Deploy the audio post-production worker template with OpenTofu

Stand up a CPU FFmpeg worker that polls an input bucket and writes mastered MP3 files to an output bucket using the validated OpenTofu template audio-worker. The worker runs on a private subnet with no floating IP; it reaches Object Storage over outbound HTTPS.

Prerequisites#

You need:

• A Quake AI account with application credentials
• OpenTofu 1.6.0 or later (installation guide)
• The AWS CLI installed (installation guide)
• OpenStack credentials sourced into the shell for minting EC2-compatible keys. See the OpenStack CLI guide.
• A copy of the audio-worker template from the template reference page
• Enough project quota for one c2a.large instance, one 40 GB boot volume, one private network, one router, and two Object Storage buckets

Step 1: Mint credentials and configure variables#

Mint EC2-compatible credentials and export the standard AWS variables:

openstack ec2 credentials create
export AWS_ACCESS_KEY_ID=YOUR_ACCESS_KEY
export AWS_SECRET_ACCESS_KEY=YOUR_SECRET_KEY
export AWS_ENDPOINT_URL_S3=https://object.us-east-1.rumble.cloud

Copy terraform.tfvars.example to terraform.tfvars and set:

key_name           = "YOUR_KEY_NAME"
input_bucket_name  = "YOUR_PROJECT_AUDIO_IN"
output_bucket_name = "YOUR_PROJECT_AUDIO_OUT"

s3_access_key = "YOUR_ACCESS_KEY"
s3_secret_key = "YOUR_SECRET_KEY"

Pick globally unique bucket names within your project. Defaults for worker_flavor, poll_interval_seconds, and target_lufs are documented on the Audio post-production worker reference page.

Step 2: Apply the template#

From the template directory, run:

tofu init
tofu plan
tofu apply

Type yes when prompted. Provisioning takes a few minutes while cloud-init installs FFmpeg and starts the worker service.

When the run finishes, note input_bucket and output_bucket from the outputs. The template allocates zero floating IPs by default.

Step 3: Verify loudness normalization#

Upload a short audio file to the input bucket. The worker accepts .wav, .flac, .mp3, .m4a, .aac, and .ogg keys:

ffmpeg -f lavfi -i "sine=frequency=440:duration=3" -ar 44100 /tmp/deploy-tone.wav
aws s3 cp /tmp/deploy-tone.wav "s3://$(tofu output -raw input_bucket)/deploy-tone.wav" \
  --endpoint-url "$AWS_ENDPOINT_URL_S3"

Wait at least one poll interval (60 seconds by default) for the worker to process the file. List the output bucket:

aws s3 ls "s3://$(tofu output -raw output_bucket)/" --endpoint-url "$AWS_ENDPOINT_URL_S3"

You should see deploy-tone-master.mp3. Download it to confirm:

aws s3 cp "s3://$(tofu output -raw output_bucket)/deploy-tone-master.mp3" /tmp/deploy-tone-master.mp3 \
  --endpoint-url "$AWS_ENDPOINT_URL_S3"
file /tmp/deploy-tone-master.mp3

The file command should report an MP3 audio stream. If nothing appears after two poll intervals, confirm the worker instance status is Active in the Console.

Next steps#

• Audio post-production worker template
• Podcast and audio publishing
• Customize template image or flavor
• S3 Storage with ACLs template

Clean up#

Empty both buckets, then run tofu destroy from the project directory:

aws s3 rm "s3://$(tofu output -raw input_bucket)" --recursive --endpoint-url "$AWS_ENDPOINT_URL_S3"
aws s3 rm "s3://$(tofu output -raw output_bucket)" --recursive --endpoint-url "$AWS_ENDPOINT_URL_S3"
tofu destroy

Delete the EC2-compatible credential with openstack ec2 credentials delete YOUR_ACCESS_KEY if you created one only for this deployment.