activemq-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bbuzzard <Billy.Buzz...@bnsflogistics.com>
Subject RE: Does Publisher Flow Control work across a Network Bridge?
Date Thu, 17 Sep 2015 16:22:02 GMT
Tim,
                I'm using ActiveMQ-5.5.1 (yes I know there are newer versions, but we're not
ready to upgrade our production code just yet.  We have a star topology with multiple end
point brokers feeding a central broker (hub) that controls routing of messages from end point
to end point.

e.g.        Broker A (end point) --> Broker (hub) --> Broker B (end point) --> Broker
(hub) --> Broker C
                (All end point brokers send their messages to the same queue on the hub called
"hub")

Here's the problem I'm having.  Files from Broker A go through the hub on their way to Broker
B.  Broker B captures information from the files and sends the files back to the hub where
they can be routed to Broker C.  Most of the time the memory usage stays below 10%, but once
and a while it climbs up very rapidly 25%, 50%, 75% and then all logging stops in Broker B
and file transfers between the hub and Broker B appear to stop until I restart Broker B. 
If I had to guess I would say that the high water mark is reached in the hub (since it's receiving
files from other brokers not shown above) and it halts the Producer Flow from Broker B, but
I'm not seeing anything that indicates the flow can be resumed.

Can you suggest some ways to isolate this problem?  I was thinking of using the existing Producer
Flow Control test and modifying the test to match my topology, but that might take a little
time for me to come up to speed on.

From: Tim Bain [via ActiveMQ] [mailto:ml-node+s2283324n4702062h1@n4.nabble.com]
Sent: Thursday, September 17, 2015 9:04 AM
To: Billy Buzzard
Subject: RE: Does Publisher Flow Control work across a Network Bridge?

I've observed it in practice, so I've never felt the need to search for a
unit test that shows the same thing, sorry.  Luckily they're all open
source, so you can do that yourself if you're so inclined.
On Sep 17, 2015 7:09 AM, "bbuzzard" <[hidden email]</user/SendEmail.jtp?type=node&node=4702062&i=0>>
wrote:

> I assumed it worked that way, but I wasn't sure.  Thank you.
>
> Is there a specific Test Case that proves this?
>
> From: Tim Bain [via ActiveMQ] [mailto:
> [hidden email]</user/SendEmail.jtp?type=node&node=4702062&i=1>]
> Sent: Wednesday, September 16, 2015 11:04 PM
> To: Billy Buzzard
> Subject: Re: Does Publisher Flow Control work across a Network Bridge?
>
> In that situation, broker 2 will PFC broker 1, and then (later, once broker
> 1 hits a usage limit) broker 1 will PFC the producer.
>
> In a NoB, each broker acts as a producer or a consumer on each
> networkConnector, so the brokers experience the same interactions that
> clients do, and any interaction between two clients (sending a message,
> PFC, etc.) happens because that interaction has happened repeatedly across
> each network link where the brokers play the roles of producer and/or
> consumer.
> On Sep 16, 2015 12:03 PM, "bbuzzard" <[hidden
> email]</user/SendEmail.jtp?type=node&node=4702052&i=0>>
> wrote:
>
> > I have two brokers, each on its own machine, connected together via a
> > Network
> > Bridge where one end uses a full duplex connection.  I'm sending in
> > messages
> > to the queue shared by both machines at a high rate and I'm expecting
> > Publisher Flow control to kick in and not let me publish anymore messages
> > until the consumer on the other machine clears out some messages.
> >
> > Does Publisher Flow Control work across two machines using a Network
> Bridge
> > or is it limited to a single machine?
> >
> >
> >
> > --
> > View this message in context:
> >
> http://activemq.2283324.n4.nabble.com/Does-Publisher-Flow-Control-work-across-a-Network-Bridge-tp4702039.html
> > Sent from the ActiveMQ - User mailing list archive at Nabble.com.
> >
>
> ________________________________
> If you reply to this email, your message will be added to the discussion
> below:
>
> http://activemq.2283324.n4.nabble.com/Does-Publisher-Flow-Control-work-across-a-Network-Bridge-tp4702039p4702052.html
> To unsubscribe from Does Publisher Flow Control work across a Network
> Bridge?, click here<
> >.
> NAML<
> http://activemq.2283324.n4.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml
> >
>
>
>
>
> --
> View this message in context:
> http://activemq.2283324.n4.nabble.com/Does-Publisher-Flow-Control-work-across-a-Network-Bridge-tp4702039p4702058.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.

________________________________
If you reply to this email, your message will be added to the discussion below:
http://activemq.2283324.n4.nabble.com/Does-Publisher-Flow-Control-work-across-a-Network-Bridge-tp4702039p4702062.html
To unsubscribe from Does Publisher Flow Control work across a Network Bridge?, click here<http://activemq.2283324.n4.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=4702039&code=QmlsbHkuQnV6emFyZEBibnNmbG9naXN0aWNzLmNvbXw0NzAyMDM5fC05MjA3NTk2MzE=>.
NAML<http://activemq.2283324.n4.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>




--
View this message in context: http://activemq.2283324.n4.nabble.com/Does-Publisher-Flow-Control-work-across-a-Network-Bridge-tp4702039p4702064.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message