On 2015/03/25 13:25, Petr Topiarz wrote: > > > > > The direction on the "pass" rule is related to which packets that rule > > matches. So a packet comes in and matches a rule and creates a state > > table entry. If that rule has a queue specifier, the state entry is > > marked with the queue name. > > > > Now when you have any packet matching that state, that queue assignment > > is used; if the outbound interface that this packet will be sent over > > has a queue with that name, traffic will be queued. > > > > > > OK, so do you suggest that inbound traffic cannot be shaped? If that was > true, then PF after 5.5 release would not be able to do what it was able to > do before with altq. Now that is sad... > >
No, queuing incoming packets was never supported. What are you going to do with them anyway? You can't stop the other device on the network from sending to you. I think you have some confusion between "queueing inbound" and "creating state on inbound traffic". When you clear that up, the rest should all fall into place.
