I think you need to dig a little deeper. To prove/disprove your theory, I
typed the following command from the Command Prompt:
for /L %a in (10,1,60) do "cmd /c start /b cscript //nologo test.vbs >>
test%a.txt"
As you probably know, it will launch 50 instances of test.vbs. This script
file contains the following lines:
wscript.echo Start time
wscript.sleep 120000
wscript.echo End time
Immediately after the command line had finished, I typed this command:
tasklist | find /i /c "cscript.exe"
As expected, I saw 50 instances of cscript.exe. Furthermore, after each
script had finished its run, each and every file "test%a.txt" contained two
lines: One with the time stamp of the start time, the other with the end
time stamp.
Slight correction: I saw 51, not 50 concurrent cscript instances.
Perhaps it also has something to do with the resources each instances makes
use of...
/Al
I did some further testing and logged onto the server myself and ran
30+ instances of the exact script that gets executed myself and had no
issues. This would cause issue immediately when running through
ControlM. I also downloaded Process Explorer from Microsoft and the
interesting thing that I see is that all these seperate instances of
the script running all tie back to a parent process. In the instance
of when I run after logging on I see that the parent process is
explorer. However when I look at the scripts that are being run by
ControlM they are running under a parent process p_ctmag.exe which is
the ControlM Agent. I really wonder if this executable somehow can
handle only a maximum number of threads running under it due to
limitations regarding Memory or something along those lines. Is this
possible at all? It seems odd to me that the parent process would
have anything to do with this but in process explorer you can see that
for each child process there are additional threads attached to the
parent process. Any other way to debug this?
I couldn't see their knowledgebase since they only open it to customers; you
may need to explore that side of this though. It's definitely not a problem
intrinsic to WSH. This actually may be some kind of a queuing issue for
Control-M that can be resolved by tuning the application.
"Spencer" <spe...@tabbert.net> wrote in message
news:2cbfd4e9-51b2-425d...@k37g2000hsf.googlegroups.com...
> On Sep 2, 9:01 pm, "Al Dunbar" <AlanD...@hotmail.com.nospaam> wrote:
>> "Pegasus (MVP)" <I....@fly.com.oz> wrote in message
> I did some further testing and logged onto the server myself and ran
Well I decided to scrap the effort on the .vbs script and rewrite it
in good ol' dos batch scripting. Can't beat parsing text files in
dos..sure is fun. Anyways I am not sure what the problem is but we
continue to have it. I also noticed the following items in the event
viewer on Friday when we had some additional errors.
Event Type: Information
Event Source: Application Popup
Event Category: None
Event ID: 26
Description:
Application popup: CScript.exe - Application Error : The application
failed to initialize properly (0xc0000142). Click on OK to terminate
the application.