> ## Documentation Index
> Fetch the complete documentation index at: https://docs.chainstack.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Chainstack Self-Hosted v2.21.0: August 4, 2026

> Avalanche, Arc, and World Chain are now available as deployment targets, and the bundled Grafana dashboards render correctly again.

This release adds three protocols as deployment targets — Avalanche, Arc, and World Chain — and fixes the bundled Kubernetes Grafana dashboards.

## Avalanche support

Avalanche Mainnet and Fuji Testnet are now available as deployment targets, running a single AvalancheGo node as a non-validator in full mode.

* Avalanche Mainnet — chain ID 43114, explorer at [snowscan.xyz](https://snowscan.xyz)
* Avalanche Fuji — chain ID 43113, explorer at [testnet.snowscan.xyz](https://testnet.snowscan.xyz)
* Endpoints — one node serves the whole Primary Network, so the node overview lists four endpoints that share a single port and differ only by path: C-Chain HTTP and WebSocket JSON-RPC, X-Chain HTTP, and P-Chain HTTP

Sync lag and peer count are reported for Avalanche nodes the same way as for any other EVM node. Mainnet restores from a snapshot, so provision about 400 GiB of storage; Fuji syncs from genesis over the network's public peers and needs about 300 GiB.

## Arc support

Arc Testnet is now available as a deployment target. Each deployment runs two components — an execution client and a consensus client — as a non-validator node in full mode, following the network's public endpoints.

* Arc Testnet — chain ID 5042002, explorer at [testnet.arcscan.app](https://testnet.arcscan.app)
* Endpoints — HTTP and WebSocket JSON-RPC, with the `eth`, `net`, `web3`, `txpool`, `trace`, and `debug` namespaces enabled

The deployment restores from a snapshot. Provision about 314 GiB of storage across the two components.

## World Chain support

World Chain Mainnet and World Chain Sepolia are now available as OP Stack deployment targets, running a World Chain execution client alongside OP-Node as the rollup client.

* World Chain Mainnet — chain ID 480, explorer at [worldscan.org](https://worldscan.org)
* World Chain Sepolia — chain ID 4801, explorer at [worldchain-sepolia.explorer.alchemy.com](https://worldchain-sepolia.explorer.alchemy.com)
* Endpoints — HTTP and WebSocket JSON-RPC on the execution client, plus the rollup HTTP RPC on OP-Node

Like the other OP Stack deployments, both networks require an Ethereum Layer 1 RPC endpoint and a Beacon API endpoint that you supply. Both restore from a snapshot — provision about 1300 GiB of storage for Mainnet and 100 GiB for Sepolia.

See [System requirements](/docs/self-hosted/requirements) for the full specification and [Supported clients and protocols](/docs/self-hosted/supported-clients-and-protocols) for the deployment catalog.

## Monitoring dashboards

The bundled Kubernetes node and pod Grafana dashboards render correctly again, and kube-state-metrics is now scraped with the labels those dashboards expect. Panels that previously came up empty are populated. See [Monitoring](/docs/self-hosted/monitoring) for how to reach the dashboards.

## Component versions

| Component                | Version |
| ------------------------ | ------- |
| cp-ui                    | v3.10.1 |
| cp-deployments-api       | v1.1.0  |
| cp-workflows             | v3.23.0 |
| cp-auth                  | v0.5.1  |
| cp-bolt                  | v1.12.0 |
| blockchain-node-exporter | v1.4.0  |