Telemetry: Cloud Storage

Contents

  1. 1 Set Up Cloud Storage
    1. 1.1 Install depot_tools
    2. 1.2 Authenticate into Cloud Storage
  2. 2 Upload to Cloud Storage
    1. 2.1 Upload your files into the bucket “chromium-telemetry”
    2. 2.2 Check the .sha1 files into repository
    3. 2.3 Download the file in python

In order to reduce the size of the Chromium source checkout, Telemetry stores binaries in Cloud Storage. This includes:

  • Recordings of webpages, known as "page set archives" or "WPR archives."
  • Support binaries, such as device/host_forwarder, ipfw, minidump_stackwalk, crash_service.
  • credentials.json, which is in Cloud Storage to provide access rights.

Many benchmarks require these files to run, and will fail without them.

Set Up Cloud Storage

Install depot_tools

Follow these instructions to install depot_tools.

Authenticate into Cloud Storage

Some files in Cloud Storage include data internal to Google or its partners. To run benchmarks that rely on this data, you need to authenticate. Run the command below and follow the instructions for authentication with your corporate account.

$ python depot_tools/third_party/gsutil/gsutil config

When prompted with “What is your project-id?”, just enter 0.

Upload to Cloud Storage

Upload your files into the bucket “chromium-telemetry”

Put the target file in the directory you want it to be when downloaded from Cloud Storage, say path/to/target. Use this command to upload:

$ python depot_tools/upload_to_google_storage.py --bucket chromium-telemetry path/to/target

A SHA1 file path/to/target.sha1 will be generated for each uploaded file.

Check the .sha1 files into repository

$ git add path/to/target.sha1

Download the file in python

from telemetry.page import cloud_storage cloud_storage.GetIfChanged('path/to/target', cloud_storage.PUBLIC_BUCKET)
Comments