OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

wsbpel message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: [wsbpel] Issue 21


Hi,

I would like to propose to close Issue 21 with no further action. As Satish
pointed out, the spec already includes a statement highlighting that faulted
scopes will not be compensated even if the fault handler catches and manages
the fault:

"The completion of the activity of a fault handler, even when it does not
rethrow the fault handled, is never considered successful completion of the
attached scope and compensation is never enabled for a scope that has had an
associated fault handler invoked."

Thank you,

Edwin

-----Original Message-----
From: Yaron Goland [mailto:ygoland@bea.com] 
Sent: Tuesday, February 17, 2004 5:24 PM
To: edwink@collaxa.com; Satish Thatte
Subject: RE: [wsbpel] Issue 21

Edwin, do you want to propose that we close Issue 21 with no further
action or would you like me to do it?
	Thanks,
		Yaron

> -----Original Message-----
> From: Edwin Khodabakchian [mailto:edwink@collaxa.com]
> Sent: Wednesday, February 11, 2004 9:29 AM
> To: 'Satish Thatte'; ygoland@bea.com
> Cc: 'Edwin Khodabakchian'
> Subject: RE: [wsbpel] Issue 21
> 
> 
> Satish, Yaron,
> 
> "The completion of the activity of a fault handler, even when 
> it does not
> rethrow the fault handled, is never considered successful 
> completion of the
> attached scope and compensation is never enabled for a scope 
> that has had an
> associated fault handler invoked." is clear enough. My bad.
> 
> Edwin
> 
> 
> -----Original Message-----
> From: Satish Thatte [mailto:satisht@microsoft.com] 
> Sent: Sunday, February 01, 2004 8:45 PM
> To: edwink@collaxa.com; ygoland@bea.com
> Cc: edwink
> Subject: RE: [wsbpel] Issue 21
> 
> In some sense Java/C# cannot make the distinction that we are 
> discussing
> here because they can only speak of the distinction between 
> scopes that
> rethrow faults or suppress them.  BPEL does support both 
> possibilities.  The
> distinction introduced by "can we compensate this" has no 
> parallel in Java
> or C#.  So I would not say that we have here an inconsistency 
> relative to
> those languages.  
>  
> Did you have in mind adding something more expository than 
> the following
> simple statement in the spec
>  
> The completion of the activity of a fault handler, even when 
> it does not
> rethrow the fault handled, is never considered successful 
> completion of the
> attached scope and compensation is never enabled for a scope 
> that has had an
> associated fault handler invoked.
>  
> ??
>  
> Satish
> 
> ________________________________
> 
> From: Edwin Khodabakchian [mailto:edwink@collaxa.com]
> Sent: Sun 2/1/2004 6:56 PM
> To: Satish Thatte; ygoland@bea.com
> Cc: 'edwink'
> Subject: RE: [wsbpel] Issue 21
> 
> 
> 
> Yaron, Satish,
> 
> Sorry for the delay. Here is the context:
> 
> There were 2 discussions ([1], [2]) in May/June regarding 
> faulted scopes and
> compensation handling.
>  [1] fault handling (Alex Boisvert, Satish Thatte) 5/26/2003
>  [2] installing compensation handlers for faulted scopes
>      (Ram Jeyaraman, Assaf Arkin) 5/28/2003
> 
> One of the interesting points made in those discussions is that BPEL
> faultHandlers are responsible for undoing whatever might have happened
> before the fault but should *NOT* include logic that would 
> represent an
> alternative way of completing the activity that threw and 
> exception. The
> main reason here is that the scope would remain as faulted and no
> compensation would be installed to compensate the work done 
> by the fault
> handler.
> 
> This behavior/pattern is different from how try/catch is used 
> in Java/C# and
> therefore is this is not already documented in the spec it 
> might be worth
> doing it. The goal of issue 21 was to make this explicit by using a
> different term than catch but I would agree now that renaming 
> the construct
> would be overkill.
> 
> Best,
> 
> Edwin
> 
> -----Original Message-----
> From: Satish Thatte [mailto:satisht@microsoft.com]
> Sent: Wednesday, January 21, 2004 5:47 PM
> To: ygoland@bea.com; edwink@collaxa.com
> Subject: RE: [wsbpel] Issue 21
> 
> Yes it doesn't appear.  Edwin, you might have some language 
> in mind that
> has changed but still offends or confuses.  Please send up a pointer.
> 
> -----Original Message-----
> From: Yaron Goland [mailto:ygoland@bea.com]
> Sent: Wednesday, January 21, 2004 11:28 AM
> To: edwink@collaxa.com; Satish Thatte
> Subject: RE: [wsbpel] Issue 21
> 
> I believe that phrase has been deleted from the spec. At least I
> couldn't
> find it in the latest editor's draft. Satish?
> 
>                 Yaron
> 
> > -----Original Message-----
> > From: Edwin Khodabakchian [mailto:edwink@collaxa.com]
> > Sent: Tuesday, January 20, 2004 11:39 PM
> > To: ygoland@bea.com
> > Cc: 'Satish Thatte'; 'edwink'
> > Subject: RE: [wsbpel] Issue 21
> >
> >
> > Hi Yaron,
> >
> > I hope that you are doing well. Sorry for the delay in
> > getting back to you
> > regarding issue #21 (faultHandlers to be renamed
> > cancellationHandlers).
> >
> > I have no problem dropping the issue.
> >
> > One request I have though is to clarify within the
> > specification what people
> > really mean when they say "faultHandlers/catch are not
> > expected to do any
> > forward looking work".
> >
> > Thank you,
> > Edwin
> >
> >
> 
> 
> 
> 
> 
> 
> 
> 



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]