stackblaze scale
Sets the replica count or autoscaling of the linked app’s web and worker processes. Everything in one command is applied as one release. Same command as stackblaze apps scale.
stackblaze scale [web=N] [worker=N] [--autoscale <process=min:max>]... [--autoscale-off <process>]... [options]stackblaze scale -r <n> [--process web|worker]web=N and worker=N set a fixed replica count, 0 to 50. 0 stops that process.
Options
Section titled “Options”| Flag | Description |
|---|---|
--autoscale <process=min:max> |
Turn on autoscaling between min and max replicas, e.g. web=2:5 (repeatable). min is at least 1 |
--autoscale-off <process> |
Turn off autoscaling for a process, back to a fixed count (repeatable) |
--skip-deploy |
Stage the change instead of deploying it; ship it later with stackblaze pending apply |
-m, --message <text> |
Release message shown in the deployment history |
-r, --replicas <n> |
Replica count for --process, 0 to 50 (older single-process form) |
--process <name> |
web (default) or worker, for -r |
Also accepts the scope and output flags: -p, --pipeline, --phase, -a, --app, --token, --api-url, -o, --output, --json.
Examples
Section titled “Examples”stackblaze scale web=3 # 3 web replicasstackblaze scale web=2 worker=1 # both processes, one releasestackblaze scale --autoscale web=2:5 # autoscale web between 2 and 5 replicasstackblaze scale --autoscale-off web # back to a fixed replica countstackblaze scale web=4 --skip-deploy # stage, apply laterstackblaze scale -r 0 --process worker # older single-process formHow it works
Section titled “How it works”The CLI sends only the requested change; the server applies it and records one release in the deployment history, with your -m message. Replica counts are checked against your plan’s per-service limit. The text output shows the previous count, for example web → 3 replica(s) (was 1).
To stop every process of an app, including cron jobs, and bring it back later as it was, use stackblaze down and stackblaze resume.