cloudreactor-ruby-client-wrapperio

Gem GitHub Snyk Vulnerabilities for GitHub Repo

Overview

This Ruby gem consists of 2 parts:

1) The CloudReactor API client allows ruby applications to programmatically create, monitor, and manage Tasks and Workflows. Most notably you can start and stop Tasks and Workflows by creating Task Executions and Workflow Executions.

2) The Wrapper I/O module allows ruby applications to communicate with the CloudReactor wrapper which is the parent process.

See the CloudReactor landing page to see the benefits of monitoring and managing your tasks with CloudReactor.

API client

The API client allows your program to control various entities in CloudReactor, most notable Tasks and Workflows. The code and documentation are generated by OpenAPI generator.

To start an existing Task, create a TaskExecution linked to the Task, with a status of CloudReactorAPIClient::TaskExecutionStatus::MANUALLY_STARTED:

config = CloudReactorAPIClient::Configuration.new
config.api_key['tokenAuth'] = 'YOUR API KEY'
api_client = CloudReactorAPIClient::ApiClient.new(config)

task = CloudReactorAPIClient::NameAndUuid.new
task.name = 'Sample Task'
task_execution = CloudReactorAPIClient::TaskExecution.new
task_execution.task = task
task_execution.status = CloudReactorAPIClient::TaskExecutionStatus::MANUALLY_STARTED
te_api = CloudReactorAPIClient::TaskExecutionsApi.new(api_client)
te_api.task_executions_create(task_execution)

Wrapper I/O

While running a ruby process that is wrapped by the CloudReactor proc_wrapper, you can use CloudReactorWrapperIO::StatusUpdater to send updates about your process to the wrapper. The wrapper will send those updates to the CloudReactor API server during its next heartbeat. Updates can include:

success count
error count
skipped count
expected count
last status message
any additional properties that can be serialized into JSON

Updates are communicated via a local UDP socket which means they are theoretically unreliable. However, in practice, updates are almost always picked up correctly.

Example usage

# This is necessary even in Rails, due to the module name not matching the
# package name.
require 'cloudreactor_wrapper_io'

# Use the environment variables
# PROC_WRAPPER_ENABLE_STATUS_UPDATE_LISTENER
# PROC_WRAPPER_STATUS_UPDATE_SOCKET_PORT
# PROC_WRAPPER_STATUS_UPDATE_SOCKET_BIND_PORT
# to determine configuration. These environment variables are typically passed
# to the proc_wrapper module which then passes them on to your process.
status_updater = CloudReactorWrapperIO::StatusUpdater.new()

begin
  status_updater.send_update(
    success_count: 1,
    error_count: 2,
    last_status_message: 'running'
  )
ensure
  status_updater.close()
end

License