datera-automation-toolkit

An external quality assurance toolkit.

Pre-requisites before beginning testing

1. Cluster not deployed.
2. Datera nodes installed running Datera OS.
3. At least one "client" or head node with 2 high-speed TCP/IP connections to the Datera nodes running CentOS7.5.
4. External NTP.
5. Two /25 networks for front end traffic.
6. Two /25 networks for back end traffic.
7. Clients must have access to yum repositories, or communicate a way to install distribution packages, such as iscsi, multipath, etc. The packages will be installed as part of the cluster deploy process: iscsi-initiator-utils sg3_utils sysstat fio lsscsi python-pip curl
8. Execution host for python tests must have access to common python packages and pip repositories.
9. Execution host must have sshpass, rsync, and wget
10. Two json description files. You may call these whatever you want but they must end in .json
    • equipment description: everything needed to test a cluster. See sample/equipment.json for an example of what this looks like
    • cluster init description: everything needed to deploy a cluster. See sample/init.json for an example of what this looks like.

Installation & Usage

All commands below should be run from the directory on the executor where the toolkit has been unpacked or cloned.

Install python dependencies on executor

pip install -r requirements.txt

Deploy cluster

python deploy_cluster.py -c path/to/equipment/description.json path/to/init/description.json

The deploy_cluster script will log to a textfile in the current directory. The filename will be prefixed with "datera_cluster_deploy-", eg datera_cluster_deploy-1581051882.txt.

Verify that logs can be collected from cluster

It's recommended to you first run the "logcollect" test to ensure that logs can be gathered from the clients & storage nodes in the event of test failures.

python run_tests.py -c path/to/equipment/description.json --test-logcollect

Begin testing

python run_tests.py -c path/to/equipment/description.json