[mythtv-users] mythbackend still eats memory: the current status

[Nick Morrott]

On 20/05/2009, Udo van den Heuvel <udovdh at xs4all.nl> wrote:
> Nick Morrott wrote:
> > I haven't had the time to look into it any further than monitoring
> > memory usage and disabliing active EIT last year (I did this 2008-07
> > after initial investigation and the graph is available at
> >
> http://www.insidethex.co.uk/mythtv/mbe-mem-usage-active-eit-disabled.png)
> >
>
>  EIT appears to play a certain role there. Over here the issue was already
> happening before I enabled EIT, but still EIT was in need of fixes at time
> of that graph.
>
>  Hmmm.
>  No VIA problem.
>  No Terratec problem.
>  No Fedora 10 problem.
>  What remains?
>
>  And who can run valgrind?
>  Can you?

No promises, but probably over the summer when the box is doing a lot
less recording, and I have more time available. Although my results
posted earlier in the thread show that there does appear to be
mythbackend memory growth over time, it has not been large enough to
cause any real issues that impact its usability.

The machine should be powerful enough to run valgrind (although I will
have to see during testing) and I will post any results to the list. I
will probably keep that 'production' box on 0.21-fixes for testing,
but I have a second box to play with that I could put trunk on and see
if I can see the same symptoms with that. I'm also planning on doing
some XMLTV-related MythTV development over in the next few months, so
I will be running trunk for that.

Nick

-- 
Nick Morrott

MythTV Official wiki:
http://mythtv.org/wiki/
MythTV users list archive:
http://www.gossamer-threads.com/lists/mythtv/users

"An investment in knowledge always pays the best interest." - Benjamin Franklin