Skip to main content

Run a Testnet Node

Nervos has introduced a Testnet for developers to experiment with integrations and Scripts in a safe environment, free from real-world risks. Due to fluctuations in the network's hash rate, the Testnet may occasionally reset.

  • At least 200 GB of available disk space
  • Stable internet connection
  • Syncing from genesis may take up to several days depending on your CPU, bandwidth, and disk performance

Step 1: Download the CKB Binary​

Download the latest version of CKB from the official GitHub Releases page. Make sure to download the correct version for your operating system.

info

This guide uses version ckb_v0.115.0 as an example.

Step 2: Navigate to the CKB Directory​

Unzip the downloaded archive into a location that's easy to find. We recommend:

  • MacOS: ~/Documents/ckb_v0.115.0/...
  • Windows: C:\ckb_v0.115.0\...

Open a terminal and navigate into the folder using the following command:

cd /Users/(NAME)/Documents/ckb_v0.115.0_aarch64-apple-darwin-portable/ckb

Remember to replace “(NAME)” with your actual name on the device.

Step 3: Verify Installed Versions​

Run the following commands:

./ckb --version
./ckb-cli --version

Notice for Unix-like systems, you need to explicitly use ./ to tell the shell to run ckb file in the current folder.

You should see the installed version printed in the terminal, like this:

ckb 0.115.0 (d845cbf 2024-04-01)
ckb-cli 1.8.0-rc1 (cf4159b 2024-03-28)
note

If you see a message like 'ckb' cannot be opened because the developer cannot be verified, this is macOS blocking unsigned binaries. To bypass:

  1. Go to System Settings → Privacy & Security
  2. Scroll down and click “Open Anyway”
  3. Rerun the command

Step 4: Initialize the Node for Mainnet​

Enter the following command to set up default configuration and data folders.

ckb init --chain testnet

If you're using a Unix-like system, run ./ckb init --chain tesnet instead.

Step 5: Start the Node​

Now that your node is initialized, you can start syncing to the Mainnet. Make sure you run the following command from the folder where you initialized the node in Step 4.

ckb run

If you're using a Unix-like system, run ./ckb run instead.

If your node is running successfully, you’ll see log messages similar to this:

...
2025-05-30 01:57:58.941 +00:00 verify_blocks INFO ckb_chain::verify  [verify block] new best block found: 9768557 => 0x1412c40a9ffcb18754e0e8349e383179ea47c4f2183a2e1e830a274ba37b5222, difficulty diff = 0x14f821cdc0fe600e, unverified_tip: 9768562
2025-05-30 01:57:58.941 +00:00 verify_blocks INFO ckb_chain::verify  block: 9768557, hash: 0x1412c40a9ffcb18754e0e8349e383179ea47c4f2183a2e1e830a274ba37b5222, epoch: 7507(1713/1800), total_diff: 0x54a33f4b40bf2ab7232c3, txs: 3
2025-05-30 01:57:58.942 +00:00 verify_blocks INFO ckb_chain::verify  [verify block] new best block found: 9768558 => 0x53fce8b632b3e4d011dd682aef73e018a885560a363599d5f00818a1ee3f3762, difficulty diff = 0x14f821cdc0fe600e, unverified_tip: 9768562
2025-05-30 01:57:58.943 +00:00 verify_blocks INFO ckb_chain::verify  block: 9768558, hash: 0x53fce8b632b3e4d011dd682aef73e018a885560a363599d5f00818a1ee3f3762, epoch: 7507(1714/1800), total_diff: 0x54a3409ac2dc06c7092d1, txs: 2
2025-05-30 01:57:58.943 +00:00 verify_blocks INFO ckb_chain::verify  [verify block] new best block found: 9768559 => 0x6b4a3c48a2035681f429cd206af3a6906cd72ce9c48b8a7f56eb2844ea8a56f1, difficulty diff = 0x14f821cdc0fe600e, unverified_tip: 9768562
2025-05-30 01:57:58.944 +00:00 verify_blocks INFO ckb_chain::verify  block: 9768559, hash: 0x6b4a3c48a2035681f429cd206af3a6906cd72ce9c48b8a7f56eb2844ea8a56f1, epoch: 7507(1715/1800), total_diff: 0x54a341ea44f8e2d6ef2df, txs: 3

These logs indicate that your node is downloading and verifying blocks in real time.

The first time you run a node, it will begin syncing the entire blockchain from the genesis block. This process can take up to several days, depending on your internet speed, CPU, and disk performance.

You don’t need to keep it running continuously — you can stop the process at any time using Ctrl + C. When you restart the node from the same folder, it will resume syncing from where it left off, as long as you haven't deleted or modified the data folder.

note

Many operating systems automatically enter low-power or sleep mode after a period of inactivity — even while a terminal process is running. This can silently pause your node, halting sync without an error message. Try disable automatic sleep/suspend in your system settings to ensure your node continues syncing properly.

Check Sync Progress​

To check whether your node has fully synced with the Mainnet, open a new terminal window and run:

ckb-cli rpc sync_state

If you're using a Unix-like system, run ./ckb-cli rpc sync_state instead.

Look for this line in the response:

"ibd": false

The "ibd" flag stands for initial block download. When it is false, your node is considered fully synced.

What Happens During IBD?​

While your node is still syncing (”ibd”: true), it behaves differently from a fully participating peer:

  • It only connects to one remote peer (not multiple peers) to download blocks.
  • It does not relay transactions or blocks to other peers.
  • It does not respond to most P2P requests (such as relay or identify).
  • It cannot help other nodes sync.

Once IBD is complete, your node becomes fully-synced.

Make Your Node Connectable​

By default, your Testnet node can sync from others, but it won’t accept inbound connections unless you configure it to.

Track Your Node’s Connectivity​

  1. Ensure your node is running
  2. Update the ckb.toml file in your ckb folder by following this instruction.
  3. Visit Node Probe, then switch to the CKB Testnet using the dropdown menu in the top-right corner.
  4. Search your node ID and you should be prompted that your node is connectable and visible on the network.

Node Discovery Criteria​

To be visible on Node Probe, your node must meet all of the following criteria:

  • Supports core protocols: The node must support the discovery, identify, relay, and sync protocols. If you follow the steps in this guide, this shouldn’t be a problem.
  • Reachability confirmed: The node must be reported as reachable by at least four other nodes in the network.
  • Well-synced and active: The node is well-synced (not necessarily fully synced) and is actively validating transactions and blocks.
  • Stable uptime: The node must remain online and meet the above conditions for at least 4 epochs (~16 hours).