Init script improvements

[Oliver <oh...@gmail.com>]

Hi,

Yesterday I gave the ActiveMQ init scripts (referenced here
http://activemq.apache.org/unix-service.html) a bit of a ribbing,
which was picked up by the eagle-eyed Bruce Snyder (seriously, I
didn't think anyone read my blog) who asked me to join this list and
suggest improvements. I've certainly made some to my company's
internal ActiveMQ package and would be happy to suggest improvements
that can be made, but I'd also like to follow any kind of existing
protocol that may be necessary.

How shall I start this process? The patches would also likely affect
other files, to do this "the right way" and I'd also be able to
contribute SPEC file patches (my reference platform is CentOS 5.5).
Anything that can get the ball rolling would be great, and I'm always
keen to have fixes applied upstream so that I don't need to do as much
monkeypatching in-house.

Best Regards,
Oliver

Re: Init script improvements

[Stan Lewis <sl...@fusesource.com>]

Contributions are always welcomed, you should have a look at:

http://activemq.apache.org/contributing.html

which lays out the process of how to submit patches.  Thanks!

On Tue, Apr 5, 2011 at 1:59 PM, Oliver <oh...@gmail.com> wrote:
> Hi,
>
> Yesterday I gave the ActiveMQ init scripts (referenced here
> http://activemq.apache.org/unix-service.html) a bit of a ribbing,
> which was picked up by the eagle-eyed Bruce Snyder (seriously, I
> didn't think anyone read my blog) who asked me to join this list and
> suggest improvements. I've certainly made some to my company's
> internal ActiveMQ package and would be happy to suggest improvements
> that can be made, but I'd also like to follow any kind of existing
> protocol that may be necessary.
>
> How shall I start this process? The patches would also likely affect
> other files, to do this "the right way" and I'd also be able to
> contribute SPEC file patches (my reference platform is CentOS 5.5).
> Anything that can get the ball rolling would be great, and I'm always
> keen to have fixes applied upstream so that I don't need to do as much
> monkeypatching in-house.
>
> Best Regards,
> Oliver
>

Re: Init script improvements

[Gary Tully <ga...@gmail.com>]

The basic information is http://activemq.apache.org/contributing.html

essentially create a jira issue and attach a patch in diff format and
grant the ASF rights to use it (tick the check box on the file attach
page).
That is the protocol. The JIRA issue makes it easy for folks to find
the improvement and provides a place to discuss that change.
If the changes are radical, start a thread on the dev list with a
reference to the JIRA.

all contributions are most welcome.

On 5 April 2011 18:59, Oliver <oh...@gmail.com> wrote:
> Hi,
>
> Yesterday I gave the ActiveMQ init scripts (referenced here
> http://activemq.apache.org/unix-service.html) a bit of a ribbing,
> which was picked up by the eagle-eyed Bruce Snyder (seriously, I
> didn't think anyone read my blog) who asked me to join this list and
> suggest improvements. I've certainly made some to my company's
> internal ActiveMQ package and would be happy to suggest improvements
> that can be made, but I'd also like to follow any kind of existing
> protocol that may be necessary.
>
> How shall I start this process? The patches would also likely affect
> other files, to do this "the right way" and I'd also be able to
> contribute SPEC file patches (my reference platform is CentOS 5.5).
> Anything that can get the ball rolling would be great, and I'm always
> keen to have fixes applied upstream so that I don't need to do as much
> monkeypatching in-house.
>
> Best Regards,
> Oliver
>



-- 
http://blog.garytully.com
http://fusesource.com

Re: Init script improvements

[Oliver <oh...@gmail.com>]

Hi again,

In followup to my original email, I've been looking at the scripts we
have in-house for ActiveMQ and there are some things which probably
warrant further discussion. What kind of parameters do we want to
support for passing to the JVM? I notice there is some kind of support
in the /bin script for activating different brokers from the get-go -
should this be parametrised and passed from (for example)
/etc/sysconfig/activemq?

I would just blindly submit patches but my gut feeling is that they
would be rejected without further discussion and consensus on the best
way of doing things.