OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Dakera Deploy

Dakera Deploy is a public infrastructure repository for deploying Dakera’s AI-agent memory server with Docker Compose, Kubernetes manifests or Helm. Its resources cover local and persistent single-node setups, three-node high availability, and monitoring.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Dakera Deploy — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Dakera Deploy when you need deployment resources for the Dakera memory server, whether for a local instance, persistent single-node setup or clustered deployment.

Who it is for

  • Teams setting up self-hosted persistent memory for AI agents.
  • Developers who need a local Dakera instance for development.
  • Operators deploying Dakera with Compose, Kubernetes or Helm.

Documented uses

  • Start a local in-memory server for development.
  • Configure a persistent single-node deployment using MinIO.
  • Set up a three-node cluster with load balancing and shared cluster configuration.
  • Choose deployment resources for Kubernetes or Helm.
Before you choose

Requirements & limits

The repository is publicly accessible and identifies itself as open source. The supplied README shows an MIT license badge.

  • The local Compose profile uses in-memory storage, while separate profiles provide MinIO-backed persistence.
  • The HA Compose profile requires a cluster secret of at least 16 characters that is the same on every node.
  • Telemetry is enabled by default; the deployment documentation describes opt-out settings and the information transmitted.
  • The repository includes Kubernetes manifests and points to a separate Helm chart.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Start a local in-memory instance with Docker Compose

Bring up the documented local single-instance Dakera server for development or testing.

Before you start

  • Docker
  • A local checkout of the dakera-deploy repository

Steps

  1. Open the repository’s docker directory.

  2. Run the command below.

    Documented command
    docker compose -f docker-compose.local.yml up -d

  3. Use the REST API at http://localhost:3000 or the gRPC API at localhost:50051; the health endpoint is http://localhost:3000/health.

Expected outcome

A local Dakera server using in-memory storage, with REST, gRPC and health endpoints.

Guide limitations

  • The local profile uses in-memory storage; the documentation recommends a MinIO-backed or default profile for data persistence across restarts.
Documentation 1 ↗
Source-based practical guide

Configure and start a three-node Compose cluster

Prepare and start the repository’s documented high-availability Dakera cluster.

Before you start

  • Docker
  • A local checkout of the dakera-deploy repository
  • A cluster secret of at least 16 characters, set identically for every node in the environment file used to start the stack

Steps

  1. Open the repository’s docker directory.

  2. Set DAKERA_CLUSTER_SECRET in the environment file used for the stack, using the same value on every node.

  3. Start the cluster with the command below.

    Documented command
    docker compose -f docker-compose.ha.yml up -d

  4. Use the load-balanced REST endpoint at http://localhost:3100, gRPC at localhost:50151, or cluster status at http://localhost:3100/admin/cluster/status.

Expected outcome

The documented three-node setup with Traefik load balancing, gossip-based clustering and MinIO storage.

Guide limitations

  • The cluster secret is mandatory; the repository states that a v0.12.1 cluster node without it exits.
  • This recipe covers the supplied Compose cluster profile, not Kubernetes or Helm setup.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
18 October 2026
Original website title
Dakera-AI/dakera-deploy · Official homepage title · retained in its source language
How we review sourcesSuggest a correction