ryanwinchester

ryanwinchester

I have an Oban Workflow that I want to limit so that only 1 Workflow will run for a user at a time, but not block for other users.

From the config it looks like I can do this per worker, but my workflow could have concurrent workers.

I don’t know if there is a way to do this with Oban config

help!

Any ideas?

Showing Posts 1 to 9

julismz

julismz

Depending in the users amount, I would use a GenServer per user, saving in the state the jobs queue and a tick function wich take the first in the list, execute and update the state…

If you want to persist the state in case of crashes and you are not using a cluster, you can save the state in ETS… and load it again in the init callback.

If you are using a cluster, you can use Horde and recover the state with :continue flag.

I’m not sure if I would use Oban for this case…

ryanwinchester

ryanwinchester OP

it’s an Oban Workflow that I am trying to run uniquely

sorenone

sorenone

Oban Core Team

Yes! Indeed, there is. With Pro v1.6 you can set a global partition by workflow_name or some other unique attribute of that users workflow. (That’s really up to you). It’s more about concurrency and not uniqueness.

config :my_app, Oban,
  queues: [
    queue_name: [limit: 20, global_limit: [allowed: 1, partition: [meta: :workflow_name]]]
  ]

In this case, each node can run up to 20 jobs but only 1 of that particular workflow. It relies on you setting a distinct workflow name for each user.

Workflow.new(workflow_name: "some-workflow-#{user_id}")
ryanwinchester

ryanwinchester OP

Nice! Is 1.6 ready for prod use yet?

Just reading 1.6 docs now.

Can I use the new context in the partition?

e.g.

global_limit: [allowed: 1, partition: [meta: :workflow_name, context: :user_id]]

*edit: looks like no. workflow_name will work, though. thanks

dimitarvp

dimitarvp

Worker name or workflow name? :thinking:

ryanwinchester

ryanwinchester OP

yeah workflow_name :sweat_smile:

sorenone

sorenone

Oban Core Team

We run it in production. It’s in rc for a while, at least a month. Totally up to you.

lessless

lessless

*edit: looks like no. workflow_name will work, though. thanks

Bummer. Is that because Oban doesn’t support partitioning by context keys yet?

sorentwo

sorentwo

Oban Core Team

The context is stored by itself, it isn’t stored with every job. That means it’s impossible to partition by anything within the context because it’s simply not available on every workflow job.

— All posts loaded —

Where Next? Top

Trending in Questions Top

stjefim
Hello! Suppose you are building workflow (order / task / payment) processing system with the following requirements: Each workflow con...
New
water
I’m aware of 2 such projects, are there others? git: GitHub - team-alembic/realworld: A fullstack Phoenix LiveView application with back...
New
itzmidinesh
Hi all, I’m working on a Phoenix project where I need to make external HTTP requests and I’m trying to choose between Req and HTTPoison....
New
peck
Hi everyone. My team and I have been working on a fairly modest app based around video streaming and chat, but we’ve landed a customer t...
New
leoschweizer
Hi, we run different LLM models across a cluster, with each node loading a configurable subset. Model-specific queues let nodes consume ...
New
csmr
Hi all, I’m new here and would like to hear peoples views on my Elixir/Phoenix based game-engine! :smiley: Mutonex gamedesign was finish...
New
Stef-Verniers
Hello I’ve been coding in Elixir for almost two years now because of my current job where we implemented Elixir in our application. But ...
New

Other Trending Topics Top

dominicletz
Hey there, I haven’t seen a list of useful Elixir specific skills for the popular AI tools yet so I’m starting one here with my first. ...
New
mudasobwa
I fully migrated to my own harness from Anthropic/Gemini and I think it’s time to share it. Welcome DSH, the DeepSeek Harness, fully writ...
New
KristerV
Hey. Is there anyone here who creates agents in their apps? Not talking about using agents, but creating them. I’m finding it pretty diff...
New
OndrejValenta
(I just needed to vent somewhere and LinkedIn is full of hope, or hype, I’m not sure which exactly) AI dream has many faces, but general...
New
webofbits
With AI doing more of the implementation work, I’ve been wondering how much coding I should deliberately keep doing myself. My main conc...
#ai
New
kip
In 2021 I started a new library called Tempo with the objective of modelling time as a set of intervals - not as instants. In 2022 I gave...
New

We're in Beta

About us Mission Statement

Options

Thread Display Mode




Thread Preview

Skip Thread Previews