I don't particularly like the name, but as its not really intended for
regular use it doesnt necessarily bother me enough to change it. I might
have gone for 'legacy session exception behaviour' or something to that
effect.

I am more interested that its use doesn't entirely accomplish what I
suggested adding it for though, since only some of the new behaviour is
reversed by its use. I think the new behaviour seems reasonable, but I
still believe an interim ability to fully restore the old (broken,
admittedly) behaviour if required be would be prudent given how long it has
been that way (even supposing it meant a second system property to allow
choosing whether you get some or all of the new behaviour).

Robbie

On 18 July 2012 00:13, Justin <[email protected]> wrote:

> Hi, Rajith.  I'm prepared to accept this, but I'm wondering whether
> "is_hard_error" is the right name for the system property.  Anyone have
> thoughts?
>
>
> On Mon, 16 Jul 2012, Rajith Attapattu wrote:
>
>  Hi Justin,
>>
>> Could please port the following two commits to the 0.18 release branch ?
>> They are associated with QPID-3575.
>> The latter (r1362161) has a system prop to revert to the old behaviour
>> if need be (as per Robbie's suggestion).
>>
>> http://svn.apache.org/viewvc?**rev=1362162&view=rev<http://svn.apache.org/viewvc?rev=1362162&view=rev>
>> http://svn.apache.org/viewvc?**rev=1362161&view=rev<http://svn.apache.org/viewvc?rev=1362161&view=rev>
>>
>> Regards,
>>
>> Rajith
>>
>> ------------------------------**------------------------------**---------
>> To unsubscribe, e-mail: 
>> [email protected].**org<[email protected]>
>> For additional commands, e-mail: [email protected]
>>
>>
>>
> ------------------------------**------------------------------**---------
> To unsubscribe, e-mail: 
> [email protected].**org<[email protected]>
> For additional commands, e-mail: [email protected]
>
>

Reply via email to