Lifeline Plugin

🌟 This plugin is available through Oban.Pro

The Lifeline plugin records queue activity as heartbeats and periodically rescues orphaned jobs, i.e. jobs that are stuck in the executing state because the node was shut down before the job could finish. Without the Lifeline plugin you need to manually rescue jobs stuck in the executing state.

Lifeline must be included used in order for Oban.Web to run properly.

Using and Configuring

To use the Lifeline plugin add the module to your list of Oban plugins in config.exs:

config :my_app, Oban,
  plugins: [Oban.Pro.Plugins.Lifeline]
  ...

There isn't any configuration necessary. By default the plugin will record heartbeats every 1 second, prune old heartbeats every 5 minutes, and rescue orphaned jobs every 1 minute. If necessary you can configure any or all of those intervals:

plugins: [{
  Oban.Pro.Plugins.Lifeline,
  delete_interval: :timer.minutes(10),
  record_interval: :timer.seconds(10),
  rescue_interval: :timer.minutes(5)
}]

This configuration will record 10x fewer heartbeats per minute, retain them 2x as long and attempt to rescue 5x less frequently. It optimizes for less database activity, at the expense of fidelity and recovery speed.

Note that rescuing orphans relies on recent heartbeats. Be sure that the delete_interval is always longer than the rescue_interval or it will look like all executing jobs are orphaned.

Rescuing Exhausted Jobs

When a job's attempt matches its max_attempts its retries are considered "exhausted". Normally, the Lifeline plugin transitions exhausted jobs to the discarded state and they won't be retried again. It does this for a couple of reasons:

  1. To ensure at-most-once semantics. If a long running job interacted with a non idempotent service and was shut down while waiting for a reply you may not want that jot to retry.
  2. To prevent infinitely crashing BEAM nodes. Poorly behaving jobs may crash the node (through NIFs, memory exhaustion, etc.) We don't want to repeatedly rescue and rerun a job that repeatedly crashes the entire node.

Discarding exhausted jobs may not always be desired. Use the retry_exhausted option if you'd prefer to retry exhausted jobs when they are rescued, rather than discarding them:

plugins: [{Oban.Pro.Plugins.Lifeline, retry_exhausted: true}]

During rescues, with retry_exhausted: true, a job's max_attempts is incremented and it is moved back to the available state.

Implementation Notes

Some additional notes about heartbeat records and how you can expect Lifeline to operate:

  • Orphan rescuing is guaranteed to only rescue jobs that belong to dead queue processes or nodes.

  • Heartbeat records are written to the oban_beats table very efficiently, as a single batch.

  • Heartbeat records are only retained for five minutes, by default. This prevents bloat or exhausting the row limit on free tier databases.

  • Only a single node will delete heartbeats or rescue orphans at any given time, which prevents potential deadlocks and churn.

Instrumenting with Telemetry

The Lifeline plugin uses Oban.Telemetry.span/3 to emit standardized plugin events.

eventmetadata
[:oban, :plugin, :start]:plugin, :action
[:oban, :plugin, :stop]:plugin, :action, :duration
[:oban, :plugin, :exception]:plugin, :action, :duration, :kind, :reason, :stacktrace

The :action key is one of :delete, :record or :rescue.