This tool allows Alfresco Premier Services engineers or Alfresco administrators to gather information in order for a Healthcheck report to be rolled-out. It must be executed on a unix-like machine, preferably not the servers themselves. It could be for example, a Mac OSX laptop or a Linux VM satisfying requirements below.
- Ansible 2.5+
- Python 2.7+ (may work with 2.6 when disabling some auto-detection features, see [/doc/])
- unzip
- git
- Apache Maven
You need SSH access to the target infrastructure together with administrative rights on the machine (small packages may be required - e.g. python2-lxml). The playbook has been written and tested with Ansible 2.7 (also works with Ansible 2.9).
- Network SSH access to servers from the machine running hc-harvester
- Internet access on the machine running hc-harvester
- SSH user with appropriate rights (see /doc/user-permissions.md for more details)
- Access to basic package managers repositories
Edit the inventory
file and add your hostnames to the appropriate section (ansible inventory group):
- repo_tiers: Hosts running Alfresco
repository
webapp (regardless of whether this is used for ingestion, serving user requests or tracking) - share_tiers: Hosts running Alfresco
Share
webapp - index_tiers: Hosts running Solr webapp or Alfresco Search Services
Hosts are specified one per line and can be DNS names, IP addresses or ssh_config aliases
Credentials are needed to:
- Generate JMX dumps: administrative user credentials (e.g: admin/admin)
- Query the database: database read-only user credentials (e.g: alfresco/admin)
Credentails must be provided using the command below:
$ ansible-vault edit yml/roles/alfresco/vars/secrets.yml
The Vault password is set to "alfresco" by default. If you want to know more about password management see /doc/passwords.md
If you have a very basic setup - all hosts installed using the Alfresco installer with no other custom configuration - then you are good to go. Otherwise please review the advanced configuration topic before running hc-harvester.
As stated in the requirements section, make sure you can login to your hosts using ssh for Linux servers before trying to run the harvester.
The recommandation is to configure your local ssh/config
file so you can login to the machine in a non-interactive manner with the appropriate user (please refer to ssh documentation).
Once you have prepared your inventory file and provided the necessary information (like passwords and maybe other login configuration), you will need to initialize the tool:
$ git submodule init
$ git submodule update
$ cd alfresco-db-queries
$ mvn install:install-file -Dfile=./lib/ojdbc6.jar -DgroupId=com.oracle -DartifactId=ojdbc6 -Dversion=11.2.0.3 -Dpackaging=jar && mvn package
$ cd -
You only need to do this once (unless you update the alfresco-db-queries tool). You can now start the play by running:
$ ./hc-harvester
You will be asked some easy questions you need to answer - use True
or False
when needed (case matters!). You may also be prompted for passwords.
- Complete most common environement (Linux, Tomcat, PostgreSQL)
- Add appropriate tasks for Windows
- Add other J2EE servers tasks
- Plenty of stuff!