Guides
Start here
Section titled “Start here”Quick startCreate a project, add a service, and deploy from the canvas.
The basicsProjects, canvas, services, variables, and environments.
StackBlaze for AgentsConnect Cursor, Claude, or any MCP client to your stack.
CLIInstall stackblaze, log in, and deploy from the terminal.
Config as codeDescribe a project as stackblaze.yaml.
Languages & frameworksNixpacks runtimes and per-framework deploy notes.
Platform
Section titled “Platform”Migrate from HerokuMove the service, config vars, and databases with minimal downtime.
Develop
Section titled “Develop”ServicesGit, Docker, templates, cron, and databases as canvas tiles.
VariablesService config, secrets, and database-linked values.
EnvironmentsProduction, custom environments, and pull-request previews.
Staged changesReview canvas edits, then deploy.
Public HTTPGive a service a public URL and TLS.
Private servicesAlways-on processes with no public domain.
Cron jobsScheduled runs on standard cron syntax.
Static sitesBuild a frontend and serve the output as a service.
Deploy from GitHubConnect a repo and auto-deploy on every push.
DockerBuild from a Dockerfile or run a registry image.
ImportCompose, render.yaml, and fly.toml onto the canvas.
Auto deploy on pushEvery push to the deploy branch triggers a build.
PR previewsA live URL for every pull request, torn down on close.
Deploy hooksTrigger a redeploy from CI or an external system.
RollbacksRevert to any previous successful deployment.
MonorepoRoot directories and build filters for multi-app repos.
Horizontal scalingReplica counts and spreading load across instances.
AutoscalingScale replicas from CPU or memory thresholds.
Memory & CPU limitsPod sizes, requests, and what happens at the cap.
Zero-downtime deploysRolling updates that wait for health checks.
Health checksReadiness probes so traffic only hits healthy instances.
Databases
Section titled “Databases”PostgreSQLManaged Postgres with Standard and HA topologies.
MySQLManaged MySQL for apps that expect the MySQL protocol.
RedisValkey-compatible cache and queue.
MongoDBDocument store with a MongoDB-compatible API.
CockroachDBDistributed SQL with Postgres compatibility.
ClickHouseColumnar analytics database.
OpenSearchFull-text search and log analytics.
MilvusVector database for embeddings and similarity search.
CassandraWide-column store for high-write workloads.
ScyllaDBCassandra-compatible low-latency store.
CouchDBHTTP document database with replication.
MemcachedIn-memory cache for simple key-value workloads.
Backups & recoveryScheduled backups and restore workflows.
Point-in-time recoveryRestore Postgres to any moment inside the retention window.
VolumesDisk attached to a service; survives restarts and redeploys.
KafkaManaged Kafka for event streams.
RabbitMQManaged RabbitMQ for work queues.
Networking
Section titled “Networking”Public networkingPlatform domains, custom hostnames, and automatic TLS.
SSL / TLSCertificates, HTTPS redirects, and cipher defaults.
Private networkingService-name DNS inside an environment.
WebSocketsLong-lived connections without extra platform config.
RedirectsHTTP redirects at the edge.
Observability
Section titled “Observability”Metrics, logs, and tracesOpen Coroot from a service; search logs from the dashboard or MCP.
TroubleshootingBuild failures, crash loops, and health-check misses.
Account
Section titled “Account”VariablesService config, secrets, and database-linked values.
Secret filesMount files that should not live in git.
Team & permissionsOwner, Admin, Developer, and Viewer roles.
Two-factor authTOTP 2FA and recovery codes.
IP allowlistingRestrict dashboard and app access by source IP.
Reference
Section titled “Reference”API referenceREST API for projects, services, stackblaze.yaml, and tokens.
CLIInstall stackblaze, log in, and deploy from the terminal.
Limits & quotasPlan caps for services, databases, builds, and bandwidth.
Pricing & plansHow usage maps to Hobby, Pro, and Enterprise plans.