Server & Storage Sync
Keep servers and buckets identical, automatically
Replicate folders between Linux, Windows and macOS servers, and between S3, Google Cloud Storage and Azure buckets. Run jobs continuously or on a schedule, see every transfer in one dashboard and stop maintaining rsync scripts and cron jobs.
What you get
- Real-time or scheduled
- Watch mode picks up file changes as they happen. Or run jobs every 5 minutes, hourly or nightly.
- Delta transfer
- Only changed blocks are sent and compressed on the way. Large files with small edits finish in seconds.
- One-way or two-way
- Mirror a source to many targets, or keep two locations in sync with a conflict policy you choose.
- Any endpoint
- Servers via the lightweight agent or SFTP, plus S3-compatible storage, Google Cloud Storage and Azure Blob.
- Verified copies
- Each file is checksummed at both ends. Failed or partial transfers retry until they match.
- Bandwidth control
- Limit throughput by job or by hour so sync never competes with production traffic.
Set up a job from the command line
Install the agent
One static binary for Linux, macOS and Windows. It only makes outbound connections, so no ports need to be opened.
curl -fsSL https://get.syncblaze.cc | sh
blaze login --token $BLAZE_TOKENMirror uploads to a second server and a bucket
Changes on web1 reach web2 and an off-site bucket within seconds.
blaze sync create uploads \
--source web1:/var/www/uploads \
--target web2:/var/www/uploads \
--target s3://media-eu/uploads \
--mode mirror --watch --delta \
--exclude "*.tmp" --bwlimit 50MB/sSupported endpoints
| Servers | Linux (x86_64, ARM64), Windows Server 2019+, macOS 13+ via agent; any host via SFTP |
|---|---|
| Object storage | SyncBlaze S3, Amazon S3, Cloudflare R2, Backblaze B2, Wasabi, MinIO and other S3-compatible services |
| Other clouds | Google Cloud Storage, Azure Blob Storage |
| Sync modes | Mirror (one-way), copy (no deletes), bidirectional with conflict policy |
| Conflict policies | Newest wins, source wins, keep both (renames the losing copy) |
| Scheduling | Real-time watch, interval (from 5 minutes) or cron expression |
| Security | TLS 1.3 in transit, scoped agent tokens, optional client-side encryption |
| Monitoring | Per-job history, transfer logs, email, Slack and webhook alerts |
Common uses
- Web server clusters
- Keep uploads and static assets identical behind a load balancer.
- Cloud migrations
- Copy buckets between providers, then keep them in sync until cutover.
- Off-site copies
- Send a live copy of important folders to another region or provider.
- Media pipelines
- Move renders and footage from edit machines to storage as soon as they are saved.
- Branch offices
- Distribute shared files to servers in several locations.
- Data lakes
- Feed new files from application servers into analytics buckets.
Questions
Is this the same as backup?
No. Sync makes the target match the source, so a deletion on the source is also deleted on a mirror. For recovery from mistakes or ransomware, use Backups, which keep older versions.
How is transfer counted?
Data sent between endpoints after compression and deduplication counts toward your monthly transfer. Checksums and metadata are free.
Do I need to open firewall ports?
No. The agent connects outbound over HTTPS to SyncBlaze. Traffic between your servers is relayed or sent directly when they can reach each other.
Can I sync between two third-party buckets?
Yes. For example Amazon S3 to Cloudflare R2. The job runs on SyncBlaze workers, so no server of yours is involved.
Works with the rest of SyncBlaze
Every plan includes all four services. Usage from each counts against the same plan limits, so you pay for one account, not four.
Start on the free plan
No card needed. Upgrade when your project grows, or talk to us about a custom setup.