MultiversX launches One-Click Node service in collaboration with Tencent Cloud

MultiversX launches One-Click Node service in collaboration with Tencent Cloud

8 April 2024

Earlier this year, MultiversX and Tencent Cloud entered into a strategic collaboration which includes supporting both internal and community services.

Today we are excited to present one of our co-developed products, the One-Click Node service for Observer node types.

The Observing squad is a service for all MultiversX shards which participates in broadcasting transactions or querying the blockchain data. All 4 interconnected shard observer processes will be deployed on the same virtual instance.

Our solution focuses on bringing these features in a compact and cost-effective format, hosting all 4 chains (metachain + 3 shards) plus the proxy service into one single virtual server which is right sized for efficiency.

We have developed this solution with the below requirements in mind:

  • Automated deployment

  • Easy operation

  • Integrated node commands

  • Support for multiple modes: lite and Full-Archive-db-lookup

  • Support for multiple networks: mainnet, testnet, devnet

  • Integrated security

With these goals in scope we developed this solution using tools which power-up automation such as Hashicorp’s Terraform coupled with Server bootstrapping best practices

We have chosen Terraform also because the user-end solution deployment is a mere parameter declaration in a plain text file with an easy-to-understand format.

Below you can see the reference architecture of the solution:

Here are some of the features of the One–Click Node solution:

  • Terraform based module which only requires parameter declaration and basic deployment steps (init -> plan -> apply)

  • The solution uses Tencent Cloud Lighthouse servers, which is a resource bundle, designed for maximum performance/cost ratio

  • Integrated commands for start, stop, restart or upgrade the node

  • Complete operation dashboard which includes the service status, Stateful firewall, or monitoring features

  • For the archive mode, we enable the addition of 3 additional disks, which will contain the additional data. The provisioning of these disks is fully automated and due to local snapshots the retrieval is being done over 10GBPS links, which ensures a lightning fast archive provisioning

  • We have also embedded the CIS Server Level 2 OS hardening commands for increased security

  • Hassle free deployment just by setting a couple of parameters which are essential for the deployment: region, name, type, purchase period or extra firewall rules

  • Code available publicly in github which enables the MultiversX community to participate and enhance

The deployment requires some prerequisite steps to be taken. These steps are described in the official README but here is a summary of that:

  1. Create a Tencent Cloud Account and go through the steps of setting it up

  2. Environment configuration described in detail here

Once the prerequisites have been completed the deployment of the solution is an easy process which includes

  1. Creating a new folder and a text file in it. Use one of the samples to fill in the parameters

  2. Solution stack deployment (3 commands only), described here

We have only exposed the parameters which are essential to the deployment in order to simplify the deployment approach. However, if you need to further customize the deployment, feel free to check all the available parameters or add custom ones. These tasks should be undertaken by advanced users.

When configuring the parameters for the deployment don’t forget to set the desired chain on which the node will be spawned. For testnet and devnet you can choose a smaller configuration in detriment of performance, if your goal is simply to test basic features. Of course we recommend going for the system requirements as advertised by the MultiversX team.

Once the solution is deployed there will be some scripts running in the background which bootstrap (deploy) the entire Observer solution. Depending on the type of node chosen these can take from 5 minutes to more than an hour. Make sure that you check the status in the “Run commands” panel of your Lighthouse instance, as depicted below.

Do not use the node until the execution of the commands is completed successfully.

Operating the node is easy through the dashboard, The possible operations are also depicted in the image above and they enable controlling the shard services, attached disks, firewall rules, ssh keys, snapshots, etc

We recommend that you perform all the operations on deployment time (e.g adding extra firewall rules) because any configuration change in the dashboard will be removed if you choose to redeploy the solution.

In the monitoring tab you will find metrics related to the server’s resources.

After node deployment it is normal to see some of the resources being fully utilized as the node is still synchronizing with the network. Depending on network speed and quality of peers this could take up to a couple of hours.

Because Terraform keeps a state file with the resources, the “source of truth” will be this terraform solution and “the result” will be seen in the dashboard. Of course we don’t discourage the fiddling with the instance from the console or directly on the OS, but be aware that the solution might break if handled improperly.

This solution will be constantly improved and because it is an observer stack, the node keys are unimportant, therefore the solution can be redeployed at will whenever a new update is needed or any configuration change has taken place.

Some of the features that we would like to add in the future are:

  • Email notifications for important events like provisioning completion

  • Notifications when network sync is completed

  • Sovereign shard support

We hope that you find this article useful and are happy to help you with any issue.

The One-Click-Node service will be managed by our trusted partner Materia Logic which handles all incoming onboarding and support requests.

To get started please check our landing page:

Cloud Integration Partner:

Cloud Integration Partner: