You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by Swampcritter <mi...@morris.com> on 2008/05/27 16:37:40 UTC

Proof of Concept

I am trying out Active MQ for the first time, so bear with me. I have a
number of questions, so I will break them up to make it easier to reply to.

Currently where I am working -- one of the developers created an in-house
file transmission application (aka 'fileman') that uses dynamic ports to
send data back and forth between different systems. The problem is that the
software is not very stable and also a security nightmare as you pretty much
have to open every port between the base IP on two or more servers. And to
make matters worse, the original programmer never documented the 'fileman'
code nor how it actually works.

Coming from an IBM MQ Series background, I was hoping of possibly using
Active MQ as a way to get rid of this old file application transference and
dependence on cron and use something that would be a bit more secure and
easier to work with.

My question is 'Can Active MQ be setup to use triggers' -- i.e. when a
message is sent by a remote client interface to the MQ broker, can the local
queue/topic discover the inbound message and trigger an outside process that
can dump the message from the queue (or topic)?

-- Michael

-- 
View this message in context: http://www.nabble.com/Proof-of-Concept-tp17491928s2354p17491928.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.


Re: Proof of Concept

Posted by Swampcritter <mi...@morris.com>.
I guess I am not making myself clear, so I will post the question in another
way...

As the system administrator for our infrastructure, we currently have a
legacy application called 'Fileman' that is a C/C++ program that runs on
multiple servers and communicates with other Fileman instances by opening
every TCP port across our internal network and then transmits data files
from once instance to another instance, but on the file sender and receiver
locations (think of it as an enhanced 'rsync' listener):

i.e. server.mydomain.com:/web/domain.com/working/inbound

I am trying to see if Active MQ can be utilized as a replacement for a
message transaction system between multiple servers.

Coming from an IBM MQ Series background (i.e. v5.0 to 5.2.1), all I needed
to do originally would be to setup the queue manager, the queues and
channels between servers. I would also use IBM's Q.exe application for
putting and getting messages from the Queues via a Trigger Queue Monitor. 

Here is a link of what I am looking at accomplishing:
http://middleware.its.state.nc.us/middleware/Documentation/html/csqzal05/csqzal0539.htm

As for my programming background: I have ZERO Java programming experience. 
The best I can offer is shell scripting (ksh/bash/sh).


mtlZhen wrote:
> 
> you can DefaultMessageListenerContainer in Spring context. Please refer to
> Spring documentation.
> 

-- 
View this message in context: http://www.nabble.com/Proof-of-Concept-tp17491928s2354p17515423.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.


Re: Proof of Concept

Posted by mtlZhen <mt...@gmail.com>.
you can DefaultMessageListenerContainer in Spring context. Please refer to
Spring documentation.



Swampcritter wrote:
> 
> I am trying out Active MQ for the first time, so bear with me. I have a
> number of questions, so I will break them up to make it easier to reply
> to.
> 
> Currently where I am working -- one of the developers created an in-house
> file transmission application (aka 'fileman') that uses dynamic ports to
> send data back and forth between different systems. The problem is that
> the software is not very stable and also a security nightmare as you
> pretty much have to open every port between the base IP on two or more
> servers. And to make matters worse, the original programmer never
> documented the 'fileman' code nor how it actually works.
> 
> Coming from an IBM MQ Series background, I was hoping of possibly using
> Active MQ as a way to get rid of this old file application transference
> and dependence on cron and use something that would be a bit more secure
> and easier to work with.
> 
> My question is 'Can Active MQ be setup to use triggers' -- i.e. when a
> message is sent by a remote client interface to the MQ broker, can the
> local queue/topic discover the inbound message and trigger an outside
> process that can dump the message from the queue (or topic)?
> 
> -- Michael
> 
> 

-- 
View this message in context: http://www.nabble.com/Proof-of-Concept-tp17491928s2354p17500864.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.


Re: Proof of Concept

Posted by Rob Davies <ra...@gmail.com>.
On 27 May 2008, at 15:37, Swampcritter wrote:

>
> I am trying out Active MQ for the first time, so bear with me. I  
> have a
> number of questions, so I will break them up to make it easier to  
> reply to.
>
> Currently where I am working -- one of the developers created an in- 
> house
> file transmission application (aka 'fileman') that uses dynamic  
> ports to
> send data back and forth between different systems. The problem is  
> that the
> software is not very stable and also a security nightmare as you  
> pretty much
> have to open every port between the base IP on two or more servers.  
> And to
> make matters worse, the original programmer never documented the  
> 'fileman'
> code nor how it actually works.
>
> Coming from an IBM MQ Series background, I was hoping of possibly  
> using
> Active MQ as a way to get rid of this old file application  
> transference and
> dependence on cron and use something that would be a bit more secure  
> and
> easier to work with.
>
> My question is 'Can Active MQ be setup to use triggers' -- i.e. when a
> message is sent by a remote client interface to the MQ broker, can  
> the local
> queue/topic discover the inbound message and trigger an outside  
> process that
> can dump the message from the queue (or topic)?
>
> -- Michael
>
> -- 
> View this message in context: http://www.nabble.com/Proof-of-Concept-tp17491928s2354p17491928.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>

You would best to look at camel to act as a smart endpoint to dump the  
message from the queue - see http://activemq.apache.org/camel/walk-through-an-example.html




cheers,

Rob

http://open.iona.com/products/enterprise-activemq
http://rajdavies.blogspot.com/





Re: Proof of Concept

Posted by Swampcritter <mi...@morris.com>.
Actually that is what I was hoping that Active MQ already has built-in (like
MQ Series has) as I have no idea how to program in Java to even consider
creating a module.

-- Michael


ttmdev wrote:
> 
> Sounds like what you're after is a MQ-like 'trigger monitor'. Perhaps
> another option is to develop a plugin module that provides 'trigger
> monitor' like functionality. 
> 

-- 
View this message in context: http://www.nabble.com/Proof-of-Concept-tp17491928s2354p17500298.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.


Re: Proof of Concept

Posted by ttmdev <jo...@ttmsolutions.com>.
Sounds like what you're after is a MQ-like 'trigger monitor'. Perhaps another
option is to develop a plugin module that provides 'trigger monitor' like
functionality. 

Joe 


Swampcritter wrote:
> 
> I am trying out Active MQ for the first time, so bear with me. I have a
> number of questions, so I will break them up to make it easier to reply
> to.
> 
> Currently where I am working -- one of the developers created an in-house
> file transmission application (aka 'fileman') that uses dynamic ports to
> send data back and forth between different systems. The problem is that
> the software is not very stable and also a security nightmare as you
> pretty much have to open every port between the base IP on two or more
> servers. And to make matters worse, the original programmer never
> documented the 'fileman' code nor how it actually works.
> 
> Coming from an IBM MQ Series background, I was hoping of possibly using
> Active MQ as a way to get rid of this old file application transference
> and dependence on cron and use something that would be a bit more secure
> and easier to work with.
> 
> My question is 'Can Active MQ be setup to use triggers' -- i.e. when a
> message is sent by a remote client interface to the MQ broker, can the
> local queue/topic discover the inbound message and trigger an outside
> process that can dump the message from the queue (or topic)?
> 
> -- Michael
> 
> 

-- 
View this message in context: http://www.nabble.com/Proof-of-Concept-tp17491928s2354p17495505.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.