Re: Question on IDL Error: Unable to free memory: from array descriptor [message #34688] |
Mon, 07 April 2003 10:10  |
Dick Jackson
Messages: 347 Registered: August 1998
|
Senior Member |
|
|
Hi Ken,
"Ken Knapp" <Ken.Knapp@noaa.gov> wrote in message
news:6146a9fd.0304030627.361c402e@posting.google.com...
>
> PROBLEM:
> When processing satellite code, sometimes IDL bombs out with the
> following message:
> Loop over image
> % Unable to free memory: from array descriptor.
> Interrupted system call
> % Execution halted at: PROC_TAU 1634
> /home/kknapp/work/lib/idl/goes/gasp9.pro
>
> [...]
>
> SYSTEM:
> Linux Redhat 7.1
> IDL 5.6 (recent upgrade)
> -plenty of RAM (1Gb) and swap
David's idea (too much memory) is plausible, in that it can make us
think we're invincible to memory shortages! A client of mine got a 2GB
machine, and we promptly found a way to exhaust it with some insanely
large images and processing methods.
Since the error message mentioned "array descriptor" I thought of an
issue that came up for me a while ago. The issue is summarized at:
http://www.dfanning.com/misc_tips/submemory.html
Any big array subscripting going on around line 1634? (don't know why
"Unable to free memory" would be the problem, but have a look anyway)
Cheers,
--
-Dick
Dick Jackson / dick@d-jackson.com
D-Jackson Software Consulting / http://www.d-jackson.com
Calgary, Alberta, Canada / +1-403-242-7398 / Fax: 241-7392
|
|
|
|
|
Re: Question on IDL Error: Unable to free memory: from array descriptor [message #34808 is a reply to message #34706] |
Thu, 17 April 2003 07:00  |
K. Bowman
Messages: 330 Registered: May 2000
|
Senior Member |
|
|
In article <3E8D9A8F.D04B00A3@noaa.gov>, Ken Knapp <Ken.Knapp@noaa.gov>
wrote:
> But, it worked fine before and I never saw this error until the 5.6
> upgrade. I suspect there is cross talk between IDL and some other
> process, but I don't know how to test that or figure out what is causing
> it.
I've been getting the same kind of messages, sometime arrays, sometimes
other things:
% Unable to free memory: freeing string memory.
Interrupted system call
IDL> PRINT, !VERSION
{ x86 linux unix linux 5.6 Oct 26 2002 32 64}
This is a non-reproducible error. Re-running the same program in a new
IDl session (usually) results in no errors.
While I was writing this message, this happened to a different IDL
process on the same Linux system.
Segmentation fault (core dumped)
This is really not good.
Ken Bowman
|
|
|