Skip to content

ECR and Processor Image

The processor Lambda
Processor Lambda container — runs PyTECGg calibration per SQS message; adopted when a Batch on Fargate parallel-execution quota increase was rejected. Image from `ghcr.io/platformfuzz/tec-processor-image` mirrored to ECR.
runs as a container image sourced from ghcr.io/platformfuzz/tec-processor-image. Terraform
HashiCorp Terraform — provisions all platform infrastructure through five modules in `terraform/`.
creates an ECR
Amazon Elastic Container Registry — stores the mirrored processor Lambda container image at deploy time.
repository; you mirror the image before applying the full processing stack.

From the demo repo root, target only the ECR resource:

Terminal window
terraform -chdir=terraform apply -target=module.processing[0].aws_ecr_repository.processor_image

Type yes when prompted. This creates the tec-processor-image ECR repository without deploying the processor Lambda yet.

Step 2 — Mirror the image from GHCR to ECR

Section titled “Step 2 — Mirror the image from GHCR to ECR”
Terminal window
./scripts/sync-processor-image.sh

By default the script syncs the latest tag, which must match processor_image_tag in Terraform (default: latest).

To use a specific tag:

Terminal window
./scripts/sync-processor-image.sh <tag>

Before mirroring, confirm the GHCR tag is single-platform:

Terminal window
docker manifest inspect ghcr.io/platformfuzz/tec-processor-image:latest
  • Docker with buildx support (docker buildx version)
  • AWS CLI credentials with ECR push permissions
  • ECR repository created in Step 1
Terminal window
./scripts/sync-processor-image.sh <tag>
terraform -chdir=terraform apply -var="processor_image_tag=<tag>"
SymptomFix
ECR push denied

Confirm AWS credentials and aws ecr get-login-password region

Multi-arch manifest rejected

Pick a single-platform tag or inspect with docker manifest inspect

Processor Lambda fails on apply

Ensure tag in ECR matches processor_image_tag variable

See Common Issues — Processor Lambda for post-deploy image problems.

Continue to Staged Apply once the image is in ECR.