brought to you by True Staking
Just basic Linux server monitoring. Sometimes, the simplest solution is the best solution. For all the trends, graphs, metrics, statistics you might need, there is Prometheus. For knowing your validator has no issues, there is VVCM.
Server checks include:
- node delinquent alert
- validator service status
- out of memory error condition
- loss of network connectivity
- disk space
- nvme heat, lifespan, and selftest
- cpu load average
Running the installer (setup.sh) generates an api key unique to your server, and creates a service (vvcm.service) triggered every 2 minutes by vvcm.timer. You select what you want to monitor, and the checks are run every 2 minutes. At the beginning of each check series, an "I'm alive" message is sent (https via curl) to our backend server. In the (hopefully) rare event that an alert is generated, an additional outbound https call via curl to monitor.truestaking.com submits the alert, and the backend forwards the alert to you via telegram or email.
Note: If our backend server doesn't receive an "I'm alive" message from your validator within 5 minutes, then it sends the "Is Alive Error" alert.
Each server should have a unqiue api key. Run the installer (setup.sh) on each server you want to have monitoring on.
Monitoring preferences can be update and monitoring can be paused using update_monitor.sh. A server account can be deleted and all monitoring stopped using delete_account.sh.
Feedback is welcome, and we hope this benefits the Velas Validator Community!
To install, run:
curl https://raw.githubusercontent.com/truestaking/vvcm/main/setup.sh -O
chmod +x ./setup.sh
./setup.sh
The install will guide your through the setup with a series of y/n questions, and accept your validator address (node delinquency check), email address (notifications) and telegram id (notifications). When finished, you will find all the files in /opt/velas/vvcm . The ENV file reflects your configuration choices -- please do not edit that file directly. Instead, use update_monitor.sh to change your monitoring configuration.
Installing creates the folder structure:
/opt/velas/vvcm # Top level folder
...
- env # Environment variables
- monitor.sh # Monitor script ran every 2 minutes by vvcm.service, triggered by vvcm.timer
- update_monitor.sh # Update script used to start/stop monitoring and change local and remote variables
- delete_account.sh # Remove a server account (requires server API key)
Tested on Ubuntu 20.04
Please don't hesitate to reach out with questions, concerns, or suggestions!