river-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Firmstone <j...@zeus.net.au>
Subject Re: TaskManager uses
Date Sun, 25 Jul 2010 20:33:59 GMT
Hi Patricia,

Yes I'll run the tests, upload your patches against the current tree to 
JIRA, I'll upload my results there too.  I've set up an issue there for 
TaskManager, I can't remember the reference number off the top of my head.

Regards,

Peter.

Patricia Shanahan wrote:
> Peter - This is a request for help running some tests. As you know, I 
> don't yet have a viable test environment.
>
> I have reached some hypotheses about what is going on with various 
> suspicious-looking sequence number based runAfter tests. It is 
> possible that testing of a modified version might eliminate one or 
> more of my hypotheses. Unfortunately, confirming them is a matter of 
> analysis, unless we can track down an original implementer to answer 
> some questions.
>
> For example, there are some places where I believe the sequence number 
> check is harmless because the tasks were added to the TaskManager in 
> ascending sequence number order. The TaskManager javadoc does not make 
> it entirely clear that the runAfter test is only shown tasks that were 
> added before "this", so some programmers may have added a sequence 
> number check unnecessarily. I would like to insert an assertion, or 
> logging, or some other reporting mechanism that will tell me if I'm 
> wrong.
>
> If you can help with running the tests, please let me know what form 
> you would like the changes, and the recommended mechanism for getting 
> the reports.
>
> Thanks,
>
> Patricia
>
>
> On 7/24/2010 6:50 PM, Patricia Shanahan wrote:
>> My notes so far on TaskManager uses are at
>> http://www.patriciashanahan.com/apache/TaskManager_uses.docx or
>> http://www.patriciashanahan.com/apache/TaskManager_uses.pdf (same
>> content, different formats).
>>
>> If you have any insights that should be added please e-mail them to me.
>>
>> Thanks,
>>
>> Patricia
>>
>
>


Mime
View raw message