20020918: NMAP2 memory issue.

Stonie,

The gn driver (used by gdradr to display the radar data) had a memory leak
previously that I fixed with the 5.6.h distribution on Monday (the crnexz.c
routine caling zlib could exit with memory still allocated to the zlib
inflater).

Each nmap2 session (again would be affected by the gn bug previously)
has up to 8 loop windows....I haven't noticed a problem here, but
I can run it out of memory if I have a bunch of really long loops
(eg 50 radar images in loop1, more in loop2 etc.).

Just to double check....did you download the new 5.6.h distribution or
is this the ppreview release you are using?

Steve

On Wed, 18 Sep 2002, Stonie R. Cooper wrote:

> Steve,
> 
> I've been busy getting ready for a trip, so didn't pay a lot of attention to 
> my workstation today, but I noticed something tonight that I thought I would 
> run by you.
> 
> I have a Linux 2.4.18 system; dual CPU, 512MB RAM, lots of harddrive space.
> 
> I have the KDE (2.2) desktop running ontop of XFree86 4.2.99.
> 
> I have 16 virtual desktops, and I had 6 NMAP2 sessions and one NWX session 
> running.  They had been running all day.  Four of the NMAP2 sessions had 
> autoupdated satellite and radar (with various components layered - surface 
> data, watches, warnings, tropical plots), the fifth was a static VGF 
> displayed, and the sixth was a loop of model data.
> 
> I had ran this way with nawips 5.6.e.1 for days without issue . . . but after 
> a day with 5.6.h, all my memory was swallowed, and my swap was running full . 
> . . in the two years I have had this system, I never recall using swap before 
> - let alone using it all.
> 
> Top showed just a minimal amount of processing (both cpu's were about 90% or 
> more idle) but the nmap2 sessions showed a great amount of memory use across 
> the six.  The kernel could not spawn any additional processes, with messages 
> showing up in /var/log/messages:
> 
> Sep 18 02:10:12 helium kernel: Out of Memory: Killed process 13050 (httpd).
> Sep 18 02:10:19 helium kernel: Out of Memory: Killed process 13051 (httpd).
> 
> As soon as I started shutting down the nmap2 sessions, most of the memory was 
> freed . . . but upon further investigation, I see a gdradr session with 
> nearly 8% of the memory just to itself.
> 
> I was initially going to blame the new 16bit/24bit friendly code for a memory 
> leak . . . but I don't recall gdradr ever having that much memory usage, 
> either.  Any thoughts?  Have you seen this?  Incidently, I'm running X in 
> 16bit mode to balance speed and resolution.
> -- 
> Stonie R. Cooper
> Planetary Data, Incorporated
> ph. (402) 782-6611
>