Linehaul

The Linehaul Statistics Daemon.

Linehaul is a daemon that implements the syslog protocol, listening for specially formatted messages corresponding to download events of Python packages. For each event it receives it processes them, and then loads them into a BigQuery database.

Usage

General configuration

$ python -m linehaul -h
Usage: linehaul [OPTIONS] COMMAND [ARGS]...

  The Linehaul Statistics Daemon.

  Linehaul is a daemon that implements the syslog protocol, listening for specially
  formatted messages corresponding to download events of Python packages. For each
  event it receives it processes them, and then loads them into a BigQuery database.

Options:
  --log-level [spew|debug|info|warning|error|critical]
                                  The verbosity of the console logger.  [default:
                                  info]
  -h, --help                      Show this message and exit.

Commands:
  migrate  Synchronizes the BigQuery table schema.
  server   Runs the Linehaul server.

Linehaul Server

$ python -m linehaul server -h
Usage: linehaul server [OPTIONS] TABLE

  Starts a server in the foreground that listens for incoming syslog events,
  processes them, and then inserts them into the BigQuery table at TABLE.

  TABLE is a BigQuery table identifier of the form ProjectId.DataSetId.TableId.

Options:
  --credentials-file FILENAME    A path to the credentials JSON for a GCP service
                                 account.
  --credentials-blob TEXT        A base64 encoded JSON blob of credentials for a GCP
                                 service account.
  --bind ADDR                    The IP address to bind to.  [default: 0.0.0.0]
  --port PORT                    The port to bind to.  [default: 512]
  --token TEXT                   A token used to authenticate a remote syslog stream.
  --max-line-size BYTES          The maximum length in bytes of a single incoming
                                 syslog event.  [default: 16384]
  --recv-size BYTES              How many bytes to read per recv.  [default: 8192]
  --cleanup-timeout SECONDS      How long to wait for a connection to close
                                 gracefully.  [default: 30]
  --queued-events INTEGER        How many events to queue for processing before
                                 applying backpressure.  [default: 10000]
  --batch-size INTEGER           The number of events to send in each BigQuery API
                                 call.  [default: 500]
  --batch-timeout SECONDS        How long to wait before sending a smaller than
                                 --batch-size batch of events to BigQuery.  [default:
                                 30]
  --retry-max-attempts INTEGER   The maximum number of times to retry sending a batch
                                 to BigQuery.  [default: 10]
  --retry-max-wait SECONDS       The maximum length of time to wait between retrying
                                 sending a batch to BigQuery.  [default: 60]
  --retry-multiplier SECONDS     The multiplier for exponential back off between
                                 retrying sending a batch to BigQuery.  [default: 0.5]
  --api-timeout SECONDS          How long to wait for a single API call to BigQuery to
                                 complete.  [default: 30]
  --api-max-connections INTEGER  Maximum number of concurrent connections to BigQuery.
                                 [default: 30]
  -h, --help                     Show this message and exit.

Schema Migrations

$ python -m linehaul migrate -h
Usage: linehaul migrate [OPTIONS] TABLE

  Synchronizes the BigQuery table schema.

  TABLE is a BigQuery table identifier of the form ProjectId.DataSetId.TableId.

Options:
  --credentials-file FILENAME  A path to the credentials JSON for a GCP service
                               account.
  --credentials-blob TEXT      A base64 encoded JSON blob of credentials for a GCP
                               service account.
  -h, --help                   Show this message and exit.

Discussion

If you run into bugs, you can file them in our issue tracker.

You can also join #pypa or #pypa-dev on Freenode to ask questions or get involved.

Code of Conduct

Everyone interacting in the Linehaul project's codebases, issue trackers, chat rooms, and mailing lists is expected to follow the PyPA Code of Conduct.