[ndnSIM] Forwarder's onIncomingInterest

[Muhammad Hosain Abdollahi Sabet]

Well, IMHO you may generated a special log to easily grep it then. Besides,
as you probably know, generated logs specify which method of which node has
generated a log line.

Sabet

خندان باشید
با تشکر

On Tue, Jun 13, 2017 at 1:44 PM, John Baugh <jpbaugh at umich.edu> wrote:

> Muhammad,
>
> Thank you for the input.
>
> I re-read the NFD documentation and it does seem to suggest that
> onIncomingInterest is called for all incoming interests (and localhost
> stuff too)
>
> I use the l3protocol getForwarder, passing the node I have "in the
> middle", which I'm treating as the router.
>
> I'm​ using point to point helper in various configurations, but all have:
>
> (Consumers) -- (router) -- (1 producer)
>
> I've tried with up to 10 consumers, connected directly using point to
> point helper.
>
> I do understand there are multiple forwarders created (one per node).  I
> use a special object inside the onIncomingInterest to determine if "this"
> is the node I want.  This helps distinguish between the router and other
> nodes.
>
> Something like:
>
> if(popMan == nullptr)
> {
>    //Don't print out data
> }
>
> Anyway, I'm only detecting a single incoming interest entering on faces
> into the router for content, when it should be around 300 times (10 / sec.
> For 30 seconds, simulated)
>
>
>
> On Jun 13, 2017 5:05 AM, "Muhammad Hosain Abdollahi Sabet" <
> mhasabet at gmail.com> wrote:
>
>> Not only all Interests *arrive to* a node, but also local Interests will
>> go to startProcessInterest. You see all of management protocol in NFD
>> relies on sending and processing local Interests.
>>
>> 2- Using NFD logs(e.g. Forwarder logs) can help you find out the *when*
>> startProccessInterest is called(Note NFD_LOG_DEBUG in the method). At least
>> that's the way I do it and it works for what I need.
>>
>> Sabet
>>
>> خندان باشید
>> با تشکر
>>
>> On Tue, Jun 13, 2017 at 12:59 PM, John Baugh <jpbaugh at umich.edu> wrote:
>>
>>> Greetings all,
>>>
>>> I resolved many other issues, and now I'm come to a bit of a concern
>>> regarding the forwarder class.
>>>
>>> I have been recompiling the Forwarder class, with some modifications to
>>> onIncomingInterest (actually, it just calls another method I wrote, and
>>> that method calls the method of a shared object called
>>> PopMan->recordInterestStats(inFace, interest))
>>>
>>> So here's what I originally understood:
>>>
>>>
>>>    - If an interest arrives at a *node*, the Forwarder's
>>>    onIncomingInterest (I think indirectly through startProcessInterest, at
>>>    least) is called automatically *every *time any interest arrives
>>>
>>> The behavior I'm observing is:
>>>
>>>    - If an interest is about to be *forwarded*, then Forwarder's
>>>    onIncomingInterest is called
>>>
>>> Now, I realize the name "Forwarder" might have been a clue...  but I was
>>> under the impression that this class sensed *all incoming interests to
>>> the node*, not just ones that are to be forwarded.
>>>
>>> 1.  Please confirm the behavior of this method
>>>
>>> 2.  If I'm right, *is there any way *to detect when an interest arrives
>>> at a node (not when the interest is to be forwarded, or cached, etc. - but
>>> EVERY TIME an interest arrives at a particular node.
>>>
>>> I had previously spent several weeks before working with the Strategy
>>> class, thinking afterReceiveInterest was called every time an interest was
>>> received at a node...  it turned out I was wrong.
>>>
>>> I was told (I think?) that I needed to override Forwarder's methods
>>> instead.  So I did.
>>>
>>> I've learned a lot, don't get me wrong.  I just am not getting the
>>> behavior I expect, and I'm starting to become extremely concerned that I've
>>> gone down the wrong path... again.
>>>
>>> Thoughts?
>>>
>>> Thanks,
>>>
>>> John
>>>
>>> _______________________________________________
>>> ndnSIM mailing list
>>> ndnSIM at lists.cs.ucla.edu
>>> http://www.lists.cs.ucla.edu/mailman/listinfo/ndnsim
>>>
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/ndnsim/attachments/20170613/db7f8ffb/attachment.html>