| Re: IDL process stalls [message #73782] |
Sat, 27 November 2010 20:59 |
trisb
Messages: 18
Registered: November 2010
|
Junior Member |
|
|
On Nov 27, 12:30 pm, jeanh
<jghasb...@DELETETHIS.environmentalmodelers.ANDTHIS.com> wrote:
> On 27/11/2010 11:29 AM, trisb wrote:
>
>
>
>
>
>> Hi everyone,
>
>> I have a problem with a very resource-intense IDL program. It uses
>> several Gigs of RAM (sometimes including swap memory) and runs at
>> around 100% CPU for a long time (ie. days). In most cases the program
>> runs just fine without problems, however sometimes it stalls. There
>> has been no exception or error thrown, it just 'hangs' in the middle
>> of the main loop of the program. The only thing that happens is that
>> the program seems still running but nothing happens anymore.
>> Incidentally the CPU and memory load goes down which indicates that it
>> somehow "lost contact". But, as mentioned, IDL itself doesn't stop or
>> note that anything went wrong. I wonder if it is related to other
>> memory-loaded tasks running on the same machine on a different CPU.
>
>> My questions: Did anyone have the same problem already? Any idea how
>> to trace the problem? I should mention that the code runs on IDL 7.0
>> on ubuntu, 8GB RAM, intel Quadcore.
>
>> Thanks for your suggestions.
>
>> T.
>
> last time I had this kind of problem, the hard disk was full... memory
> swap became a problem (not too sure why) and eventually nothing would
> respond anymore
> Jean
Thanks for the suggestion. There's still 100GB left on the hard drive.
I wonder if there is some problem with memory (de-)allocation when
swapping. Just a guess.
T.
|
|
|
|
| Re: IDL process stalls [message #73788 is a reply to message #73782] |
Sat, 27 November 2010 12:30  |
jeanh
Messages: 79
Registered: November 2009
|
Member |
|
|
On 27/11/2010 11:29 AM, trisb wrote:
> Hi everyone,
>
> I have a problem with a very resource-intense IDL program. It uses
> several Gigs of RAM (sometimes including swap memory) and runs at
> around 100% CPU for a long time (ie. days). In most cases the program
> runs just fine without problems, however sometimes it stalls. There
> has been no exception or error thrown, it just 'hangs' in the middle
> of the main loop of the program. The only thing that happens is that
> the program seems still running but nothing happens anymore.
> Incidentally the CPU and memory load goes down which indicates that it
> somehow "lost contact". But, as mentioned, IDL itself doesn't stop or
> note that anything went wrong. I wonder if it is related to other
> memory-loaded tasks running on the same machine on a different CPU.
>
> My questions: Did anyone have the same problem already? Any idea how
> to trace the problem? I should mention that the code runs on IDL 7.0
> on ubuntu, 8GB RAM, intel Quadcore.
>
> Thanks for your suggestions.
>
> T.
last time I had this kind of problem, the hard disk was full... memory
swap became a problem (not too sure why) and eventually nothing would
respond anymore
Jean
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
