gpfREXX/REXX script runs for 2-3 days then dies 
Author Message
 gpfREXX/REXX script runs for 2-3 days then dies

(I'm running REXX ubder OS/2 Warp Connect CSD IPO8000 (think that's fixpack 22
 and gpfREXX version 1 Release 0 Level B1)

I have a gpfREXX application that runs just fine for several days on end and
then just stops. It won't work again until the system is rebooted. I have no
real clues. Normal REXX comes down with it. I can't even run rexxtry getting
either error rex0004 (***Program interrupted***) (on a "call" instruction)
or rex0048 (***Failure in system service***) (on a "pull" instruction).
The required libs all seem to be fine (rexx.dll, rexxapi.dll, rexxinit.dll
and rexxutil.dll) and os2\dll is in my libpath. Very simple REXX scripts
still seem to work, however.

I have seen this once before when I don't think I had been running gpfREXX
however, I can't be certain. For this reason, I don't know if REXX or gpfREXX
is the cause. I am guessing that gpfREXX dies when calling one of our
REXX-C DLLs. The gpfREXX screens, btw, don't hang, they just stop running
(clock won't update, won't process user input, ...).

Unfortunately, due to the long delays between problems, I haven't been able
to narrow it down. My guess is that gpfREXX is at fault but this is only a
guess.

I hope someone out there has seen something like this before. Thnak you
to anyone that can help out.

Todd Fox



Fri, 06 Aug 1999 03:00:00 GMT  
 gpfREXX/REXX script runs for 2-3 days then dies


Quote:

>I hope someone out there has seen something like this before. Thnak you
>to anyone that can help out.

I am almost certain that this is caused by a memory leak. If you call a Rexx
procedure recursively (but not very often) it can run for quite a while before
the program dies.

If it is a memory leak inside GpfRexx iself or one of the c *.dll's then it will be
difficult to find. Your best bet might be to run one *.dll in separate little test
programs and see if anything breaks.

In the worst case, you might want to have a "cron" clone reboot the machine
once a day - I know a mainframe setup where they have 3 mainframes and
re-boot them in rotation every 3 days; memory leaks are very hard to resolve!

Hope this is *some* help,
frithiof jensen.

===============================================================================
    The above article is the personal view of the poster and should not be
       considered as an official comment from the JET Joint Undertaking
===============================================================================



Fri, 06 Aug 1999 03:00:00 GMT  
 gpfREXX/REXX script runs for 2-3 days then dies

Quote:

> I hope someone out there has seen something like this before. Thnak you
> to anyone that can help out.

There is a problem with both REXX & SOM introduced by FP 17 - 26 that
make strange things happen. The fixes finally made it into FP 28 and I
was instrumental in getting them into Warp 4.

I have put the new modules into the file rexxsom.zip
(ftp://ftp.cfsrexx.com/pub/rexxsom.zip)  which can be unzipped and put
into your \OS2\DLL library (replacing the Fixpak modules). This will
have to be done from the Install diskettes, utility disks, or another
OS/2 partition as these DLLs are locked when the system is running.

 --------------------------------------------------------------
 Dick Goran            author, REXX Reference Summmary Handbook
 C F S Nevada, Inc.     Contrib. Ed., Personal Systems Magazine
 953 E. Sahara Ave, Suite 9B                     [OS/2 Advisor]
 Las Vegas, NV 89104-3012          CompuServe:       71154,2002

 702-732-3847 FAX
                    http://www.cfsrexx.com
              (see our new OS/2 game - CFSPoker)



Fri, 06 Aug 1999 03:00:00 GMT  
 gpfREXX/REXX script runs for 2-3 days then dies


:>

[snip]
:>(I'm running REXX ubder OS/2 Warp Connect CSD IPO8000 (think that's fixpack 22
:>I have seen this once before when I don't think I had been running gpfREXX
:>however, I can't be certain. For this reason, I don't know if REXX or gpfREXX
:>is the cause. I am guessing that gpfREXX dies when calling one of our
:>REXX-C DLLs. The gpfREXX screens, btw, don't hang, they just stop running
:>(clock won't update, won't process user input, ...).
:>
[snip]

Todd,

Have you tried to get some help from Jean-Bernard Cl/erin (/e mimics an e
acute) the author of GpfRexx? He is a very responsive guy and always do his
best to answer questions.
He is also the author of Home Page Publisher and I think that with the doc of
it you should find a way to cantact him. If not, just write me and I'll dig
out the compuserve Id he uses.

Jean-Pierre Cabani/e  (also an e acute)



Sun, 08 Aug 1999 03:00:00 GMT  
 
 [ 4 post ] 

 Relevant Pages 

1. gpfrexx vs. vispro/rexx

2. gpfrexx vs. vispro/rexx

3. Canceling a long running Object REXX script

4. Run UNIX from rexx script

5. How to run REXX scripts in IE5

6. using Rexx to run a Unix script from NT

7. Question on running a DOS program from OS/2 REXX script

8. OT: The day the internet died..

9. DIES on the Apollo/Runs on others

10. php dies on script execution

11. expect scripts dies after starting ping for ~ 250 times in a loop

12. How to die and keep executed children running

 

 
Powered by phpBB® Forum Software