Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Handling Deployments When Provisioning JBoss domain.xml (With Ansible)

DZone's Guide to

Handling Deployments When Provisioning JBoss domain.xml (With Ansible)

· DevOps Zone
Free Resource

Download the blueprint that can take a company of any maturity level all the way up to enterprise-scale continuous delivery using a combination of Automic Release Automation, Automic’s 20+ years of business automation experience, and the proven tools and practices the company is already leveraging.

It is tricky to manage JBoss with a provisioner such as Puppet or Ansible because its domain.xml contains not only rather static configuration but also sections that change quite often such as deployments. So how can we manage the static parts of domain.xml with f.ex. Ansible while still enabling developers to deploy at will via jboss-cli (and thus changing the <deployments> sections of the file)? Here is one possible solution, based on extracting the sections from the current file and merging them into the template.

We will use XSLT stylesheets to extract the deployments sections via xsltproc run as a command, storing their output into variables that will be then used in the domain.xml template. (xsltproc has been chosen since it already was on our servers.)

Here are the Ansible tasks themselves:

## JBoss merge deployment info from the existing domain.xml if any and copy the final domain.xml
# Note: <deployments> is added at 2 places to domain.xml whenever an app is deployed via JBoss CLI, we need to keep them
- name: Make sure xsltproc is installed
  yum: name=libxslt state=installed
  tags: jboss_configuration
 
- name: Copy XSLT stylesheets for domain.xml data extraction
  copy: src={{ item }} dest=/tmp
  with_items:
    - extract_domain_deployments.xslt
    - extract_servergroup_deployments.xslt
  tags: jboss_configuration
 
- name: Check if domain.xml already present
  stat: path=/opt/jboss/jboss-{{ jboss_version_short }}/domain/configuration/domain.xml
  register: domain_xml_stat
  tags: jboss_configuration
 
- name: Extract domain/deployments from domain.xml, if present
  command: /usr/bin/xsltproc /tmp/extract_domain_deployments.xslt /opt/jboss/jboss-{{ jboss_version_short }}/domain/configuration/domain.xml
  register: domain_deployments
  when: domain_xml_stat.stat.exists is defined and domain_xml_stat.stat.exists == true
  tags: jboss_configuration
 
- name: Extract server-groups//deployments from domain.xml, if present
  command: /usr/bin/xsltproc /tmp/extract_servergroup_deployments.xslt /opt/jboss/jboss-{{ jboss_version_short }}/domain/configuration/domain.xml
  register: servergroup_deployments
  when: domain_xml_stat.stat.exists is defined and domain_xml_stat.stat.exists == true
  tags: jboss_configuration
 
- name: Copying domain.xml configuration file
  template: src=domain.xml dest=/opt/jboss/jboss-{{ jboss_version_short }}/domain/configuration/
  notify: restart jboss-eap
  tags: jboss_configuration

The extract_domain_deployments.xslt stylesheet:

<xsl:stylesheet version="1.0"
    xmlns:j="urn:jboss:domain:1.5"
    xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
 
<!-- Output the domain/deployments element of domain.xml as is, without namespaces etc. -->
 
<xsl:output omit-xml-declaration="yes"/>
<xsl:template match="/">
  <xsl:apply-templates select="j:domain/j:deployments" />
</xsl:template>
 
<xsl:template match="*" name="identity">
  <xsl:element name="{name()}">
    <xsl:apply-templates select="node()|@*" />
  </xsl:element>
</xsl:template>
 
<!-- Copy content as is -->
<xsl:template match="node()|@*" priority="-2">
  <xsl:copy>
    <xsl:apply-templates select="node()|@*" />
  </xsl:copy>
</xsl:template>
 
</xsl:stylesheet>

The (quite similar) extract_servergroup_deployments.xslt stylesheet:

<xsl:stylesheet version="1.0"
    xmlns:j="urn:jboss:domain:1.5"
    xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
 
<!-- Output the domain/server-groups/server-group/deployments element of domain.xml as is, without namespaces etc. -->
 
<xsl:output omit-xml-declaration="yes"/>
<xsl:template match="/">
  <xsl:apply-templates select="j:domain/j:server-groups/j:server-group/j:deployments" />
</xsl:template>
 
<xsl:template match="*" name="identity">
  <xsl:element name="{name()}">
    <xsl:apply-templates select="node()|@*" />
  </xsl:element>
</xsl:template>
 
<!-- Copy content as is -->
<xsl:template match="node()|@*" priority="-2">
  <xsl:copy>
    <xsl:apply-templates select="node()|@*" />
  </xsl:copy>
</xsl:template>
 
</xsl:stylesheet>

The XSLT is so complex because we only have XSLT 1.0 and need to get rid of any namespace declarations that would normally be output. (SeeE. Lenz’ Too Many Namespaces for background and better options in XSLT 2.0).

Disadvantages

When we execute Ansible with –check –diff, it will always report that it would remove deployments from the file because it doesn’t run the extraction commands.

Update: According to "Marcus Meat Ramberg"  it might be possible to force the extract commands to run even during check using always_run (since v.1.3.). Thank you!

Download the ‘Practical Blueprint to Continuous Delivery’ to learn how Automic Release Automation can help you begin or continue your company’s digital transformation.

Topics:

Published at DZone with permission of Jakub Holý, DZone MVB. See the original article here.

Opinions expressed by DZone contributors are their own.

THE DZONE NEWSLETTER

Dev Resources & Solutions Straight to Your Inbox

Thanks for subscribing!

Awesome! Check your inbox to verify your email so you can start receiving the latest in tech news and resources.

X

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}