Workflow on 4.6C (customising the organisational management buffering timeframe)

[Maue, Thomas X.]

Kjetil,
 
You have answered your own question.  What a tough road you had to follow!!
 
 
There is a configuration setting in table T77S0 that allows for how the
organizational environment is
 buffered.  The table entry is:
WFLOW INBOX X   Org. Environment Buffering Type in the Inbox.=20
 
If the indicator is 'X', buffered once per day or each logon. =20
If not set (space), read each time the inbox is called.  Note - this setting
leads to longer call times for=20
the inbox.  A lot of new systems come standard delivered with the 'S' and
causes non-buffered results.
Maybe Alan Rickayzen can shed some light here.=20
 
Use SM30 to set it and ignore the warning message.  Then promote to
production before changing=20
development back to space.                              =09
 
Thomas Maue=20
Workflow Consultant
 
> -----Original Message-----
>> From: Kjetil Kilhavn <KJETILK at statoil.com>@LYONDELL=20
> Sent: Monday, April 23, 2001 2:41 AM
> To:   SAP-WUG at MITVMA.MIT.EDU
> Subject:      Re: Workflow on 4.6C (customising the organisational
> management              buffering timeframe)
>=20
> =20
> Does anyone know how I can change the "default" between non-buffered and
> buffered? Alan, you mention there is a customising flag, but I haven't
> found it.
>=20
> I have encountered problems using the SAP_WAPI_COUNT_WORKITEMS function
> (module). Possibly they may be related to the buffering mentioned here.
> What I
> want to do is to check every user and send a message to those users who
> have
> workitems. This is part of our new integration solution between SAP and
> Lotus
> Notes.
>=20
> When I loop through an internal table of users and call the function for
> each of
> them the system dumps. The problem is related to shared memory, with the
> first
> message in the dump being "The EXPORT data cluster is too large for the
> application buffer." The dump happens whether I have commit work at the
> end of
> every loop or not. With only a fraction of the users in the development
> system
> compared to the production systems raising memory limits is not an option.
>=20
> With 'a little help from my friends' I have found out some more. Further
> down
> the call stack a decision is made whether to export results to a database
> or to
> shared memory. The decision in our development system is to export to
> shared
> memory. By digging further I have become quite certain that the entry in
> T77S0
> with key WFLOW, INBOX decides what happens. In the development system this
> has
> the value S, in the production system it has the value ' ' (space). I have
> looked for a place to customize this, but I haven't found it. Tips &
> tricks
> anyone?
> --
> KjetilK
>=20
>=20
>=20
>=20
>=20
>> From:  "Rickayzen, Alan" <alan.rickayzen at sap.com>@MITVMA.MIT.EDU> on
> 22.04.2001
>        07:44
>=20
> Please respond to "SAP Workflow Users' Group" <SAP-WUG at MITVMA.MIT.EDU>
>=20
> Sent by:  SAP Workflow <Owner-SAP-WUG at MITVMA.MIT.EDU>
>=20
>=20
> To:    SAP-WUG at MITVMA.MIT.EDU
> cc:     (bcc: Kjetil Kilhavn)
>=20
> Subject:  Re: Workflow on 4.6C
>=20
>=20
> Paul,
> Certain org management buffers are refreshed at midnight and others during
> logon. This often gives rise to the feeling that the system is unstable or
> has to be left overnight to cool down ;-).
>=20
> If you use the diagnosis tool, it will refresh the buffers for you while
> you
> test and you won't see the behaviour that you describe.
>=20
> There is also a customizing flag for turning off the buffers (useful in
> the
> development or demo environment), but make sure they're turned on in the
> production system so that performance is not reduced.
>=20
> Alan Rickayzen
>=20
> -----Original Message-----
>> From: Paul, Reena [mailto:Reena.Paul at delta.com]
> Sent: Freitag, 16. M=E4rz 2001 15:46
> To: SAP-WUG at MITVMA.MIT.EDU
> Subject: Workflow on 4.6C
>=20
>=20
> Hi All,
>=20
>     Workflow seems to be unstable on my system. I developed a workflow and
> used the workflow test environment. The work item comes to my inbox, but
> when I execute it, gives me the error 'No administrator found for the
> task'.
> The next day, I was able to execute the work item with no errors and I
> made
> no changes to the workflow.
>=20
> I had developed another workflow wherein the last step was to send mail.
> All
> other work items worked fine but the send mail had an error. I deleted the
> step and re-created the step, but still didn't work. I did it 5 times and
> I
> got the same error (in the process it created 5 unwanted tasks).  I
> created
> the step the next day and it worked fine.
>=20
> I am currently testing on my sandbox. But we have 15 workflow requirements
> and I need to start the development on my Dev box. I am not sure I would
> like to wait the next day for the workflow to work. Nor would I like to
> create unnecessary task because then I will have to start cleaning up my
> transports.
>=20
> I have never had this problem in 4.5B.
>=20
> Am I missing some thing or do I have to do some thing else.
> I have done synchronize buffer, Workflow verification and nothing seems to
> work.
>=20
> If anyone could give me any input.
>=20
> Thanks,
> Reena Paul
> Workflow consultant
>=20
>=20
>=20
> -------------------------------------------------------------------
> Statoil: The information contained in this message may be CONFIDENTIAL and
> is
> intended for the addressee only. Any unauthorised use, dissemination of
> the
> information or copying of this message is prohibited. If you are not the
> addressee, please notify the sender immediately by return e-mail and
> delete this
> message.
> Thank you.
 
 
**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.
 
 
www.mimesweeper.com
**********************************************************************