You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@geronimo.apache.org by Ashish Jain <as...@gmail.com> on 2010/04/21 13:59:58 UTC
Modify the redeploy behaviour
Hi All,
The current behavior of redeploy is to set the state of all the childs in
the stop state and parent in the running state on success.
One can go ahead and manually start all the childs. This may be fine if your
deployment is on one single machine. However in a
scenario in which there are thousand of systems it may be a problem to start
all the childs manually. In my opinion this is not a
correct behavior. Instead all the childs along with parent should be
started on successful redeployment.
I have generated a fix for this. If we are ready to change this behavior I
will share further details on the fix.
Comments please?
Thanks
Ashish
Re: Modify the redeploy behaviour
Posted by Ashish Jain <as...@gmail.com>.
Ivan, thanks for you advice. I have created a JIRA(
https://issues.apache.org/jira/browse/GERONIMO-5265) and uploaded a patch.
Kindly provide your comments.
On Wed, Apr 21, 2010 at 6:58 PM, Ivan <xh...@gmail.com> wrote:
> It seems a good idea, but I think that "all the children" should be the
> children which are in the running state before the deployment.
>
> 2010/4/21 Ashish Jain <as...@gmail.com>
>
> Hi All,
>>
>> The current behavior of redeploy is to set the state of all the childs in
>> the stop state and parent in the running state on success.
>> One can go ahead and manually start all the childs. This may be fine if
>> your deployment is on one single machine. However in a
>> scenario in which there are thousand of systems it may be a problem to
>> start all the childs manually. In my opinion this is not a
>> correct behavior. Instead all the childs along with parent should be
>> started on successful redeployment.
>>
>> I have generated a fix for this. If we are ready to change this behavior I
>> will share further details on the fix.
>>
>> Comments please?
>>
>> Thanks
>> Ashish
>>
>>
>
>
> --
> Ivan
>
Re: Modify the redeploy behaviour
Posted by Ivan <xh...@gmail.com>.
It seems a good idea, but I think that "all the children" should be the
children which are in the running state before the deployment.
2010/4/21 Ashish Jain <as...@gmail.com>
> Hi All,
>
> The current behavior of redeploy is to set the state of all the childs in
> the stop state and parent in the running state on success.
> One can go ahead and manually start all the childs. This may be fine if
> your deployment is on one single machine. However in a
> scenario in which there are thousand of systems it may be a problem to
> start all the childs manually. In my opinion this is not a
> correct behavior. Instead all the childs along with parent should be
> started on successful redeployment.
>
> I have generated a fix for this. If we are ready to change this behavior I
> will share further details on the fix.
>
> Comments please?
>
> Thanks
> Ashish
>
>
--
Ivan