Skip to main content

Set up a Collator Node

This is a step-by-step how-to so you can get your Hydration collator up and running. In this guide, we use Ubuntu 20.04 LTS.

Required technical specifications

The following technical specifications are required as a minimum for running a Hydration collator node:

  • OS: Ubuntu 20.04
  • CPU: Intel Core i7-7700K @ 4.5Ghz (or equivalent single core performance)
  • Memory: 64GB RAM
  • Storage: NVMe SSD with a capacity of at least 200GB (the data footprint will grow over time)

These are the minimum technical requirements which have been verified by the team. Want to make sure that your machine has sufficient resources to run a node? Run a performance benchmark to find out.

Create a technical hydra user and add it to Sudoers

sudo adduser hydra
sudo usermod -aG sudo hydra
su - hydra

Download and configure the hydradx binary

Pick a 12.x release, we are using v12.1.0 from our assets repository:

sudo mv hydradx /usr/local/bin
sudo chmod +x /usr/local/bin/hydradx
sudo chown hydra:hydra /usr/local/bin/hydradx

Command to run your collator

Best is to run your collator as a service using systemctl. To do so, create a file, namely hydradx-collator.service under /etc/systemd/system/hydradx-collator.service:

sudo vim /etc/systemd/system/hydradx-collator.service

Then paste the following:

Description=hydradx validator
ExecStart=/usr/local/bin/hydradx \
--prometheus-external \
--base-path /var/lib/hydradx \
--collator \
-- \
--execution wasm \
--telemetry-url "wss:// 0" \
--base-path /var/lib/hydradx


Before starting your node, let's create the base-path folder and give it the necessary permissions and ownership:

mkdir /var/lib/hydradx
chown hydra:hydra /var/lib/hydradx

Make sure you have enough volume for your base-path by using df -h command.

Note that --prometheus-external is optional, but we highly recommend it so you can be able to export prometheus metrics and monitor your node's health through Grafana. For more details about monitoring, please visit this link.

If you need to monitor both the parachain and relaychain metrics, --prometheus-externaloption should be setup in both parts. You also need to set a separate port for the relaychain part as follows: --prometheus-port YOUR_CUSTOM_PORT_NUMBER

Depending on your setup, you might also want to override certain parameters like the websocket, rpc or your node p2p port. Please use hydradx --help for more information about the available options.

After saving your file, run the following commands to start your node:

sudo systemctl enable hydradx-collator
sudo systemctl start hydradx-collator.service

Your node should now be up and running. Make sure your hydra user has the necessary permissions to access your base-path and key file.

If you need to troubleshoot your running service, you can use the journalctl command with the -f option for tailing:

journalctl -fu hydradx-collator

Generate your session key

In order to generate keys for your node, run the following command:

curl -H "Content-Type: application/json" -d '{"id":1, "jsonrpc":"2.0", "method": "author_rotateKeys", "params":[]}' http://localhost:9933

Once done, you will have an output similar to:


Set your session key

To associate the generated session keys with your Controller account, navigate to the following menu item in the Polkadot/apps on the Polkadot parachain Hydration: Developer > Extrinsics.

Fill in the fields:

  • using selected account: select your Controller account;
  • submit the following extrinsic: select session on the left side and setKeys on the right;
  • keys: enter your session key you just generated;
  • proof: 0;

What's next?

Make sure that your node is fully synced. Once this is done, let us know in the dedicated Discord channel (only if you have been preselected as a collator).