emBee --

I had this fixed on my copy of it (which, sadly, was lost in a crash).

The problem is that the xmlCache is being instantiated multiple times. That
thing chews up memory, and the current design (last I checked) still creates
multiple ones. Making it a singleton (i.e. creating it once and then caching
it for subsequent uses) fixed the massive memory issues in PK).

--sgp

On Sat, Jul 3, 2010 at 4:45 AM, Martin Bähr <
mba...@email.archlab.tuwien.ac.at> wrote:

> hi,
>
> andres has been working on trying to find the packagekit memory leak
> today. it looked like he had it but on my system it still uses to much
> memory
>
> even on my moderate update set (40 jobs) the old and new packagekit died
> with lack of memory. both allocate more than 500MB
>
> my memory looks like this:
> $ free
>             total       used       free     shared    buffers     cached
> Mem:       2047116    1157828     889288          0      59784     745888
> -/+ buffers/cache:     352156    1694960
> Swap:            0          0          0
>
> no swap, but 2GB should still be plenty, the machine is not doing anything
> else. (i am logged in through ssh, so other than the login screen there is
> no
> gui stuff)
>
> i think we need to dig deeper with python memory use profiling...
>
> i initially only sent that to andres, but i am sharing it here in hope
> that there may be others who can provide some insight into memory
> profiling and other comments.
>
> greetings, martin.
> --
> cooperative communication with sTeam      -     caudium, pike, roxen and
> unix
> searching contract jobs:  programming, training and administration -
> anywhere
> --
> pike programmer      working in china
> community.gotpike.org
> foresight developer  (open-steam|caudium).org
> foresightlinux.org
> unix sysadmin        iaeste.at
> realss.com
> Martin Bähr          
> http://www.iaeste.or.at/~mbaehr/<http://www.iaeste.or.at/%7Embaehr/>
> is.schon.org
> _______________________________________________
> Foresight-devel mailing list
> Foresight-devel@lists.rpath.org
> http://lists.rpath.org/mailman/listinfo/foresight-devel
>
_______________________________________________
Foresight-devel mailing list
Foresight-devel@lists.rpath.org
http://lists.rpath.org/mailman/listinfo/foresight-devel

Reply via email to