How does the semantics of message routing work (involving resources)?
I have come up with the following (for a message sent to [EMAIL
PROTECTED]<mailto:[EMAIL PROTECTED]>):
1. Is the destination on this server?
a. No -> Open or use S2S connection and route the packet there.
2. Does it have a resource?
a. No -> Find the lowest resource and route.
3. Is the resource online?
a. Yes -> Route it to that connection.
b. No -> Is the user online?
i. Yes ->
*What to do here?* (Assuming that [EMAIL PROTECTED]<mailto:[EMAIL PROTECTED]>
or [EMAIL PROTECTED]/resource<mailto:[EMAIL PROTECTED]/resource> is online)
ii. No ->
Store it in DB for later.
I know this has come up before, but what about duplicate destinations? If a
*nuclear reactor is overheating* packet is sent to [EMAIL
PROTECTED]<mailto:[EMAIL PROTECTED]> and he has three resources:
[EMAIL PROTECTED]/mobile<mailto:[EMAIL PROTECTED]/mobile> (0)
[EMAIL PROTECTED]/pc<mailto:[EMAIL PROTECTED]/pc> (1)
[EMAIL PROTECTED]/allfail<mailto:[EMAIL PROTECTED]/allfail> (0)
Thus now he has identical resources in case a message is sent to his mobile,
and it needs urgent action (but he isn't online). Can't messages be routed to
all resources with identical priorities?
Thanks for your time.
