comp.lang.idl-pvwave archive: archive » SPAWN EXIT_STATUS not the same as shell. Ideas?

Home » Public Forums » archive » SPAWN EXIT_STATUS not the same as shell. Ideas?

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

SPAWN EXIT_STATUS not the same as shell. Ideas? [message #92816]

Fri, 04 March 2016 11:43

Paul Van Delst[1]
Messages: 1157
Registered: April 2002

Senior Member

Hello,

When I run a command in the shell I get the following result with a
non-zero exit status:

-----%<-----
$ lblrtm

Program received signal SIGSEGV: Segmentation fault - invalid memory
reference.

Backtrace for this error:
#0 0x7F1874C5AB07
#1 0x7F1874C5B11E
#2 0x3000C3269F
#3 0x45C2E5 in eminit_
#4 0x473F4A in xmerge_
#5 0x408F96 in xlayer_
#6 0x429333 in MAIN__ at lblrtm.f90:0
Segmentation fault (core dumped)

$ echo $?
139
-----%<-----

When I run the same command in my IDL script via spawn, I see the following:

-----%<-----
IDL> SPAWN, 'lblrtm', stdout, stderr, EXIT_STATUS = exit_stat
IDL> foreach s, stderr do print, s

Program received signal SIGSEGV: Segmentation fault - invalid memory
reference.

Backtrace for this error:
#0 0x7F75D828DB07
#1 0x7F75D828E11E
#2 0x3000C3269F
#3 0x45C2E5 in eminit_
#4 0x473F4A in xmerge_
#5 0x408F96 in xlayer_
#6 0x429333 in MAIN__ at lblrtm.f90:0

IDL> print, exit_stat
0
-----%<-----

Umm...anyone know why my IDL EXIT_STATUS (0) is different from my shell
exit status (139)?

Do I need an extra switch in the SPAWN command or...?

Thanks for any info.

cheers,

paulv

p.s.
IDL> print, !version
{ x86_64 linux unix linux 8.3 Nov 15 2013 64 64}

Report message to a moderator

Re: SPAWN EXIT_STATUS not the same as shell. Ideas? [message #92817 is a reply to message #92816]

Fri, 04 March 2016 12:14

Paul Van Delst[1]
Messages: 1157
Registered: April 2002

Senior Member

Just followup, this is my original code:

; Run LBLRTM
; ...Spawn the executable
SPAWN, 'lblrtm', stdout, stderr, EXIT_STATUS = exit_stat
; ...Check the result via the exit status
run_error = ( exit_stat NE 0 )
IF ( run_error ) THEN BEGIN
FOREACH s, stderr DO MESSAGE, s, /CONTINUE
MESSAGE, 'Error running LBLRTM for '+FILE_BASENAME(GT5_file) + $
'. Trying again...', /INFORMATIONAL
ENDIF

I added the following directly after the SPAWN:

; ...Check the result via the stderr output
potential_run_error = (N_ELEMENTS(stderr) GT 1)
IF ( potential_run_error ) THEN BEGIN
lc_stderr = STRLOWCASE(stderr)
; ...Look for common indications of an actual error
error_string = ['sigseg', 'segmentation fault']
FOR i = 0, N_ELEMENTS(error_string)-1 DO BEGIN
error_idx = WHERE(STRPOS(lc_stderr, error_string[i]) NE -1, $
error_count)
IF ( error_count GT 0 ) THEN BEGIN
exit_stat = -1
BREAK
ENDIF
ENDFOR
ENDIF

Anyone have a smarter method?

This gets around the problem I'm having but it would sure be nice if
EXIT_STATUS worked properly.

cheers,

paulv

On 03/04/16 14:43, Paul van Delst wrote:
> Hello,
>
> When I run a command in the shell I get the following result with a
> non-zero exit status:
>
> -----%<-----
> $ lblrtm
>
> Program received signal SIGSEGV: Segmentation fault - invalid memory
> reference.
>
> Backtrace for this error:
> #0 0x7F1874C5AB07
> #1 0x7F1874C5B11E
> #2 0x3000C3269F
> #3 0x45C2E5 in eminit_
> #4 0x473F4A in xmerge_
> #5 0x408F96 in xlayer_
> #6 0x429333 in MAIN__ at lblrtm.f90:0
> Segmentation fault (core dumped)
>
> $ echo $?
> 139
> -----%<-----
>
> When I run the same command in my IDL script via spawn, I see the
> following:
>
> -----%<-----
> IDL> SPAWN, 'lblrtm', stdout, stderr, EXIT_STATUS = exit_stat
> IDL> foreach s, stderr do print, s
>
> Program received signal SIGSEGV: Segmentation fault - invalid memory
> reference.
>
> Backtrace for this error:
> #0 0x7F75D828DB07
> #1 0x7F75D828E11E
> #2 0x3000C3269F
> #3 0x45C2E5 in eminit_
> #4 0x473F4A in xmerge_
> #5 0x408F96 in xlayer_
> #6 0x429333 in MAIN__ at lblrtm.f90:0
>
> IDL> print, exit_stat
> 0
> -----%<-----
>
> Umm...anyone know why my IDL EXIT_STATUS (0) is different from my shell
> exit status (139)?
>
> Do I need an extra switch in the SPAWN command or...?
>
> Thanks for any info.
>
> cheers,
>
> paulv
>
>
> p.s.
> IDL> print, !version
> { x86_64 linux unix linux 8.3 Nov 15 2013 64 64}

Report message to a moderator

Previous Topic:	Set graphics variables once, function graphics
Next Topic:	Problems on the Savitzky-Golay smoothing filter

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 15:17:53 PDT 2025

Total time taken to generate the page: 0.00486 seconds