Our verified Trial workspace has 30 days and $4.98 remaining. The existing production service is online, but repeated Dockerfile builds fail about five seconds after scheduling. The only visible build output is twice:

scheduling build on Metal builder "builder-cpzdgy"

Dashboard diagnosis reports Infrastructure Error. Please inspect the underlying build failure and builder health; if appropriate, reschedule on a healthy builder.

Region: sfo

Last successful deployment: 0432f9db-8baa-4e8b-acaf-34432ab75e30

Failed clean release archive: e4a2fe07-85e4-4263-81c7-0794a45ad095

Failed NO_CACHE=1 attempt: 30515221-eed1-4332-a356-975b1bc8bb34

Previous failed deployment: 9cb8939a-d879-415a-acb5-b77235e29200

Checks completed: repaired npm lockfile; clean local npm ci --ignore-scripts succeeds (821 packages); Linux x64 musl dry-run succeeds; uploaded a minimal release archive with --path-as-root containing Dockerfile, lockfile, app, Prisma schema/migrations and runtime scripts. The no-cache retry still failed, and the temporary NO_CACHE setting has been removed.

These local checks do not reproduce a complete Linux Docker build, so an application build error hidden by the log gap cannot be excluded. No exhausted-credit or limited-access notice is displayed.

Please confirm whether a trial-specific build restriction is involved, or provide the missing build error. We do not want to upgrade on an unsupported assumption that payment fixes this.