v7.1.0 Release Notes

Mach5 Search v7.1.0 is the latest production release. It adds native OpenSearch compatibility, system diagnostics and warehouse management, autoscaling and placement controls, broader query behavior, stronger authorization, and more durable operational workflows.

Highlights

• Native OpenSearch compatibility. The legacy Java OpenSearch deployment is removed from the standard v7.1 deployment. /opensearch is served by MDX’s OpenSearch-compatible API, with broader administrative, search, update, analysis, statistics, and template API coverage.
• System warehouse and diagnostics by default. Helm enables the system warehouse and diagnostics receiver by default and bootstraps the warehouse in the fixed _system namespace.
• Autoscaling and shared warehouses. Shared capacity, namespace-level warehouse attachment, resource and admission priorities, worker placement, and autoscaling provide more control over workload distribution and infrastructure cost.
• Expanded KQL and query behavior. KQL gains broader Azure Data Explorer-style semantics for joins, nested data, dynamic values, dates, IP addresses, regular expressions, aggregations, and materialized views. SQL and search correctness improves across joins, aggregations, scoring, highlighting, and resource handling.
• Stronger authorization and secret handling. Authorization is applied more consistently across APIs and tools, while cloud credentials in dynamically created workloads use secret references rather than direct exposure.
• More durable operations. Data-asset lifecycle state, compaction, ingestion admission control, backpressure, auditing, and diagnostics are improved for retries, failures, recovery, and resource pressure.

Upgrade Guide

Review the v7.1.0 migration guide before upgrading from v6.2.0. It includes the OpenSearch endpoint change, system warehouse values, direct-service authorization, multi-field projection behavior, Helm upgrade procedure, and post-upgrade validation steps.

OpenSearch Compatibility

The legacy Java OpenSearch deployment (opensearch-legacy) is removed in v7.1. The /opensearch endpoint is served by MDX’s OpenSearch-compatible API.

Integrations that depend on Java OpenSearch plugins or implementation-specific behavior should be tested. Search, bulk, scrolling, aliases, aggregations, analyzers, highlighting, dashboards, and saved searches may differ where the native compatibility layer and Java OpenSearch previously differed. Unsupported operations now return clearer client errors.

System Warehouse and Diagnostics

The system warehouse and diagnostics receiver are enabled by default. Helm creates or reconciles the system warehouse in _system using the configured systemWarehouse.config.

systemWarehouse:
  enabled: true
  name: default
  config:
    resource:
      num_mediators: 1
    enabled: true
    memory_policy: legacy
diagnosticsreceiver:
  enabled: true

Extend systemWarehouse.config with the sizing, storage, placement, or engine settings required by the environment. To disable Helm management, set both systemWarehouse.enabled and diagnosticsreceiver.enabled to false. See the migration guide for the complete configuration and upgrade implications.

Warehouses, Querying, and Indexing

v7.1 adds autoscaling, shared warehouses, namespace-level warehouse attachment, placement controls, and explicit resource and admission priorities. Existing warehouses are adopted where possible during startup rather than recreated.

KQL support expands with broader Azure Data Explorer-style semantics for joins, nested and dynamic data, date and IP functions, regular expressions, aggregations, and materialized views. SQL and search behavior improves for joins, aggregations, scalar strings, date and time handling, scoring, highlighting, and large-result/resource-exhaustion handling.

Mapping limits are enforced more consistently, and partition-aware indexing and query execution are expanded. Review workloads using large mappings, long field names, strict mappings, nested or multi-valued fields, custom routing, or partitioning.

Data Assets, Packages, and Ingestion

Data-asset mapping and package installation preserve more lifecycle state across retries, restarts, updates, failures, and recovery. Existing Kafka formats, including the Protobuf support documented for v6.2.0, remain available after upgrading.

Security and Operations

Authorization is stricter and more comprehensive across MCP/agentic tools, administrative OpenSearch operations, indexes, warehouses, data assets, dashboards, workflows, notebooks, scripts, and connections. Cloud credentials in dynamically created workloads are protected through secret references.

Compaction and ingestion provide stronger admission control, auditing, and diagnostics. Under resource pressure, writes may be delayed or rejected with a backpressure response rather than allowing system instability to grow.

Local and Single-Node Deployments

The local and single-node deployment model is more closely aligned with Kubernetes deployment behavior, while still differing in profiling, storage backends, external services, and controller-managed operations. The local runtime has a smaller default storage-backend footprint than the distributed deployment.

Supported Documentation Versions

Use the following stable documentation entry points:

• 7.1.0 — latest
• 6.2.0 — supported
• 6.1.0 — supported

Useful entry points include 7.1.0 documentation, Kafka ingestion, Mach5 One, and current documentation.