Re: [Brahms-dev-l] dst2tree and memory

Hi,

wow,  1200MB ++ seems alot... I remember long ago seeing around 400MB...
hmmm, well several changes have been made since. My guess is that it
leaks a histogram. Many histograms are remade every time it starts on a
new DST, but before that the histogram that only dealt with the previous
run number is written to file and deleted. So it should use a constant
amount of memory.

But of course it could be something more tricky...

Cheers,
	Truls

On Mon, 2007-03-12 at 14:43 -0400, Flemming Videbaek wrote:
> I think it has a leak, the mem. use has grown from 1280 to 1486 in the last 3/4 hour for the particular task I am watching
> 
> flemming
> 
> 
> --------------------------------------------
> Flemming Videbaek
> Physics Department 
> Bldg 510-D
> Brookhaven National Laboratory
> Upton, NY11973
> 
> phone: 631-344-4106
> cell:       631-681-1596
> fax:        631-344-1334
> e-mail: videbaek @ bnl gov
> ----- Original Message ----- 
> From: <i.c.arsene_at_fys.uio.no>
> To: "Flemming Videbaek" <videbaek_at_bnl.gov>
> Sent: Monday, March 12, 2007 2:30 PM
> Subject: Re: [Brahms-dev-l] dst2tree and memory
> 
> 
> >       Hi Flemming,
> > 
> >    I don't know if this can be the reason, but dst2tree from the new
> > banapp fill lots of diagnostic histograms (if the setting contains
> > many runs, can be thousands of histograms). Up to now i did not
> > encountered any problem with memory leaking.
> > 
> >                       Regards,
> >                 Ionut
> > 
> >> Hi
> >>
> >> I am curious if the dst2tree has memory leaks . Several jobs are running
> >> right now (catalin), that consumes ~1200Mby of virtual memory.
> >> Is that the normal condition ? I am not requesting that any jobs be
> >> stopped, but will like to understand. In fact one machine rcas0062 dies
> >> earlier today
> >> because it ran out of swap space.
> >>
> >> flemming
> >>
> >>
> >>
> >> --------------------------------------------
> >> Flemming Videbaek
> >> Physics Department
> >> Bldg 510-D
> >> Brookhaven National Laboratory
> >> Upton, NY11973
> >>
> >> phone: 631-344-4106
> >> cell:       631-681-1596
> >> fax:        631-344-1334
> >> e-mail: videbaek @ bnl gov_______________________________________________
> >> Brahms-dev-l mailing list
> >> Brahms-dev-l_at_lists.bnl.gov
> >> http://lists.bnl.gov/mailman/listinfo/brahms-dev-l
> >>
> > 
> >
> _______________________________________________
> Brahms-dev-l mailing list
> Brahms-dev-l_at_lists.bnl.gov
> http://lists.bnl.gov/mailman/listinfo/brahms-dev-l
> 

_______________________________________________
Brahms-dev-l mailing list
Brahms-dev-l_at_lists.bnl.gov
http://lists.bnl.gov/mailman/listinfo/brahms-dev-l