๐Ÿค–Automatic Upgrades

Cosmovisor is a process manager for Cosmos SDK applications, providing a mechanism for automatic node upgrades. It monitors the governance process for signals indicating that a software upgrade is required and automatically downloads and applies the necessary updates. This tool helps to facilitate smoother and more efficient network upgrades, thereby ensuring a high level of continuity and reliability of blockchain services.

1. Install Titand

Install titand

2. Install Cosmovisor

circle-info

To install Cosmovisor you also need to install Go (guide)

Follow the document of cosmosarrow-up-right to install Cosmovisor

Ensure add go bin /usr/local/go/bin and your GOPATH to PATH environment variable. You will also want to add below path into $HOME/.profile to persistent.

# add go binary
export PATH=$PATH:/usr/local/go/bin
# add go package binary
export PATH=$PATH:$(go env GOPATH)/bin

3. Init and config Node

Follow Joining Testnet until step Create service file.

4. Config Cosmovisor

  1. Add environment variable to ~/.profile

Correct DAEMON_HOME if you do not use default home location for node

  1. Reload your .profile

  1. Create folders for Cosmovisor

  1. Init Cosmvisor with titand binary

  • If you install titand from source. Run follow command to init your genesis binary

  • If you download prebuild binary titand from our github. Copy prebuild binary and lib into cosmovior

After that your .titand folder will be like this (will not contain lib folder if you install from source)

Verify your setup with command (ensure version is same as your titand)

  1. Update service file /etc/systemd/system/titand.service to use cosmovisor

Replace <cosmovisor_absoule_path> by output of command which cosmovisor.

Replace <home_directory> by your home directory that contain folder .titand . Or other directory if you chose your own folder for titan node.

circle-info

By default, Cosmovisor will back up the current data before upgrading the chain process (this means your disk space should always have more than 40% free). If your node is configured to keep only a small amount of historical state, the backup process may be fast. However, if your node is a full node that stores a large amount of historical data, you may want to set UNSAFE_SKIP_BACKUP=true to help speed up the upgrade process and avoid consuming disk space.

  1. Start service

  1. Check node status

You should get something like this:

If have any issue, you can check error log by

Last updated