Hi, Lloyd.
Lloyd L Chambers wrote:
> What has --force meant historically? Is replacing a war with an ear
> consistent with that past meaning?
The --force=true option means "deploy this app even if there is already
an app by that name deployed in the domain."
In v2 the CLI default was --force=false, which was actually an
oversight. I think the intent was for the default to be --force=false
so that the user must take explicit action to overwrite a
previously-deployed app. This helps avoid me accidentally clobbering
your app if we are both working on an app named "Test" in the same domain.
Because of the difficulty GlassFish v3 will have in figuring out
unambiguously what module type a given app is, I think Jerome's point is
that requiring the explicit --force=true in v3 will need to suffice for
avoiding overwriting a module of one type with a module of another type.
- Tim
>
> On May 19, 2008, at 12:32 PM, Jerome Dochez wrote:
>
>>
>> On May 19, 2008, at 12:21 PM, Jason Lee wrote:
>>
>>> On Mon, May 19, 2008 at 1:56 PM, Lloyd L Chambers
>>> <Lloyd.Chambers_at_sun.com> wrote:
>>>> I think it would be better to reject deployments that are of a
>>>> different
>>>> type; eg replacing 'war' with 'ear'. This is not a common case, so
>>>> it's
>>>> likely to be a unintentional mistake. Better to have the user
>>>> undeploy the
>>>> 'war', the deploy the 'ear'.
>>>
>>> I agree. I'd hate to be the maintainer of foo.war and start getting
>>> support calls on my broken app, only to find out that another
>>> developer/admin deployed foo.ear and silently replaced my app. In an
>>> ideal world, there would be organizational controls in place to help
>>> mitigate that kind of scenario, but we're not always in such a world.
>>> :)
>>
>> right, so this is exactly what we are planning to do by default and
>> as long as the --force = true (or redeploy) is not used. So it takes
>> an obvious step like specifying the --force=true or use the redeploy
>> command to override the foo.war with a foo.ear, I think it should be
>> enough right ?
>>
>> I am not sure that changing an obvious step like using --force=true
>> (or redeploy command) into a undeploy/deploy cycle is an added security.
>>
>>
>>
>>>
>>>
>>> --
>>> Jason Lee, SCJP
>>> Software Architect -- Objectstream, Inc.
>>> Mojarra and Mojarra Scales Dev Team
>>> https://mojarra.dev.java.net
>>> https://scales.dev.java.net
>>> http://blogs.steeplesoft.com
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe_at_glassfish.dev.java.net
>>> For additional commands, e-mail: dev-help_at_glassfish.dev.java.net
>>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe_at_glassfish.dev.java.net
>> For additional commands, e-mail: dev-help_at_glassfish.dev.java.net
>>
>
> ---
> Lloyd L Chambers
> lloyd.chambers_at_sun.com
> Sun Microsystems, Inc
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe_at_glassfish.dev.java.net
> For additional commands, e-mail: dev-help_at_glassfish.dev.java.net
>