hi asankha,

thanks for your answeres!
can you give me a hint if my USe/NOT USE scenraio makes sense
according to your opinion?

regards chris

On 7/7/06, Asankha C. Perera <[EMAIL PROTECTED]> wrote:
Hi Chris

Let me attempt to answer some of your questions.

regards
asankha

References
[1] http://wiki.apache.org/incubator/Synapse/InProgress

> 1) since synapse would be a really company central part of software in
> the future what is about clustering / health ckeck. how to monitor if
> synapse is working correctly? how to set it up for HAV
> what could be a scenario for an active/ active clustering approach ?
> would there be any disadvantages? (does reliable messaging work then
> (sequences))
Synapse is implemented over the Axis2 engine and uses Apache Sandesha
for WS-RM, Rampart for WS-Sec etc. Hence the answer to the above
question will be related to those projects as well. Yes, we will allow
Synapse to run in a clustered environment - but its not one of the main
priorities right now at this stage, although its of very high priority.
> 2) is there a tool to check the config if the syntax would be bootable ?
Not right now, but soon you will be able to validate a configuration
against a schema
> can config be reloaded/refresehd during runtime?
Yes, its designed to do this which is a key requirement, and right now
we are at the stage of integrating with a registry to achieve this.
Check [1] under registry thoughts
> 3) when exactly would "you" use synapse and when NOT to use it?
> my point of view would be:
>
> USE IT:
> when ws of the same type are spread over many hosts (dislocation)
> when security layer should be added to an existing ws (e.g. b2b)
> when throtteling / QOS should be added (e.g. b2b)
> when A2A integration is done (so the ws call exits a system that can
> be seen as one unit (SAP->SIEBEL)
> when protocol layer is different (e.g. JMS->SOAP)
> mapping must be done (e.g. to maintain two different interfaces map
> the old one to the new one)
>
> NOT USE IT:
> when ws are called internally (CRM->CRM) in most cases the ws are
> managed very well way inside an enterpris ready sofware
> when massive data exchange is done ? performance?
>
> have i left something out or misunderstood?
>
> 4) how does UDDI fit into the scenario of a ws mediation? how does it
> work together / integrate?
We are working on integration with a registry right now - but not UDDI.
Refer to [1] under registry thoughts
> 5) are there similar products / projects that have similar feature
> sets like synapse ? any feature matrix known? any experiences?
>
> 6) what would be the roadmap for the near future? is the synapse as it
> is today ready for production issues?
Synapse is hoping to graduate out of incubation at Apache, and we are
planning to make the 1.0 release soon.
> 7) what are perofrmance impacts when almost always using the mediation
> gateway?
>
> 8) is routing over more than one hop easilly done and or does it make
> sense?
> e..g to have a synapse server in DMZ this applies security and
> throttling and routes to an internal synapse server -> this synapse
> server routes to the original service provider


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




--
-----------------
Christian Brennsteiner

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to