stackabletech / stackabletech/nifi-operator

NiFi 1.25.0 UI access unstable with more than one replica

Open
#630 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

type/bug
Dominant language
Rust
Stars
48
Forks
12
Avg merge
11h 26m
Merged PRs (30d)
10

Description

In NiFi 1.25.0 the access JWT changed its issuer/sub to a specific pod. Due to the round robin of the service / nodeport (it should be fine with load balancers and sticky sessions), it happens that you are connecting with a JWT from pod X to pod Y which is rejected and basically logs you out or results in strange behavior... This makes the UI more or less not usable for proper production.

We should create a single pod service (or adapt the create-reporting-task service that talks to only one pod) to avoid this behavior. It probably makes sense to deploy one (single-pod) service per pod (listener??).

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by tracing the Kubernetes service and NodePort routing, then inspect the create-reporting-task service mentioned in the issue and how requests are distributed across NiFi replicas. Compare JWT issuer/sub values across pods and verify the chosen routing approach keeps the UI usable with more than one replica.

Written by the indexing model from the issue text.

Assessment

Tech stack
kubernetes, rust
Domain
authentication, infrastructure, networking
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.