JAP_WS_LOCAL.py seems memory leak?
Opened this issue · 8 comments
GoogleCodeExporter commented
What steps will reproduce the problem?
1.remote:JAP_WS_REMOTE_PYTHON_OPENSHIFT-2.0.0-SNAPSHOT-3
2.local: JAP_WS_LOCAL_PYTHON-2.0.0-SNAPSHOT-2
3.run: python2 JAP_WS_LOCAL.py
What is the expected output? What do you see instead?
At firest,the process takes up 14M memory.After watching Youtube for about 2
hours,it eats 96M.
What version of the product are you using? On what operating system?
os: Archlinux + Xfce4.10 + Chromium
- python2 2.7.3-4
- twisted 12.3.0-1
- python2-pyopenssl 0.13-2
- autobahn 0.5.9
Please provide any additional information below.
Maybe memory leak,or bad garbage collection.
Original issue reported on code.google.com by westmin...@gmail.com on 25 Jan 2013 at 2:35
GoogleCodeExporter commented
Can you test attached JAP_WS_LOCAL_PYTHON?
Thanks :)
Original comment by jeroen.v...@gmail.com on 25 Jan 2013 at 8:54
Attachments:
GoogleCodeExporter commented
Better,but still the same that only taking up memory from 14M+ to 60M+ slowly
in 3 hours browsing.Why don't recover the memory?BTW,APPP always uses about 15M
memory.
Original comment by westmin...@gmail.com on 26 Jan 2013 at 3:46
GoogleCodeExporter commented
I think the problem is PYOPENSLL/OPENSSL:
- http://twistedmatrix.com/trac/ticket/5487
- http://journal.paul.querna.org/articles/2011/04/05/openssl-memory-use/
Do you have any ideas?
Original comment by jeroen.v...@gmail.com on 26 Jan 2013 at 5:08
GoogleCodeExporter commented
Sorry,I'm just an end-user,know little ABC programming.
The test version seems better as leaving the process idle for a long time it
releases memory from 60M+ to 24M+,but keep using,it only keeps increasing.
Original comment by westmin...@gmail.com on 27 Jan 2013 at 12:49
GoogleCodeExporter commented
Maybe memory fragmentation?
http://revista.python.org.ar/2/en/html/memory-fragmentation.html
Original comment by jeroen.v...@gmail.com on 28 Jan 2013 at 9:13
GoogleCodeExporter commented
IRC log (irc://irc.freenode.net/twisted)
Original comment by jeroen.v...@gmail.com on 29 Jan 2013 at 6:29
Attachments:
GoogleCodeExporter commented
So what's wrong?Upstream issue?
Original comment by westmin...@gmail.com on 31 Jan 2013 at 1:32
GoogleCodeExporter commented
It could be normal because of memory fragmentation (and the way the python VM
works).
Original comment by jeroen.v...@gmail.com on 31 Jan 2013 at 1:56