[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
RE: [smila-dev] ODE Processing timeout vs. message
|
I created a bug for this @ https://bugs.eclipse.org/bugs/show_bug.cgi?id=302402
Thomas Menzel @ IT-Solutions GmbH
-----Original Message-----
From: smila-dev-bounces@xxxxxxxxxxx [mailto:smila-dev-bounces@xxxxxxxxxxx] On Behalf Of Juergen.Schumacher@xxxxxxxxxxx
Sent: Mittwoch, 10. Februar 2010 12:17
To: smila-dev@xxxxxxxxxxx
Subject: RE: [smila-dev] ODE Processing timeout vs. message
Hi,
Thomas Menzel wrote:
> I noticed recently that the timeout for the processing (pipeline.timeout=10) gets a bit in the way of
> configuring the MaxMessageBlockSize (MMBS).
>
> The total processing time is usually/often a linear function of the MMBS causing the timeout to be hit
> when increased unless the timeout is set to a very high value.
>
> As a consequence it hard to config the MMBS independently of the timeout.
>
> I suggest that the timeout is defined on a per record/item and then the effective timeout for the process
> is is calc'ed by timeout * records in message.
OK with me.
> Spinning this even further I think that such timeouts should be spec'ed on pipeline level and not globally,
> since pipelines might have very different timeout needs - or even better on pipelet/processing level. However,
> there might be already such BPEL? mechanisms. Anybody know of them?
Yes, the is timeout for each BPEL invocation (see invokeProcess() in org.eclipse.smila.ode.ODEServer),
so it can be set dependent on message size and process name, so it should be easy to configure
pipeline-specific timeouts and use them. Just someone would have to implement it (:
Cheers,
Juergen
_______________________________________________
smila-dev mailing list
smila-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/smila-dev
