Local setup broken in 2026: Persistent ImagePullBackOff for RabbitMQ/Redis due to unmaintained Bitnami dependencies

[metesulak]

The local setup for FarmVibes.AI is effectively broken in 2026 due to unaddressed changes in Bitnami's container images and Helm charts. Despite multiple attempts following the official quickstart and troubleshooting guides, the installation fails at the Helm deployment stage for RabbitMQ and Redis with persistent ImagePullBackOff errors and Helm context deadline exceeded timeouts.

This is not a user configuration issue – it is a direct result of Bitnami's 2025 migration (moving free images/charts behind restrictions/paywalls and deprecating public Docker Hub repositories). The project's Terraform modules still point to outdated OCI repositories (oci://registry-1.docker.io/bitnamicharts), and even switching to legacy repositories does not reliably resolve the pull failures. The default Helm timeouts are insufficient for any slow or rate-limited pulls, leading to consistent failures.

Steps to Reproduce

1. Clone the repository and run farmvibes-ai local setup on a fresh Ubuntu 22.04/24.04 system (WSL2 on Windows with Docker Desktop, 32GB RAM allocated).
2. The k3d cluster creates successfully, Dapr deploys, but RabbitMQ and Redis pods remain in Init/ImagePullBackOff.
3. kubectl get pods -A output:
   NAMESPACE NAME READY STATUS RESTARTS AGE
   dapr-system dapr-operator-78979574dc-wzkkb 1/1 Running 0 23m
   dapr-system dapr-placement-server-0 1/1 Running 0 23m
   dapr-system dapr-sentry-d7d77684b-ztm4j 1/1 Running 0 23m
   dapr-system dapr-sidecar-injector-6d68957886-2bv66 1/1 Running 0 23m
   default rabbitmq-0 0/1 Init:ImagePullBackOff 0 23m
   default redis-master-0 0/1 ImagePullBackOff 0 23m
   kube-system coredns-ccb96694c-wwk8p 1/1 Running 0 23m
   kube-system helm-install-traefik-crd-l58tn 0/1 Completed 0 23m
   kube-system helm-install-traefik-nq9c2 0/1 Completed 1 23m
   kube-system local-path-provisioner-5cf85fd84d-6ms8p 1/1 Running 0 23m
   kube-system metrics-server-5985cbc9d7-kdpmm 1/1 Running 0 23m
   kube-system svclb-traefik-826efb2b-c22f5 2/2 Running 0 23m
   kube-system traefik-5d45fc8cc9-z7qlt 1/1 Running 0 23m

4. Helm deployments time out after ~5 minutes with "context deadline exceeded".

Expected Behavior

The local setup should complete successfully on a standard machine without requiring manual Terraform edits, external registries, or cloud resources.

Actual Behavior

Installation fails repeatedly at the same stage, making the local/quickstart path unusable for new users.

Environment

• OS: Ubuntu 24.04 (WSL2 on Windows 11)
• Docker: Latest Docker Desktop
• k3d: Latest
• FarmVibes.AI version: Latest main branch (git pull as of January 2026)
• Hardware: High-spec machine (32GB RAM, multiple cores allocated)

Additional Context

This issue has been reported in various forms since Bitnami's 2025 changes (see related discussions in Helm/Bitnami communities). The project has not been updated to use alternative messaging/caching providers or mirrored images, rendering the local installation path obsolete.

As an open-source project from Microsoft promoting sustainable agriculture and AI accessibility, it is disappointing that the primary quickstart method (local setup) is non-functional for over a year without any acknowledgment or workaround in the documentation. Users are forced to resort to paid cloud resources (Azure AKS) to use the tool, which contradicts the spirit of an open-source platform.

Please prioritize updating the dependencies to compatible, publicly available images or provide a maintained local alternative. Attached is the full local setup log for reference.

Thank you for addressing this critical breakage.

[nitinya9av]

Detailed Findings

Issue: Local setup broken: Persistent ImagePullBackOff for RabbitMQ/Redis due to unmaintained Bitnami dependencies

Investigation Date: February 11, 2026

Status: Confirmed - Issue is still present

---

Detailed Diagnosis

Image Availability Testing Results

RabbitMQ Images (ALL FAILED)

• ❌ bitnami/rabbitmq:4.0.4-debian-12-r1 (hardcoded in constants.py) - Not found
• ❌ bitnami/rabbitmq:latest - Not found
• ❌ bitnami/rabbitmq:4.1.3-debian-12-r1 (Helm chart default) - Not found
• ❌ bitnami/rabbitmq:3.10.8-debian-11-r4 (older version) - Not found

Redis Images

• ❌ bitnami/redis:7.4.1-debian-12-r2 (hardcoded in constants.py) - Not found
• ❌ bitnami/redis:7.4 - Not found
• ✅ bitnami/redis:latest - Works!

Verification Commands

# RabbitMQ - FAILS
$ docker pull bitnami/rabbitmq:latest
Error response from daemon: failed to resolve reference "docker.io/bitnami/rabbitmq:latest": docker.io/bitnami/rabbitmq:latest: not found

# Redis - SUCCEEDS
$ docker pull bitnami/redis:latest
latest: Pulling from bitnami/redis
49dc82c1d740: Pull complete
Digest: sha256:6b21b5e84007db6e2dc82607ecef0dc5664f7e2f76381286a0b3af0e68afba0d
Status: Downloaded newer image for bitnami/redis:latest

---

Root Cause Analysis

Primary Issue: Bitnami RabbitMQ Repository Unavailable

Bitnami RabbitMQ images are completely unavailable on Docker Hub - not just the specific version tags, but the entire repository appears to be gone/restricted. This is consistent with Bitnami's 2025 migration policy moving free images behind paywalls.

Secondary Issue: Image Incompatibility

When attempting to work around this by using the official rabbitmq:3.13-management image with modified Terraform configuration, the pod fails in the init container:

Events:
  Type     Reason     Age                    From               Message
  ----     ------     ----                   ----               -------
  Normal   Scheduled  6m44s                  default-scheduler  Successfully assigned default/rabbitmq-0
  Normal   Pulling    6m44s                  kubelet            Pulling image "docker.io/rabbitmq:3.13-management"
  Normal   Pulled     6m28s                  kubelet            Successfully pulled image "docker.io/rabbitmq:3.13-management"
  Normal   Created    4m53s (x5 over 6m28s)  kubelet            Created container prepare-plugins-dir
  Normal   Started    4m53s (x5 over 6m28s)  kubelet            Started container prepare-plugins-dir
  Warning  BackOff    103s (x23 over 6m25s)  kubelet            Back-off restarting failed container prepare-plugins-dir

Reason: The Bitnami Helm chart's prepare-plugins-dir init container expects Bitnami-specific directory paths like /opt/bitnami/rabbitmq/ which don't exist in the official RabbitMQ image structure.

---

Pod Status During Failure

$ kubectl get pods -A
NAMESPACE     NAME                                      READY   STATUS                  RESTARTS   AGE
dapr-system   dapr-operator-78979574dc-hj7xw            1/1     Running                 0          8m9s
dapr-system   dapr-placement-server-0                   1/1     Running                 0          8m9s
dapr-system   dapr-sentry-d7d77684b-gnjk7               1/1     Running                 0          8m9s
dapr-system   dapr-sidecar-injector-6d68957886-6qpr4    1/1     Running                 0          8m9s
default       rabbitmq-0                                0/1     Init:ImagePullBackOff   0          8m6s  ❌
default       redis-master-0                            1/1     Running                 0          8m5s  ✅
kube-system   coredns-ccb96694c-5nrmm                   1/1     Running                 0          8m31s
kube-system   helm-install-traefik-4zk57                0/1     Completed               1          8m32s
kube-system   helm-install-traefik-crd-c78qf            0/1     Completed               0          8m32s
kube-system   local-path-provisioner-5cf85fd84d-flbwt   1/1     Running                 0          8m31s
kube-system   metrics-server-5985cbc9d7-k2mq8           1/1     Running                 0          8m31s
kube-system   svclb-traefik-eb125728-lwv6c              2/2     Running                 0          8m18s
kube-system   traefik-5d45fc8cc9-sb4k5                  1/1     Running                 0          8m18s

Note: Redis successfully starts with latest tag, confirming that the issue is specific to RabbitMQ availability.

---

Files Requiring Updates

1. src/vibe_core/vibe_core/cli/constants.py (lines 17-18)

Current (Broken):

REDIS_IMAGE_TAG = "7.4.1-debian-12-r2"  # Image does not exist
RABBITMQ_IMAGE_TAG = "4.0.4-debian-12-r1"  # Image does not exist

2. src/vibe_core/vibe_core/terraform/local/modules/kubernetes/rabbitmq.tf

Current:

• Uses oci://registry-1.docker.io/bitnamicharts/rabbitmq Helm chart
• Chart is tightly coupled to Bitnami image structure
• Incompatible with official RabbitMQ images

Requires: Complete replacement of Helm chart or migration to alternative solution.

---