Terraform Google provider "Error 403: Quota exceeded" in CI

By Kaveh Alemi·Latchkey

Google returned a quota error: creating the resource would exceed a project quota (CPUs in a region, addresses, or an API rate). The request is valid; the project is not entitled to that much, or is calling too fast.

What this error means

terraform apply fails with "googleapi: Error 403: Quota 'CPUS' exceeded. Limit: N in region X" or a RESOURCE_EXHAUSTED / rateLimitExceeded variant.

Error: Error creating Instance: googleapi: Error 403: Quota 'CPUS'
exceeded. Limit: 24.0 in region us-central1., rateLimitExceeded

Common causes

A regional resource quota is exhausted

The project already uses its allowed CPUs, addresses, or disks in the region, so the new resource is rejected.

An API rate quota is tripped

A burst of requests on a large config exceeds a per-minute API quota, returning rateLimitExceeded transiently.

How to fix it

Request a quota increase or free usage

1. Read which quota and region the message names.
2. Request an increase for that quota, or delete unused resources to reclaim it.
3. Re-run apply once headroom exists.

gcloud compute regions describe us-central1 \
  --format="table(quotas.metric,quotas.usage,quotas.limit)"

Lower parallelism for rate quotas

For a transient rateLimitExceeded, reduce concurrent operations so requests stay under the per-minute API quota.

terraform apply -parallelism=5

How to prevent it

• Track regional resource quotas against planned usage.
• Request quota before provisioning large fleets.
• Keep parallelism modest to avoid tripping API rate quotas.

Frequently asked questions

Related guides

References

• Terraform documentation
• GitHub Actions documentation