Are there any plans to implement a receiving proxy in a future release of
SwiftMQ? As you have said before, you must be connected to a local router
(with a local topic/queue defined) to receive messages from it. In
situations where there is one primary queue defined on a remote router,
with multiple clients that need to read messages that the thread count on
that remote router would increase dramatically (depending on connected
clients of course).
If there was a receiving proxy, then clients could connect to local
routers (increasing the thread count on the local system), and the local
router could have one connection to the remote router.
Just curious, for now we will send messages to the remote queues via the
local router (which has a connection to the router with the defined
queues) and connect to the remote router (with the queues defined) to read
messages. Assuming that sent messages will be persisted on the local
system if the remote router is down... this is correct isn't it? If the
local router goes down and comes back up it will still retain messages
destined for the remote router... right?
Can you recommend what the optimal maximum number of clients to one router
is?
I don't know if you have gotten a response back from Ravi yet, but it
seems that multiple receivers across multiple vm's works
correctly. I am not really sure why it didn't before, but it works now
(perhaps due to the upgrade to 1.2).
For JNDI lookup, can you use a JSSE socket and then use a JSSE & plain
socket for connections? I ran into trouble with the Explorer when I had a
connection factory that was JSSE with a Plain JNDI socket factory.
If we have multiple slave routers (which do not have any queues
define) and one primary router (with queues), what effect does round robin
scheduling have? It seems that with this case that the each slave would
only know about the primary via a static route. If for some reason the
connection to the primary router is down from slave0, but slave0 can
connect to slave1, and slave1 can connect to the primary, would messages
sent to slave0 (destined for the primary) be routed through slave1? If
yet, would slave0 have to have a static route to slave1?
Sorry about the huge number of questions, but they have been building up
as I experiment with the router topology and our usage of JMS.
--jason
------------------------------------------------------
SwiftMQ developers mailing list * http://www.swiftmq.com
To unsubscribe from this list, send an eMail to
[EMAIL PROTECTED] and write in the body of your message:
UNSUBSCRIBE developers <your-email-address>
Archive: http://www.mail-archive.com/developers@mail.iit.de/
