Production Checklist

Before running SynapS3 as a long-lived single-host service, verify local disk, database health, background workers, and recovery paths.

Network Exposure

Surface	Recommended exposure
S3 API	Expose only to trusted clients or an authenticated edge.
Dashboard and Admin API	Keep on 127.0.0.1:9090; use SSH tunneling for remote access.
Metrics	Scrape from the private network or host-local agent only.

Do not publish the dashboard or Admin API directly to the internet. Settings, wallet, task retry, and S3 user endpoints are operational control surfaces.

Runtime Data

• Put /var/lib/synaps3 or ~/.synaps3 on durable storage.
• Back up config.toml, db/, and cache metadata before upgrades.
• Watch free space on the database volume and cache volume.
• Keep config.toml, .env, databases, cache data, and wallet material out of git.

Secrets and Wallet

• Store SYNAPS3_FILECOIN_PRIVATE_KEY in a host environment, .env, or secret manager.
• Confirm synaps3 admin status reports a healthy wallet after startup.
• Deposit USDFC before expected uploads. This example deposits 2 USDFC:

synaps3 wallet deposit 2 # 2 USDFC

Expected result: the wallet operation is accepted and later appears in the dashboard or GET /api/v1/wallet/operations.

Configuration Review

Check the effective settings:

synaps3 admin settings get

Review these values first:

Field	Check
admin.addr	Keep 127.0.0.1:9090 unless protected by private access.
filecoin.network	calibration until you intentionally move to mainnet
filecoin.allow_private_networks	false unless provider URLs are trusted private endpoints
cache.max_size_gb	Size it for expected upload backlog
logging.format	Compose sets json; built-in default is text.

High-risk settings require explicit confirmation:

synaps3 admin settings set filecoin.network=mainnet --yes

Monitoring

At minimum, monitor:

• GET /healthz
• GET /metrics
• cache usage
• task queue depth
• exhausted task count
• worker liveness
• provider and data set health

Treat {"status":"unhealthy"} as actionable. It means database, cache, or worker checks failed.

Upgrade Readiness

Before upgrading:

curl http://127.0.0.1:9090/healthz
synaps3 admin task stats
synaps3 admin task list --status exhausted --limit 50

Expected result: health is ok, task queues are understood, and any exhausted task has an owner decision before the process is replaced.

Recovery Entry Points

• Health issue: start with Health and Metrics.
• Failed background work: use Troubleshooting.
• Version change: follow Upgrade and Recovery.