Re: [reSIProcate] DnsResult blacklistLast,greylistLast question.

[Byron Campen <docfaraday@xxxxxxxxx>]

On 12/30/13 10:24 AM, Sergey Pavlov wrote:

On 28 December 2013 01:11, Byron Campen <docfaraday@xxxxxxxxx> wrote:

On 12/27/13 3:48 AM, Sergey Pavlov wrote:

Am I understand correctly that in such case UDP transport will not be grey-listed in case of failure a transaction failure? I.e. failed transaction will not influence on other signaling with corresponding IP.

   It would be greylisted if there is a timeout, but because there are no alternatives, it would continue to be used.

In such case it is ok. BTW I am not sure about exact scenario but as I know currently the stack generates '480 Transport failure: no transports left to try'  in case when there are no alternatives.

    That happens when stuff is blacklisted only, I think.

 

 

In our case we will be  satisfied even with second option.  Indeed, greylisting can be really useful for results obtained from DNS lookup. But I think that ability to disable greylisting completely (or according to some predefined list) still should provided. In fact we have two conditions which may cause unanswered requests on UAS side:

1) bug in SIP stack implementation  which is triggered by some specific requests;

2) remote UA  overloaded, completely down or unreachable because of hardware or network malfunction.

Greylisting is good strategy for case (2) but can be real pain for case (1) even with results of real DNS lookups. Just imagine the case when all DNS records points to devices with the same SIP stack implementation  thus all of them can be greylisted by specific SIP request. Since there is no efficient method to distinguish (1) and (2) from UAC point of view it would be good to provide some means to control greylisting strategy for resipprocate stack user.

   I am ok with allowing greylisting to be disabled entirely. It is already a last-ditch measure; if you're observing actual timeouts in the field, things are pretty bad. One thing that might make sense to do is to only greylist after the tuple fails to respond to a stand-alone OPTIONS request (ie; we see a transaction timeout, and then send a quick OPTIONS request to test whether this was a problem specific to the transaction/dialog; if that OPTIONS request times out too, we greylist, or maybe even blacklist).

Yep, it makes sense but it seems this part with additional checking is completely up to you. My main goal was to understand - is there something wrong from our side (e.g. incorrect resiprocate usage) or we just faced with the issue when the existing mechanism (greylisting)  simply doesn't work well. Anyway, things are looks pretty clear now.The question is - can you develop required functionality? We may try to do it by ourselves but I doubt that our knowledge of stack internals is good enough for proper implementation.

    For making greylisting configurable, it should be really easy I think; just wrap the greylistLast() call (and add the appropriate configuration call) and you should be good to go. Controlling the handling of sip:<bar-ip-address> should not be too hard, primarily finding where to add the check I think. I can look into it tonight.

Best regards,
Byron Campen