[mythtv-users] ticket locking

[Michael T. Dean]

On 06/18/2008 02:07 PM, Brad DerManouelian wrote:
> On Jun 18, 2008, at 10:20 AM, Udo van den Heuvel wrote:
> <a bunch of stuff that doesn't matter>
>
> I can guarantee you that arguing that you have provided enough  
> information even though you have no provided the requested information  
> will not result in an investigation into your issue. I can also  
> guarantee you that the longer this thread goes on, the less likely  
> someone will take pity on your environment, stop whatever else they  
> are doing, set up an environment that matches yours, run it for a  
> couple of days to watch memory usage climb, look into the source code,  
> find the problem, fix the problem, test the fix, post the patch and  
> wait for it to be accepted into trunk.

Especially since there's a very good chance--if you actually have a
memory leak, and I'm not sure I believe you do, at this point--your
ticket was a duplicate of another (i.e. perhaps #5410, an EIT-related
memory leak-- http://svn.mythtv.org/trac/ticket/5410 ).  If you look at
that ticket, you'll see that it's not closed...  Perhaps the difference
is that the reporter for #5410 actually gave the information required to
find out what's happening.

However, on the bright side, if my suspicion is correct, you'll see the
fix when #5410 is fixed.  Then again, you might just see the same
behavior after #5140 is fixed--i.e. because you don't have a memory leak.

Of course, this is just a blind guess (as is the prevailing theory in
this thread that your /perceived/ (but unconfirmed) memory leak is due
to your aggressive recording schedule) as I'm lacking the information
necessary to do any real diagnosis.

Oh, and /please/ re-read Daniel Kristjansson's message.  It said
everything that needed to be said: anecdotal evidence with no Valgrind
output, virtual DVB recorders using buffer memory.

Mike