Large production systems may consist of hundreds of dynamic components such as instances, disks, networks, and firewall rules. These components are transient in nature, and often it is important to understand how they change over time. For example, a time series view of how many instances with a given tag were created and deleted in a given week can help you understand usage patterns of your cloud deployment.
This solution aims to help developers of large systems get answers to history-related questions.
This solution builds on the Data Pipeline Solution to create a workflow that captures snapshots of Compute Engine deployment configurations into BigQuery. By using the App Engine Cron Service, deployment configurations can be captured as time series data and queried using BigQuery. A sample queries cookbook are provided for you to get started on the analysis. The workflow is best illustrated with the following diagram:
Copyright 2014 Google Inc. All Rights Reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
[http://www.apache.org/licenses/LICENSE-2.0]
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
This is not an official Google product.
Follow the Data Pipeline Installation Guide to install the Data Pipeline Solution.
Two sample pipelines are provided for logging Compute Engine instance, disk, and zone operations data into BigQuery. The instance and disk data is processed by one pipeline. The zone operations data is processed by a separate pipeline. Instance and disk data is transient in nature, so the pipeline should be scheduled to run more frequently. The data for zone operations is not transient, so the pipleline does not have to run as often.
The cloud_history_pipeline.json pipeline included in Data Pipeline reads the current Google Compute Engine instance and disk data using the Compute Engine API interface. The data is then transformed to a JSON format that is compatible with BigQuery and loaded into BigQuery.
You can find the file in the app/static/examples directory.
Follow the Data Pipeline instruction to create the pipeline for
instance/disk data.
The operations_history_pipeline.json sample pipeline included in this package reads the current Google Compute Engine zone operations data using the Compute Engine API interface. The data is then transformed to a JSON format that is compatible with BigQuery before loading it into BigQuery.
You can find the file in the app/static/examples directory.
Follow the Data Pipeline instruction to create the pipeline for
zone operations data.
Copy the 'Run URL' link for each pipeline that you created. The url should be in the following format. The pipeline variables, if defined, are appended as query parameters.
/run/cloud-history/<unique id>
Change the directory to where you downloaded the data pipeline
package. Create a cron.yaml file in your app directory
(along with an app.yaml).
Use the following as a template and fill in your values.
cron:
- description: instance and disk data
url: <Run URL for instance and disk data>
schedule: every 15 minutes
target: <YOUR APP VERSION>
- description: operations data
url: <Run URL for operations data>
schedule: every day 06:00
timezone: America/Los_Angeles
target: <YOUR APP VERSION>Update the cron configuration while you are still in your app directory:
appcfg.py update_cron . --oauth2After the pipeline is set up, you can use the BigQuery Web User Interface for analysis. Make sure that the project where you loaded the data is the current project. You should see the cloud_history dataset. Within the dataset, there should be three tables: Disks, Instances, and zoneoperations.
Use the sample query cookbook to get started with timeline analysis.