E
Erwin Dahl
I have an all python (ie no c code) daemon program that sits in a loop
doing stuff (pinging machinies). It had been running fine in
python-2.1.3 for over a year. Last week I upgraded python to 2.2.3
and it started leaking memory very quickly (over 100 pages per loop).
I fired up a debug version of python-2.2.3 and I see that the object
ref count stays the same across loop iterations. I also tried 2.3.2
and it had no leak problem. Now I know that python-2.2.3 isn't the
current version but its what I need as a result of some of the
libraries I'm using.
My questions are:
- is this really a leak in python (or one of its libraries)?
- if gettotalrefcount() stays the same it must be a C level leak,
correct?
- if so what is the best way to figure it out?
I'm on solaris 2.8 and python is built with gcc 3.2.2. The program
doesn't use many libraries the ones that I haven't been able to pull
when debuging are socket and select so I suspect them.
-EAD
doing stuff (pinging machinies). It had been running fine in
python-2.1.3 for over a year. Last week I upgraded python to 2.2.3
and it started leaking memory very quickly (over 100 pages per loop).
I fired up a debug version of python-2.2.3 and I see that the object
ref count stays the same across loop iterations. I also tried 2.3.2
and it had no leak problem. Now I know that python-2.2.3 isn't the
current version but its what I need as a result of some of the
libraries I'm using.
My questions are:
- is this really a leak in python (or one of its libraries)?
- if gettotalrefcount() stays the same it must be a C level leak,
correct?
- if so what is the best way to figure it out?
I'm on solaris 2.8 and python is built with gcc 3.2.2. The program
doesn't use many libraries the ones that I haven't been able to pull
when debuging are socket and select so I suspect them.
-EAD