Built-in Expire Snapshots Job

Overview

The builtin-iceberg-expire-snapshots job template removes old Iceberg snapshots and their associated metadata files. Without periodic expiration, snapshot JSON files and manifest lists accumulate indefinitely, slowing table operations and wasting storage.

This job executes Iceberg's expire_snapshots stored procedure via Spark SQL.

Job Template

Property	Value
Name	builtin-iceberg-expire-snapshots
Type	Spark
Version	v1
Main class	org.apache.gravitino.maintenance.jobs.iceberg.IcebergExpireSnapshotsJob

Parameters

Required

Key	Description	Example
catalog_name	Iceberg catalog name registered in Spark	rest_catalog
table_identifier	Fully qualified table name	db.sample

Optional

Key	Description	Default
older_than	Expire snapshots older than this timestamp (yyyy-MM-dd HH:mm:ss)	5 days ago (Iceberg default)
retain_last	Minimum number of most recent snapshots to keep	1
stream_results	Flag: presence enables streaming of intermediate delete results	disabled
spark_conf	JSON map of custom Spark configurations	none

Usage

Direct job submission via REST

curl -X POST -H "Accept: application/vnd.gravitino.v1+json" \
  -H "Content-Type: application/json" \
  -d '{
    "jobTemplateName": "builtin-iceberg-expire-snapshots",
    "jobConf": {
      "catalog_name": "rest_catalog",
      "table_identifier": "db.t1",
      "older_than": "2024-01-01 00:00:00",
      "retain_last": "3",
      "spark_master": "local[2]",
      "spark_executor_instances": "1",
      "spark_executor_cores": "1",
      "spark_executor_memory": "1g",
      "spark_driver_memory": "1g",
      "catalog_type": "rest",
      "catalog_uri": "http://localhost:9001/iceberg",
      "warehouse_location": ""
    }
  }' \
  http://localhost:8090/api/metalakes/test/jobs

Expire with only retain_last

curl -X POST -H "Accept: application/vnd.gravitino.v1+json" \
  -H "Content-Type: application/json" \
  -d '{
    "jobTemplateName": "builtin-iceberg-expire-snapshots",
    "jobConf": {
      "catalog_name": "rest_catalog",
      "table_identifier": "db.t1",
      "retain_last": "5",
      "spark_master": "local[2]",
      "spark_executor_instances": "1",
      "spark_executor_cores": "1",
      "spark_executor_memory": "1g",
      "spark_driver_memory": "1g",
      "catalog_type": "rest",
      "catalog_uri": "http://localhost:9001/iceberg",
      "warehouse_location": ""
    }
  }' \
  http://localhost:8090/api/metalakes/test/jobs

Check job status

curl -sS "http://localhost:8090/api/metalakes/test/jobs/<job-id>" | jq

Generated SQL

The job builds and executes a Spark SQL statement:

CALL `rest_catalog`.system.expire_snapshots(
  table => 'db.t1',
  older_than => TIMESTAMP '2024-01-01 00:00:00',
  retain_last => 3,
  stream_results => true
)

Only non-empty optional parameters are included. The catalog identifier is backtick-quoted for safety.

Output

On success, the job logs:

Expire Snapshots Results:
  Deleted data files: 12
  Deleted manifest files: 8
  Deleted manifest lists: 3

Verification

After running the job:

# Verify job completed
curl -sS "http://localhost:8090/api/metalakes/test/jobs/<job-id>" | jq '.job.state'
# Expected: "SUCCEEDED"

# Check staging logs
cat /tmp/gravitino/jobs/staging/test/builtin-iceberg-expire-snapshots/<job-id>/stdout.log

Relationship to Other Jobs

Job	Purpose
builtin-iceberg-rewrite-data-files	Compacts small data files
builtin-iceberg-update-stats	Collects file statistics and metrics
builtin-iceberg-expire-snapshots	Removes old snapshot metadata

These jobs are complementary. A typical maintenance workflow runs update-stats first, then compaction, then expire-snapshots to clean up the snapshot history created by compaction.