We take pull requests.

On Nov 8, 2012, at 4:10 AM, nightwatch77 <[email protected]> wrote:

> That's quite bad that RSB cannot do transactional processing correctly. Fine 
> if you can live with that but imho it should be safe by default.
> And imho 10 minutes is way too long for a single transaction, especially a 
> distributed transaction. If your processing takes so long you should probably 
> use some long-running task for storing intermediate status, for example a 
> saga.
> And I confirm, there is a difference between SQL 2005 and 2008 in how they 
> handle distributed transactions - in SQL 2005 you will be promoted to a 
> distributed transaction if you open a second connection to same database 
> inside a transaction scope. SQL 2008 will not use DTC in such case.
> I've been fighting with transactional processing of messages for quite a long 
> time and have even built an SQL-only message bus just to be able to guarantee 
> ACID without resorting to DTC. It's nice to have it working but there are 
> many pitfalls and if you can, it's better to avoid using DTC at all.
> 
> Best regards
> RG
> 
> On Friday, November 2, 2012 4:31:05 PM UTC+1, Daventry wrote:
>> 
>> Hi Corey,
>>  
>> I think I replied just to you instead of posting to the group.
>>  
>> We're using Rhino Queues as transport. RSB version is 2.3.4.0
>>  
>> We're also supressing the TransactionScope because otherwise the DTC kicks 
>> in every time we open a second connection to a DB (which is SQL Server 2008 
>> R2). I thought this was a SQL Server 2005 bug and should not happen in the 
>> 2008 version...
>>  
>> Thanks.
>> 
>> On Friday, November 2, 2012 3:08:16 PM UTC, Corey wrote:
>>> The TransactionScope timeout would be much sooner than 10 minutes. I'm not 
>>> sure why you would be seeing this behavior. Which transport is being used?
>>> 
>>> On Wed, Oct 31, 2012 at 11:22 AM, Daventry <[email protected]> wrote:
>>>> Hi,
>>>>  
>>>> We've noticed that when our message is being processed, after 10 minutes 
>>>> the message is processed even though the first run hasn't errored. So they 
>>>> both run in parallel (because we have set the bus threadcount to 5).
>>>> If the first run finishes in less than 10 minutes, then we don't see such 
>>>> behavior.
>>>>  
>>>> We don't have this problem when threadcount is set to 1.
>>>>  
>>>> Regards.
>>>> -- 
>>>> You received this message because you are subscribed to the Google Groups 
>>>> "Rhino Tools Dev" group.
>>>> To view this discussion on the web visit 
>>>> https://groups.google.com/d/msg/rhino-tools-dev/-/VvCvUP1WXI8J.
>>>> To post to this group, send email to [email protected].
>>>> To unsubscribe from this group, send email to 
>>>> [email protected].
>>>> For more options, visit this group at 
>>>> http://groups.google.com/group/rhino-tools-dev?hl=en.
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "Rhino Tools Dev" group.
> To view this discussion on the web visit 
> https://groups.google.com/d/msg/rhino-tools-dev/-/WZEpY0Z1ogoJ.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to 
> [email protected].
> For more options, visit this group at 
> http://groups.google.com/group/rhino-tools-dev?hl=en.

-- 
You received this message because you are subscribed to the Google Groups 
"Rhino Tools Dev" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/rhino-tools-dev?hl=en.

Reply via email to