| Re: Problems with long IDL background jobs [message #36986] |
Mon, 10 November 2003 01:37 |
Nigel Wade
Messages: 286
Registered: March 1998
|
Senior Member |
|
|
R.G. Stockwell wrote:
> "Nigel Wade" <nmw@ion.le.ac.uk> wrote in message
> news:bofq1e$t73$1@south.jnrs.ja.net...
> ...
>
>> Does your IDL code do any plotting? I didn't think IDL actually opened any
>> connection to the X server unless you issued a command which initiates
>
> graphics.
>
>
>
> As a guess, perhaps you have some commands in your startup
> that initiate the X server connection.
> I don't know if the following commands do this:
> device,true=24 ; i do not think i need this
> device,retain = 2
> device,decomposed=0
>
> Cheers,
> bob
>
>
>
I think that any DEVICE command has the potential to open a graphics channel
as it may well need it to perform the requested operation. For example, a
device,true=24 will almost certainly do so as it's requesting a 24bit
TrueColor visual from the X server and will need to communicate with the X
server to do so. I think retain=2 will also, as it's asking for use of the X
servers backing store.
--
Nigel Wade, System Administrator, Space Plasma Physics Group,
University of Leicester, Leicester, LE1 7RH, UK
E-mail : nmw@ion.le.ac.uk
Phone : +44 (0)116 2523548, Fax : +44 (0)116 2523555
|
|
|
|
| Re: Problems with long IDL background jobs [message #36989 is a reply to message #36986] |
Sun, 09 November 2003 01:56  |
R.Bauer
Messages: 1424
Registered: November 1998
|
Senior Member |
|
|
Craig Markwardt wrote:
>
> Reimar Bauer <R.Bauer@fz-juelich.de> writes:
>> Hi
>>
>> I know this problem too. My solution is to start the session without the
>> X-Server. If I did only a ssh -C then it works while ssh -XC doesn't.
>>
>> If you have only a terminal connection then nothing which requires X is
>> started. I think the problem belongs more to the OS or user environment
>> as to idl.
>
> And, why not SET_PLOT, 'NULL' or SET_PLOT, 'Z' to *really* prevent IDL
> from doing anything with X?
>
> Craig
>
Most time you are right but what is if you don't like to change your normal
startup file.
The problem is e.g. device,decompose=0 or a window command and so on.
Reimar
--
Forschungszentrum Juelich
email: R.Bauer@fz-juelich.de
http://www.fz-juelich.de/icg/icg-i/
============================================================ ======
a IDL library at ForschungsZentrum Juelich
http://www.fz-juelich.de/icg/icg-i/idl_icglib/idl_lib_intro. html
|
|
|
|
| Re: Problems with long IDL background jobs [message #37004 is a reply to message #36989] |
Fri, 07 November 2003 12:01 |
R.G. Stockwell
Messages: 363
Registered: July 1999
|
Senior Member |
|
|
"Nigel Wade" <nmw@ion.le.ac.uk> wrote in message
news:bofq1e$t73$1@south.jnrs.ja.net...
...
>
> Does your IDL code do any plotting? I didn't think IDL actually opened any
> connection to the X server unless you issued a command which initiates
graphics.
>
As a guess, perhaps you have some commands in your startup
that initiate the X server connection.
I don't know if the following commands do this:
device,true=24 ; i do not think i need this
device,retain = 2
device,decomposed=0
Cheers,
bob
|
|
|
|
| Re: Problems with long IDL background jobs [message #37008 is a reply to message #37004] |
Fri, 07 November 2003 01:52 |
Nigel Wade
Messages: 286
Registered: March 1998
|
Senior Member |
|
|
J.K. wrote:
> I have some processing that takes about 12hrs so
> I'd like to run this in the background. I'm on
> IRIX. It must be something simple that I'm doing
> wrong. After I log out and close my X11 connections,
> I'll get a message in my log file:
>
> XIO: fatal IO error 131 (Connection reset by peer) on X server "xxxxxxxx:0.0"
> after 116 requests (116 known processed) with 0 events remaining.
>
> Am I not redirecting standard out properly?
> Below is what I'm doing. Any help appreciated.
No, that's the X client (IDL) complaing it's lost its connection to the X
server.
>
> idlstartbg.pro startup file, I only set the path, no device commands.
> !PATH=!PATH+":~/idl:~/wave/wvproc:~/wave/fermi98
>
> runidl batch file:
> #!/bin/csh
> setenv IDL_STARTUP idlstartbg.pro
> idl $1 >& log.$1 &
>
> runback.pro
> .run sdsnan
> exit
>
> sdsnan.pro is a main
> filelist = strarr(600)
> openr, unit, "list",/get_lun
> i = -1
> fn = ' '
> while (not eof(unit)) do begin
> i = i + 1
> readf, unit, fn
> filelist(i) = fn
> print, i,filelist(i)
> endwhile
> close, unit
> ; do lots of other stuff with some print statments
> end
>
> How I execute:
> nohup runidl runback >& out &
>
> exit (close X window)
>
>
> Produces:
> out: empty
> log.runback has standard out but also the X11 message and job does not
> complete.
>
> What am I doing wrong?
> If I change all print to printf in sdsnan.pro to write to a file it
> works but why can't I capture the standard out?
>
> Thanks,
> John K.
Does your IDL code do any plotting? I didn't think IDL actually opened any
connection to the X server unless you issued a command which initiates graphics.
Anyway, try issuing the command 'unset DISPLAY' before running IDL (assuming
you are using sh/ksh). And, for good measure, in IDL use the command
SET_PLOT,'PS' to change the default display device to PostScript.
--
Nigel Wade, System Administrator, Space Plasma Physics Group,
University of Leicester, Leicester, LE1 7RH, UK
E-mail : nmw@ion.le.ac.uk
Phone : +44 (0)116 2523548, Fax : +44 (0)116 2523555
|
|
|
|
| Re: Problems with long IDL background jobs [message #37009 is a reply to message #37008] |
Thu, 06 November 2003 21:32 |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
Reimar Bauer <R.Bauer@fz-juelich.de> writes:
> Hi
>
> I know this problem too. My solution is to start the session without the
> X-Server. If I did only a ssh -C then it works while ssh -XC doesn't.
>
> If you have only a terminal connection then nothing which requires X is
> started. I think the problem belongs more to the OS or user environment
> as to idl.
And, why not SET_PLOT, 'NULL' or SET_PLOT, 'Z' to *really* prevent IDL
from doing anything with X?
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@REMOVEcow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
| Re: Problems with long IDL background jobs [message #37014 is a reply to message #37009] |
Thu, 06 November 2003 09:11 |
R.Bauer
Messages: 1424
Registered: November 1998
|
Senior Member |
|
|
Hi
I know this problem too. My solution is to start the session without the
X-Server. If I did only a ssh -C then it works while ssh -XC doesn't.
If you have only a terminal connection then nothing which requires X is
started. I think the problem belongs more to the OS or user environment
as to idl.
regards
Reimar
J.K. wrote:
> I have some processing that takes about 12hrs so
> I'd like to run this in the background. I'm on
> IRIX. It must be something simple that I'm doing
> wrong. After I log out and close my X11 connections,
> I'll get a message in my log file:
>
> XIO: fatal IO error 131 (Connection reset by peer) on X server "xxxxxxxx:0.0"
> after 116 requests (116 known processed) with 0 events remaining.
>
> Am I not redirecting standard out properly?
> Below is what I'm doing. Any help appreciated.
>
> idlstartbg.pro startup file, I only set the path, no device commands.
> !PATH=!PATH+":~/idl:~/wave/wvproc:~/wave/fermi98
>
> runidl batch file:
> #!/bin/csh
> setenv IDL_STARTUP idlstartbg.pro
> idl $1 >& log.$1 &
>
> runback.pro
> .run sdsnan
> exit
>
> sdsnan.pro is a main
> filelist = strarr(600)
> openr, unit, "list",/get_lun
> i = -1
> fn = ' '
> while (not eof(unit)) do begin
> i = i + 1
> readf, unit, fn
> filelist(i) = fn
> print, i,filelist(i)
> endwhile
> close, unit
> ; do lots of other stuff with some print statments
> end
>
> How I execute:
> nohup runidl runback >& out &
>
> exit (close X window)
>
>
> Produces:
> out: empty
> log.runback has standard out but also the X11 message and job does not
> complete.
>
> What am I doing wrong?
> If I change all print to printf in sdsnan.pro to write to a file it
> works but why can't I capture the standard out?
>
> Thanks,
> John K.
--
Reimar Bauer
Institut fuer Stratosphaerische Chemie (ICG-I)
Forschungszentrum Juelich
email: R.Bauer@fz-juelich.de
------------------------------------------------------------ -------
a IDL library at ForschungsZentrum Juelich
http://www.fz-juelich.de/icg/icg-i/idl_icglib/idl_lib_intro. html
============================================================ =======
|
|
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
