avalon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stephen McConnell <mcconn...@apache.org>
Subject Re: Startable doesn't has a isRunning() interface ?
Date Thu, 09 Oct 2003 00:55:41 GMT


Yannick Menager wrote:

> About JMS.... yup that's what i was starting to think too... 


Am asuming you meant to say JMX.

If that's correct - its another viewpoint - but it's not a container 
viewpoint.  JMX is for administrators to monitor and control multple 
systems.  This subject (from my point of view - which may be truly 
suspect) is about the internal engineering to ensure that the system is 
properly functioning.  Currently the container reposibilities are 
limited to assuring lifestyle and lifecycle - the startable interface 
doesn't imply any responsibility on the container to assure *lifesign* 
(which I think is something that a container should be doing).  Assuming 
something landed in the framework space to address this, then - a 
container could expose this via JMX. In the meantime a component could 
expose this as a component specific JMX attribute.

Umm - I like the word lifesign!
Plays well with lifecycle and lifestyle.

Steve.

>
> Niclas Hedhman wrote:
>
>> On Thursday 09 October 2003 04:08, Yannick Menager wrote:
>>
>>> The container would generally not call isRunning().... But for example
>>> if I have an admin console, i would like it to be able to be 
>>> informed of
>>> the state of each component. And I would like it to be a *standard* way
>>> for any Startable Component... be it using a isRunning() type of
>>> function, or using an event-based system, as long as it's standard :)
>>
>>
>>
>> Doesn't JMX allow for any method to be viewed? So, if you have 
>> isRunning() in the component implementation it will show up, won't 
>> it? (Not good with JMX, but my own ancient similar thing shows all 
>> properties that exists in the implementing class.)
>>
>> In any event, it is not possible to introduce isRunning in the 
>> Startable interface without breaking compatibility, so the option 
>> would be a sub-interface, and I would suggest that is too much for 
>> too little...
>>
>>
>> Niclas
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
> For additional commands, e-mail: dev-help@avalon.apache.org
>
>

-- 

Stephen J. McConnell
mailto:mcconnell@apache.org




---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@avalon.apache.org
For additional commands, e-mail: dev-help@avalon.apache.org


Mime
View raw message