Open GoogleCodeExporter opened 9 years ago
Thanks for this bug report and the kind words that come with it.
I used the pmap tool and got a perfectly stable output, at least for the 15
minutes I let Flowblade run:
0000000000400000 2804K r-x-- python2.7
00000000008bc000 4K r---- python2.7
00000000008bd000 468K rw--- python2.7
0000000000932000 72K rw--- [ anon ]
000000000120c000 18988K rw--- [ anon ]
00007f2730000000 132K rw--- [ anon ]
00007f2730021000 65404K ----- [ anon ]
00007f2738000000 132K rw--- [ anon ]
00007f2738021000 65404K ----- [ anon ]
00007f273fff8000 65540K rw-s- pulse-shm-1799224795
00007f2744000000 132K rw--- [ anon ]
---etc
So this issue cannot be reproduced on my system.
First I would like to find out if this is the same bug that was discussed here:
http://sourceforge.net/p/mlt/bugs/221/
This bug has been fixed in MLT. To test if this fixes your issue too, you can
compile and install the repository version MLT.
Below is URL to page that has a script that builds and installs repository
versions of both Flowblade and MLT into your home folder. You can remove them
just by deleting the folder to continue using your installed version. The
dependency list may not be up to date, but it will get you started.
http://www.mltframework.org/bin/view/MLT/BuildScripts
Original comment by janne.li...@gmail.com
on 3 Feb 2015 at 9:07
Thank you very much for your feedback and help!
The MLT-bugreport describes indeed a very similar problem, Fortunately in my
case the memory-usage was nowhere increasing so extremely rapidly (about
60MB/s) as in the bugreport, although the result was the same - the process
consumed all the available memory and swap.
The buildscript for MLT and Flowblade worked fine after some small adjustments:
gerhard@paisley:~/flowblade/20150204$ ./start-flowblade
FLOWBLADE MOVIE EDITOR 0.17 repository/dev
---------------------------
Launch script dir: /home/gerhard/flowblade/20150204/flowblade-trunk
Running from filesystem...
MLT found, version: 0.9.3
...
and after almost half an hour everything, especially the memory-usage seems to
be running "under normal conditions". I guess I can confirm now that using MLT
0.9.3 instead of 0.9.2 has been the solution for the problem here.
Thanks again!
Original comment by Vienna...@hotmail.com
on 4 Feb 2015 at 1:58
Original issue reported on code.google.com by
Vienna...@hotmail.com
on 31 Jan 2015 at 9:13Attachments: