Skip to content

mason: add --min-instances to mason deploy for manual horizontal scaling - #511

Draft
elainewang-db wants to merge 1 commit into
databricks:mainfrom
elainemwang:mason-apps-scaling
Draft

mason: add --min-instances to mason deploy for manual horizontal scaling#511
elainewang-db wants to merge 1 commit into
databricks:mainfrom
elainemwang:mason-apps-scaling

Conversation

@elainewang-db

Copy link
Copy Markdown
Contributor

Wires --min-instances through to databricks apps compute scaling: passed to apps create for a new app, and applied via apps update before re-deploying an existing app's source (scaling isn't settable on apps deploy). Apps doesn't autoscale yet, so this is a fixed instance count.

…aling

Wires `--min-instances` through to Databricks Apps compute scaling. Apps doesn't
autoscale yet, so a fixed replica count is set by pinning compute_min_instances ==
compute_max_instances (the API rejects setting one bound alone). Instance count is
create-only: `apps create` sets it, and re-deploying an existing app only proceeds
when the requested count matches the running one — otherwise it fails with a clear
"delete and recreate to re-scale" message rather than a doomed update-API call.

Verified e2e on e2-dogfood: create at 2 instances (app RUNNING, min==max==2),
matching-count re-deploy is a no-op, and a mismatched re-scale is rejected before
any deploy.

Co-authored-by: Isaac <no-reply@databricks.com>
@elainewang-db
elainewang-db marked this pull request as draft September 3, 2026 20:52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant