Node Watch API

Service designed to log the status of many Rolltime node collectors.

Usage

The API has the following working methods:

• / GET: Retrieves a running list of node status.
• / POST: Stores a record of a node status. It needs the following arguments:
• id: Node id. Nodes should have unique ids.
• status: Either error or ok.
• message: A string with the message. Required in case of error.
• time: An ISO 8601 time stamp (up to seconds).

Example request:

$ curl -X POST localhost:9000/ \
  -d "id=collector-ny-01&status=error&message='Failed to collect data.'&time=2015-06-01T14:34:01'"

Docker Setup

Review the Dockerfile and run it linking to a MongoDB instance. make setup will try to setup its own collection in the instance (called scraper_status). This image doesn't need a volume mounted, but it needs the following environment variables in order to work appropriately:

• MONOGDB_SCRAPER_STATUS_USER_NAME: Dedicated user name for manipulating collections.
• MONGODB_SCRAPER_STATUS_USER_PASSWORD: Password for the user above.

Those should be passed when running the image.

$ docker run -d --name scraper_status \
  --link mongo:mongo \
  -e MONOGDB_SCRAPER_STATUS_USER_NAME=foo \
  -e MONGODB_SCRAPER_STATUS_USER_PASSWORD=bar \
  luiscape/hdx-monitor-scraper-status:latest