development mode ?

[maxandersen]

One of the challenges of docker images are that they are immutable/limited/cut-down and as such rarely have features enabled for making it easy to use with development tools (such as a debugger, incremental/hot deployment etc.).

I'm often asked or suggested how to use Docker from tools, because docker is awesome so it should be easy to use for development too - it often is, but once you start outlining the extra steps and variations different OS and depending on what image you are using we already lost a lot of users.

So we should try and keep it simple to at least do some basics.

It would be great though if we could either incorporate into the images conventions on how to enable such features.

Ideas:

• Have a environment variable (ENABLE_JPDA?) to enable java debugger on a predefined port for java virtual machine launches
• have a known volume mount point for exploded deployments
• have a flag to enable/disable management (probably already fine if all that is needed is to EXPOSE 9990)
• have a -dev as example that have the above enabled (assuming it can't be done in the base image)

[maxandersen]

To give some context these ideas came up around writing these blogs: http://tools.jboss.org/blog/2015-03-02-getting-started-with-docker-and-wildfly.html and http://tools.jboss.org/blog/2015-03-03-docker-and-wildfly-2.html

[chevdor]

👍 I like the idea of a "dev" flag.
See #26

[chevdor]

See #27
The "dev mode" can be done using:
docker run -p 9990:9990 -it jboss/wildfly /opt/jboss/wildfly/bin/standalone.sh -bmanagement 0.0.0.0

[willsilvano]

Hey guys!
What you do to work with hot deploy?
Here we area migrating local environment to Docker, and we like to use this image to work with Wildfly.
We are using IntelliJ to code.
What do you recomend to work with this image in local development?