examples

Examples

Each YAML here is a self-contained config that demonstrates one capability. The examples reference {{ must_env "AWS_ACCOUNT_ID" }} (or jsonnet's std.native('must_env')) and pass ebschedule validate offline — under validate, both paths fall back to a <env:AWS_ACCOUNT_ID> placeholder that satisfies the ARN check, so no env vars or AWS credentials are required for the structural pass. For apply / diff / dump / run, ebschedule auto-fills AWS_ACCOUNT_ID from sts:GetCallerIdentity if the env var is unset, so you don't need to export it explicitly when running against the calling account.

# Walk through any example without hitting AWS:
ebschedule -conf examples/03-schedule-cron-timezone.yaml validate

# Run for real against the calling account (account auto-resolved via STS):
AWS_PROFILE=my-sandbox \
  ebschedule -conf examples/03-schedule-cron-timezone.yaml apply

# Or pin AWS_ACCOUNT_ID explicitly for cross-account or CI:
AWS_PROFILE=my-sandbox AWS_ACCOUNT_ID=123456789012 \
  ebschedule -conf examples/03-schedule-cron-timezone.yaml apply

File	Demonstrates
01-rule-lambda.yaml	Smallest possible config: a Rule firing a Lambda on a rate schedule
02-rule-event-pattern.yaml	eventPattern against S3 events; Step Functions target with retry / DLQ
03-schedule-cron-timezone.yaml	EventBridge Scheduler with timezone-aware cron + a one-shot at(...) schedule
04-multi-group-schedules.yaml	Per-schedule groupName: override; one config managing several groups
05-ecs-runtask.yaml	ECS Fargate RunTask via Rule + containerOverrides (the ecschedule pattern)
06-target-types.yaml	Variety pack: Kinesis, SQS FIFO, Batch, Redshift Data, SageMaker pipeline, API Destination
07-template-funcs.yaml	env / must_env / ssm template substitution
09-base-file/	baseFile: inheritance: shared region / trackingId / tags in _base.yaml, env-specific rules in prod.yaml / staging.yaml
10-tfstate.yaml	{{ tfstate "..." }} lookup against a Terraform state file via EBSCHEDULE_TFSTATE_URL
11-jsonnet/	Jsonnet config (.jsonnet / .libsonnet); env / SSM / tfstate via std.native('env'|'must_env'|'ssm'|'tfstate'), imports, comprehensions
multi-file/	Glob-loaded configs: -conf 'examples/multi-file/*.yaml'; per-file trackingId keeps prune scoped per team

_base.yaml style files are referenced via baseFile:; they're never loaded directly. Don't glob include them in -conf '...*.yaml' patterns - load only the leaf configs:

ebschedule -conf examples/09-base-file/prod.yaml apply
ebschedule -conf examples/09-base-file/staging.yaml apply

Bootstrapping from existing AWS

If you already have rules in an account, dump them filtered by tag and use the output as a starting point:

ebschedule \
  -tag Service=my-app \
  -tag Env=prod \
  dump > my-app.yaml

dump tolerates a missing ebschedule.yaml (the default -conf path), so you don't need to pass -conf when bootstrapping from scratch.

ecschedule migration

The import-ecschedule subcommand converts an ecschedule YAML to ebschedule format:

ebschedule import-ecschedule \
  -in path/to/ecschedule.yaml \
  -account 123456789012 \
  -tracking-id my-app \
  > my-app.yaml
ebschedule -conf my-app.yaml validate

05-ecs-runtask.yaml shows the kind of output you can expect.