Cloud Computing 6 min read

How Tencent Cloud Elasticsearch Enables Multi‑AZ Disaster Recovery

Tencent Cloud Elasticsearch now supports cross‑availability‑zone deployment, requiring even‑numbered data nodes, dedicated master nodes, and replica settings to ensure continuous service when a zone fails, with detailed steps for quick setup and region limitations explained.

Tencent Cloud Developer
Tencent Cloud Developer
Tencent Cloud Developer
How Tencent Cloud Elasticsearch Enables Multi‑AZ Disaster Recovery

Tencent Cloud Elasticsearch has introduced a cross‑availability‑zone (AZ) disaster‑recovery feature, allowing clusters to be deployed across two AZs. This ensures that if one AZ experiences power or network failures, the other AZ continues to provide stable, uninterrupted service, enhancing business reliability.

Quick Experience

To try the feature, go to the Tencent Cloud Elasticsearch Service purchase page, select Deployment Mode > Multi‑AZ , choose a VPC, and configure two AZs (currently only two AZs are supported). Apart from requiring an even number of nodes to distribute evenly across the AZs, the other settings are the same as single‑AZ deployment.

Cluster Multi‑AZ Disaster‑Recovery Principle

Data nodes are automatically and evenly distributed across the selected AZs, and each node is AZ‑aware. Index shards must have at least one replica, ensuring that each replica resides in a different AZ. The service provides an internal VIP that load‑balances requests across all data nodes. The VIP includes health‑check logic: if a node fails to respond after multiple checks, it is temporarily removed from the VIP pool until it recovers, preventing client requests from hitting unhealthy nodes.

Data Node Requirements

Number of data nodes must be a multiple of the AZ count (e.g., for two AZs: 2, 4, 6, 8 …).

Index shards must have at least one replica, guaranteeing at least two copies of the data.

Dedicated Master Nodes

To improve cluster reliability, a minimum of three dedicated master nodes is required. These master nodes are distributed across three different AZs, even if data nodes are only in two AZs. This design ensures that, when one AZ becomes unavailable, the cluster still retains a majority of master nodes for election, maintaining normal operation.

Region Availability and Limitations

The multi‑AZ disaster‑recovery feature is currently available only in large regions such as Beijing, Shanghai, and Guangzhou, which support at least three AZs. Other regions will be added as Tencent Cloud expands its data‑center infrastructure.

Note: The feature is not yet open to all customers. If you need multi‑AZ capability, please contact your pre‑sales architect or sales manager.

Original Source

Signed-in readers can open the original source through BestHub's protected redirect.

Sign in to view source
Republication Notice

This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactadmin@besthub.devand we will review it promptly.

cloud computingElasticsearchhigh availabilitydisaster recoveryMulti‑AZTencent Cloud
Tencent Cloud Developer
Written by

Tencent Cloud Developer

Official Tencent Cloud community account that brings together developers, shares practical tech insights, and fosters an influential tech exchange community.

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.