Obstor

Obstor is self-hosted object storage for teams that want S3-compatible access and SFTP in one system. It suits operators building distributed storage for application, analytics, or machine-learning data. The Apache 2.0 software costs nothing, but its standalone mode is for evaluation; production needs should be planned around distributed deployment.

Overview

Obstor accepts existing S3 SDKs, command-line tools, and applications without code changes, so teams can keep familiar S3 workflows while running the storage server themselves. Its native SFTP service gives people and systems another way to upload files, which Obstor stores as S3 objects.

Connected nodes automatically shard and replicate objects, with each object stored in at least two global locations. Obstor supports continuous replication, multi-site replication, and failover orchestration. These capabilities make it a fit for teams responsible for operating storage across sites, but they also make deployment design central to the decision.

Key features

Data protection and recovery

Configurable erasure-coding parity protects against drive and node failures. Object versioning and lifecycle policies support history, recovery, and retention needs. The important constraint is that standalone servers are intended for development and evaluation; versioning, object locking, and bucket replication require distributed deployment with erasure coding. Teams that need those controls should plan for the distributed configuration rather than rely on a single server.

Encryption and access control

Server-side encryption supports SSE-C and SSE-S3, including KMS integration. Key derivation uses HMAC-SHA-256, with AES-256-GCM or ChaCha20-Poly1305 for authenticated encryption. IAM supports users, groups, policies, and service accounts using S3 and AWS IAM policy syntax. This combination is useful for teams that need familiar S3-style permissions and encryption options while managing their own infrastructure.

Integration and deployment

Event notifications can be sent to Kafka, NATS, AMQP, Redis, PostgreSQL, Elasticsearch, and webhooks, which gives teams several ways to connect storage events to existing systems. Installation options include a single binary, Docker, a Kubernetes Helm chart, or a source build. The Docker Compose example creates four distributed instances and supports up to 16 Obstor services; it is a bounded example, not a substitute for planning deployment capacity.

Pricing

Open source: free under the Apache 2.0 license, for self-hosted use. There is no paid tier described here, so the software itself has no subscription price. Self-hosting still means operating the deployment, and the standalone option is limited to development and evaluation. Readers who need versioning, object locking, or bucket replication must account for the distributed deployment with erasure coding.

Platforms

Obstor supports API, Linux, macOS, self-hosted, web, and Windows. Official standalone instructions cover macOS, GNU/Linux, and 64-bit Windows. Those standalone instructions make evaluation accessible across common desktop and server systems, but do not change the production constraint: the specified data-management features require distributed deployment.

Who it's for

Obstor is aimed at teams that can run and manage their own storage and want to retain S3-compatible tools while adding SFTP ingestion. Its stated workloads include machine learning, analytics, and application data. It is particularly relevant when multi-site replication, configurable erasure coding, and IAM are part of the storage design.

It is a weaker fit for readers who want a managed service or a simple standalone server for production. The single-server mode is expressly intended for development and evaluation, and the key recovery and replication features depend on a distributed setup.

Pros and cons

Pros: S3 compatibility without application code changes preserves existing SDK, CLI, and application workflows.

Pros: Native SFTP stores uploads as S3 objects, bringing two access paths into the same storage system.

Pros: Distributed sharding, replication, configurable erasure coding, versioning, and lifecycle policies address placement and resilience needs across nodes.

Pros: Apache 2.0 licensing and deployment options from binary to Kubernetes give self-hosting teams flexibility without a software subscription.

Cons: Standalone deployment is for development and evaluation, not a production shortcut for teams needing versioning, object locking, or bucket replication.

Cons: Distributed deployment is a prerequisite for those features, adding operational and architectural work for teams that only want a single server.

Alternatives

Storage Replication Software is the broader category to explore when comparing replication tools rather than choosing a specific storage server.

SeaweedFS is another self-hosted option with a free Community support plan for any cluster size; choose it when that stated cluster-size allowance and community support model suit your needs.

Synology Photos is a free alternative that requires a Synology NAS, making it the relevant choice for readers specifically seeking that NAS-dependent photo product.

Resilio Active Everywhere is a paid alternative with a free trial and pricing based on devices or endpoints rather than data volume or transfer usage; consider it if that pricing basis fits better.

Amazon S3 is a paid, usage-based alternative with no minimum charge and a free plan; choose it if you prefer that billing model to self-hosting Obstor.

LINBIT DRBD offers a free open-source community version and a paid Starter plan at $1,000 a year; it is an alternative for readers comparing those DRBD options.

XtreemFS is a free, open-source BSD-licensed option for readers considering another self-hosted storage project.

Cristie CloneManager is a paid migration product with a $150, paid once migration plan; consider it when migration is the task at hand.

OpenEBS is a free, open-source Kubernetes storage option; choose it when Kubernetes storage is the specific need.

Verdict

Choose Obstor if you want a no-cost, Apache 2.0 S3-compatible server with SFTP and are prepared to operate distributed storage for production-grade data management. Its strongest reason to choose it is the combination of familiar S3 access, multiple-site replication, and configurable protection in a self-hosted system. Look elsewhere if you need a managed service or expect a standalone server to provide the full set of recovery and replication features.