ApacheCon EU 2014

When HTTP 1.0 came out, it supported Basic Authentication. Since then, the majority of authentication operations moved to applications and frameworks, using internal datastores. As they were getting more popular, external authentication services like LDAP needed to be added to the application code or to framework's set of methods.
Introducing three Apache httpd modules that hope to simplify the application logic by moving the authentication, access control, and identity operations back to the Apache httpd, with application consuming the result of these operations: mod_authnz_pam for access control, mod_intercept_form_submit for PAM operations using application's native logon form submission, and mod_lookup_identity that can deliver user attributes and group membership from servers like FreeIPA via sssd and dbus.

This talk runs under Ed Fink's Banner of Open Sourcing Mental Illness, who has been running these types of talks for over a year now. Ed talks are very broadly on the subject, while still backing it up with his personal stories.

I'd like to mirror his work under that banner: I'd like to talk about depression, and I'd like to talk about burn out. Open source is work. Not just a work of passion, but real actual work. And it can be taxing. I'd like to discuss how.

Apache OFBiz community is mainly composed of Service Company. For each client project, it is necessary to use multiple environments : sand box - integration - pre-prod - prod.
This presentation is a feedback and a resume of our best practices in the OFBiz customization methodology and every steps of customer project management : Initialization, realization, user test, production environment upgrade
In the course of this presentation we will speak about main step and rules for OFBiz client project to make reliable and smooth deployment ;-)

Deploying enterprise-class distributed applications to Docker containers spanning multiple hosts is tricky. Dealing with port forwarding, firewalls and injecting configuration into containers for inter-dependencies are just some of the problems.

We will show how Apache Brooklyn's cloud abstractions can be used to simplify the deployment and management of a complex application to a virtual Docker infrastructure. Brooklyn will create and maintain the required Docker containers in the right locations for your application, and control and manage the software and services using policies to scale both the application and infrastructure based on their state.

We will demonstrate deploying simple multi-tier web applications and also a complex financial risk analytics application from OpenGamma, onto a Docker infrastructure.

Tomcat directly-supports a total of 6 connector configurations, many of which are confusing to newcomers or whose usage has been passed-down in arcane configurations that none of the users actually understand. In this talk, you’ll learn the differences between BIO and NIO, what APR can do for you, and whether or not using AJP makes any sense. You'll be able to see what a proxy can do for you and see what connectors are likely to be the best-fit for your environments. You will see some performance comparisons — including the results of a race against Apache httpd and other servers. A part of the presentation will be dedicated to the Tomcat8 new connector and WebSockets.

Creating excellent software is just half the story. Getting it deployed is the other half - and we all know how challenging that can be. Thanks to Docker containers, though, deploying open source applications has never been easier!

In this presentation, we'll take a look at the basics of Docker containers, then conduct a demo of deploying a service with Docker containers. Next we'll take a look at how to create containers and deploy a service using Docker, GearD, and more.