doesn't  
>>> seem to
>>> be causing any problems, it just looks funky.
>> I'll leave this one for Jarod ;-)
>>
> If I were Jarod I'd ask if it happens with Core 7? :-)

What's Core 7? ;)

(There is no more Core since Core and Extras merged, so its actually  
just Fedora 7 now. :)

Oh, and I have no idea. Haven't noticed it, but I haven't been  
looking either.

-- 
Jarod Wilson
jarod at wilsonet.com


-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://lists.atrpms.net/pipermail/mythdora/attachments/20070612/a3b518ae/attachment.bin