Question 1

How does Verk compare to Oban for Elixir?

Accepted Answer

Verk is Redis-based and follows Sidekiq/Resque patterns, focusing on queue isolation and reliability, while Oban uses PostgreSQL and offers features like unique jobs. Choose Verk if you're already using Redis or migrating from a Sidekiq background.

Question 2

How to deploy Verk on Kubernetes without issues?

Accepted Answer

Use a StatefulSet with stable pod names to ensure unique node_id, and set terminationGracePeriodSeconds to match Verk's shutdown timeout. Avoid Deployments to prevent node_id conflicts and job stuck scenarios, as detailed in the Kubernetes deployment guide.

Question 3

Does Verk support job priorities within a single queue?

Accepted Answer

No, Verk handles priorities through separate queues with different worker counts; it lacks built-in priority levels within a queue, so you need to manage multiple queues for priority-based processing.

Question 4

How to set up custom retry times in Verk?

Accepted Answer

Define a retry_at/2 function in your worker module to specify custom retry delays, overriding the default exponential backoff. This allows precise control, such as linear delays based on retry count.

Question 5

What happens if a Verk node crashes with jobs in progress?

Accepted Answer

Verk automatically re-enqueues jobs from crashed nodes using Redis tracking, but this requires unique node_id and careful deployment to avoid duplicates, as explained in the reliability and deployment sections.

Question 6

Can Verk integrate with monitoring systems like Prometheus?

Accepted Answer

Yes, but indirectly through the GenStage EventProducer for job lifecycle events, or by using third-party projects like Verk Stats. You'll need to build custom handlers for direct integration, as built-in metrics are limited.

verk

What is verk?

Overview

Use Cases

Best For

Related Projects

Found a gem we're missing?

Not Ideal For

Pros & Cons

Pros

Cons

Open Source Alternative To

Frequently Asked Questions