| commit | 2dd62b1ccb31338063e884b4225c107b96b4837c | [log] [tgz] |
|---|---|---|
| author | Carlos Frias <friasc@google.com> | Tue Mar 20 18:48:56 2018 -0400 |
| committer | Carlos Frias <friasc@google.com> | Tue Mar 20 18:48:56 2018 -0400 |
| tree | 44d93fe625d0c685df37aaf591d26515dc12281a | |
| parent | 945b64d75173658145a44f5125ea5c254e1d61c3 [diff] |
Updated README.md file
This is an Ansible playbook that performs a modification of the Cassandra replication factor used with Apigee components with zero downtime in a two data center topology.
This playbook uses ansible-galaxy to download roles that are used by this playbook. Once roles are downloaded the playbook and be invoked in several ways. The playbook can be invoked in its entirety, a tag can be used to invoke a subset of playbook services or the playbook may be run manually. These ways of invoking the playbook are described below.
The playbook roles and their location is found in the requirements.yml file. This file is provided to ansible-galaxy so that the roles may be downloaded.
Ansible galaxy will download the required roles when invoked with the install action and the requirements file. This is accomplished in the following way:
ansible-galaxy install -r requirements.yml
The update-replication-factor.yml script performs the update of the Cassandra replication factor for keyspaces in a Cassandra ring. This playbook assumes that you are changing Cassandra replication on a planet that is composed of two regions or two data centers.
This script expects you to provide the replication_factor setting you wish to use when the script is invoked at the command line.
The full playbook invocation is performed in the following way:
ansible-playbook update-replication-factor -e replication_factor=6 -vvv -b
This playbook enables tasks to be invoked manually. Manual invocation of this playbook is accomplished in the following way:
ansible-playbook update-replication-factor -e replication_factor=6 -vvv --step
During manual invocation the playbook will stop before every task. The user is prompted on each task
for permission to execute, stop execution or continue execution to the first task of the next play. The prompt accepts y to execute; n to stop execution or c to continue execution and stop on the first task of the next play.
ansible-playbook update-replication-factor -e replication_factor=6 -vvv --tags={{ tag name}}
ansible-playbook update-replication-factor -e replication_factor=6 -vvv --tags={{ tag name}} --step
| Playbook Name | Tag Name | Description | Usage |
|---|---|---|---|
| update-replication-factor.yml | cache | Update local cache apigee attributes | --tags=cache |
| update-replication-factor.yml | backup | Backup zookeeper and cassandra | --tags=backup |
| update-replication-factor.yml | two | Set the consistency level to TWO for RMPs and MSs | --tags=two |
| update-replication-factor.yml | local-quorum | Set the consistency level to LOCAL_QUORUM for RMPs and MSs | --tags=local-quorum |
| update-replication-factor.yml | repair | Perform nodetool repair | --tags=repair |
| update-replication-factor.yml | restart | Restart all apigee components for nodes with C*/ZK, RMPs and MSs | --tags=restart |
| update-replication-factor.yml | stop | Stop all apigee components for nodes with C*/ZK, RMPs and MSs | --tags=stop |
| update-replication-factor.yml | start | Start all apigee components for nodes with C*/ZK, RMPs and MSs | --tags=start |
| update-replication-factor.yml | update-replication-factor | Update the replication factor | --tags=update-replication-factor |
| update-replication-factor.yml | logs | Download Apigee logs from servers | --tags=logs |
This is not an officially supported Google product.