These are the recommended specs for a ZetaChain node.
|Node Type||CPU||Memory||Data Disk|
|Observer Signer||8 CPUs||32 GB Memory||300+ GB Data Disk|
|Core Validator||4 CPUs||16 GB Memory||300+ GB Data Disk|
ZetaChain nodes communicate over the following ports. Make sure these are open to the public internet.
|Full Node and Validator||26656||TCP|
|Observer Signer Only||8123||TCP|
|Observer Signer Only||6668||TCP|
Additionally, Observer Signers must allow communication between
zetaclientd on port 9090 but this port should not be publicly accessible.
If you want to access RPC/API services on your node you'll need to make sure the following ports are open.
|EVM RPC - HTTP||8545||TCP|
|EVM RPC - WSS||8546||TCP|
ZetaChain nodes have been developed and tested on x86_64 architecture. Our binary files have been compiled with Ubuntu 22.04 LTS x86_64. This guide assumes you are using Ubuntu 22.04 LTS x86_64. If you are using a different OS you may need to make some adjustments.
Set Limits on Open Files and Number of Processes
To better manage the resources of your nodes, we recommend setting some limits on the maximum number of open file descriptors (nofile) and maximum number of processes (nproc).
/etc/security/limits.conf to include or modify the following parameters:
* soft nproc 262144
* hard nproc 262144
* soft nofile 262144
* hard nofile 262144
/etc/sysctl.conf to include the following:
In a production environment we recommend monitoring the node resources (CPU load, Memory Usage, Disk usage and Disk IO) for any performance degradation.
Prometheus can optionally be enabled to serve metrics which can be consumed by Prometheus collector(s). Telemetry include Prometheus metrics can be enabled in the app.toml file. See the CosmosSDK Telemetry Documentation for more information.
See more about your Validator Monitoring here.