[mythtv-users] Shows Chopped Up and Long Queue of Comm Flag / Metadata Jobs

[George Galt]

On Tue, Aug 6, 2013 at 12:25 PM, Gary Buhrmaster
<gary.buhrmaster at gmail.com>wrote:

> On Tue, Aug 6, 2013 at 2:36 PM, George Galt <george.galt at gmail.com> wrote:
> ....
> > In defense of process monitors, I run one not because myth is unreliable,
> > but because it is so reliable that it may take several days or more
> before I
> > notice it isn't running -- by which time I've missed a few of my wife's
> > favorite programs.  Since I don't watch TV every night (or sometimes for
> > more than a week, even when home), I can't otherwise monitor the system.
>  Do
> > you have any other suggestions if a process monitor isn't the preferred
> > strategy?
>
> Monitoring is not (itself) a bad thing.  However, what you do about
> a failure detection is important.  I know the common "Windows user"
> approach is to reboot, and hope the problem goes away.  It might
> (especially with commodity hardware that occasionally just
> fails in unusual ways) but you have not addressed the root cause.
> The reality is that MythTV has bugs(*).  To improve MythTV, there
> needs to be reliable and useful (and hopefully repeatable) reports
> of problems.
>
> So, rather than just restarting your current release(**) MythTV,
> why not have your monitoring system perform the appropriate
> debugging steps as as documented at http://www.mythtv.org/wiki/Debugging
> so that bugs can be reported, and fixed?  Note that for the
> backtrace to be really useful, you will need to install the debug
> symbols.  Except in the most extremely storage constrained
> system, installing the debug symbols is a no-brainer.  Just
> install them, and have your package system maintain them
> as you upgrade so you will have them when you need them.
>
> Gary
>
> (*) There have been great strides over the last few development
> cycles to eliminate deadlocks and to deal with potential errors
> detected by static analysis, along with the usual bug fixes.  But
> there is always more to do, and MythTV is a very complex piece
> of software, not all code paths have been extensively tested.
> And any complex piece of code has bugs.
>
> (**) If you are running older releases, first move to the current
> release.  As stated above, many fixes go in at each release,
> and the devs do not have the resources to deal with bugs that
> have already been fixed in later releases.
> _______________________________________________
> mythtv-users mailing list
> mythtv-users at mythtv.org
> http://www.mythtv.org/mailman/listinfo/mythtv-users
>

Gary:

You are right, of course.  I'll look into installing the debugging symbols
and getting the proper information.  Normally, I have no issues with myth
failing, but I have had an intermittent problem with it crashing that is
associated with one particular program (the Graham Norton Show on BBC
America, recorded through an HD-PVR).  I'll see if I can get a proper
backtrace for that issue and submit it.

Thanks,

George
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.mythtv.org/pipermail/mythtv-users/attachments/20130806/ede9a7a4/attachment.html>