RTFS Scheduled Jobs Reference
Look up background jobs RTFS runs for Federation health and event processing.
RTFS runs background jobs to maintain Federation health, process events, manage database partitions, collect metrics, and perform cleanup. This reference lists all scheduled jobs, their purpose, and configuration.
Cron-Based Jobs
These jobs run on fixed cron schedules.
| Job | Purpose | Cron Property | Default Schedule |
|---|---|---|---|
| Mirror Heartbeat | Sends heartbeats to all Federation members to monitor connectivity and detect inactive or unreachable members. | rtfs.mirror.heartbeat.cron | */20 * * ? * * * (every 20 seconds) |
| Stream Recovery | Detects and recovers stalled event streams that have stopped processing due to transient errors. | rtfs.stream.recovery.job.cron | */10 * * ? * * * (every 10 seconds) |
| Stream Lag Collection | Collects worker lag metrics and identifies repositories with the highest synchronization delay. | rtfs.stream.lag.collector.job.cron | 0 */2 * ? * * (every 2 minutes) |
| File List Tasks | Processes pending file list preparation tasks required for Full Sync operations. | rtfs.file.list.tasks.job.cron | */30 * * ? * * * (every 30 seconds) |
| Readiness Probe | Performs internal readiness checks to determine if RTFS can serve traffic. | rtfs.readiness.job.cron | */15 * * ? * * * (every 15 seconds) |
| Partition Management | Evaluates database partition utilization and creates new partitions for event and message blob tables when thresholds are reached. | rtfs.partition.job.interval.cron | 0 0/30 * ? * * (every 30 minutes) |
| Sub-Partition PK Monitoring | Monitors sub-partition primary key counts and reports metrics for capacity planning. | rtfs.sub.partition.pk.monitoring.cron | 0 0 0 ? * * * (daily at midnight) |
| Events Cleanup | Removes processed events older than the retention period to control database growth. | rtfs.events.cleanup.cron | 0 0 1 * * ? (daily at 1:00 AM) |
| Full Sync Dir Cleanup | Cleans up temporary directories and files created during Full Sync operations. | rtfs.full.sync.cleanup.cron | 0 0/10 * * * ? * (every 10 minutes) |
| Metrics Collection | Collects general RTFS metrics (member counts, config consistency, and connection pool utilization). See the Open Metrics Reference for exposed metric names. | rtfs.collect.metrics.job.cron | */59 * * ? * * * (approximately every minute) |
| Call Home Metrics | Collects aggregated usage metrics for JFrog analytics. | rtfs.call.home.collect.metrics.job.cron | 0 0 3 * * ? (daily at 3:00 AM) |
| Orphan Cleanup | Removes orphaned records from the database (for example, member state entries for repositories that no longer exist). | rtfs.orphan.cleanup.cron | 0 0 2 * * ? (daily at 2:00 AM) |
| Configuration Sync | Aligns repository configuration between RTFS, Artifactory, and remote members. Detects configuration drift and repairs inconsistencies. | rtfs.repo.config.sync.job.interval.cron | 0 0 * * * ? (hourly) |
| Configuration Integrity | Validates repository configuration integrity as a secondary consistency check. | rtfs.repo.config.integrity.job.interval.cron | 0 30 * * * ? (hourly at :30) |
Interval-Based Jobs
These jobs run at fixed intervals rather than cron expressions.
| Job | Purpose | Interval Property | Default Interval | Initial Delay Property | Default Delay |
|---|---|---|---|---|---|
| Full Sync | Coordinates full synchronization between members that are in PENDING_FS state. Processes members in batches. | rtfs.full.sync.job.interval.sec | 120 seconds | rtfs.full.sync.job.initial.delay.min | 1 minute |
| Auto-Healing | Checks for members in error states and initiates recovery (resets exhausted queues, replays failed binary tasks). | rtfs.auto.healing.job.interval.sec | 20 seconds | rtfs.auto.healing.job.initial.delay.min | 1 minute |
| Metrics Export | Exports collected metrics for external consumption. | rtfs.export.metrics.job.interval.sec | 60 seconds | rtfs.export.metrics.job.initial.delay.min | 1 minute |
Continuous Processes
These services run continuously rather than on a schedule.
| Process | Purpose | Key Properties |
|---|---|---|
| Worker Runner | Continuously polls event streams and processes event batches for propagation to remote members. | rtfs.worker.batch.size (100), rtfs.worker.max.events.processed.in.batch (200) |
| Binary Tasks Handler | Continuously processes binary transfer tasks (download from source, upload to target). | rtfs.binary.tasks.handler.run.duration (PT1S), rtfs.binary.tasks.handler.tenants.thread.pool.size (100) |
Tuning Guidelines
- High-volume deployments: If you observe stream lag growing, reduce the
rtfs.stream.recovery.job.croninterval and increasertfs.worker.batch.size.
For example, to run Stream Recovery every 5 seconds instead of the default 10:
rtfs.stream.recovery.job.cron=*/5 * * ? * * *- Database growth: If database size grows faster than expected, reduce
rtfs.events.cleanup.retention.time.Daysand verifyPartitionMakerJobis running successfully. See RTFS Database Requirements for schema details. - Full Sync performance: Increase
rtfs.full.sync.job.run.members.batch.sizeto process more members per cycle. Decreasertfs.full.sync.grace.period.minto allow more frequent syncs for the same member. - Auto-healing frequency: The default 20-second interval is suitable for most deployments. Increase it only if auto-healing log noise is excessive in stable environments.
Related Topics
Updated about 7 hours ago
Did this page help you?