[rules-users] Fusion, Insert Events with timestamp in the past.

[Matteo Mortari]

Errata Corrige: "Because basically with this approach I personally found
that also *you have to*:"

I missed the *you have to* bit, sorry.


On Wed, Jun 4, 2014 at 3:55 PM, Matteo Mortari <matteo.mortari at gmail.com>
wrote:

> Concerning point #1, I'm not sure I understand what you mean, I'm sorry,
> but might be connected to my last statement about "this approach would
> solve for "network delay issue" but will induce another problem (...)"
>
> Because basically with this approach I personally found that also:
> a. data is buffered and inserted in chronological order
> b. pseudoclock is advanced anyway after a "deathwatch" period, normally to
> activate negative patterns as you mention - thus at the same time
> discarding data if arriving after the deathwatch already triggered
> c. pseudoclock is advanced anyway to trigger timers of rules
>
> This is the part which gets complex and the reason behind I mentioned I
> do mostly same as code example linked.
>
> In my use case scenario, implementation of deathwatch + advance of
> pseudoclock anyway, do fit the bill; but of course result may vary
> depending case by case.
>
>
> Concerning point #2 in my use case timer of rules do work, even if I use
> pseudoclock.  But again, result may differ depending on the rules
> specification, I suppose.
>
>
> Sorry I cannot be of more help, I thought worthy share my experience
> anyway because I felt was very similar scenario, and fortunately the system
> behaves as expected.
>
> Ciao
> MM
>
>
> On Wed, Jun 4, 2014 at 2:07 PM, SebastianStehle <mail2stehle at gmail.com>
> wrote:
>
>> Thanks for your answer.
>>
>> We already use the pseudo-clock for running unit tests, but there might be
>> some other problems:
>>
>> 1. If you do not move forward step by step you can miss the exact times
>> when
>> using negative patterns. e.g. in my example you would need to step forward
>> with AdvanceTiem with about 100ms or 1sec or so.
>>
>> 2. We also use timers, this does not work with the pseudo clock as well,
>> same problem like above.
>>
>> In my understand the example should work, if not I dont see what the
>> @timestamp config is for.
>>
>>
>>
>> --
>> View this message in context:
>> http://drools.46999.n3.nabble.com/Fusion-Insert-Events-with-timestamp-in-the-past-tp4029843p4029852.html
>> Sent from the Drools: User forum mailing list archive at Nabble.com.
>> _______________________________________________
>> rules-users mailing list
>> rules-users at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/rules-users/attachments/20140604/062aa4e9/attachment.html