Monitoring

[ppolewicz]

• there is a different acceptable vtrust range for every subnet
• monitoring of weight setting attempts of a validator
• memory usage
• swap usage
• disk space
• CPU/loadavg usage
• GPU usage
• VRAM usage
• open handles
• subnet-specific log events

[wjurkowlaniec]

hey, I created a minimal, configurable script for doing monitoring

https://gist.github.com/wjurkowlaniec/9fca4d8f1d0d74c81bc7e1c2ba1e982e

please check it out, I can work on this further.

It defaults to send data to InfluxDB, but we can do any other destination.

[wjurkowlaniec]

also I can provide some basic local grafana + influx docker images + pm2 config for running the script

[ppolewicz]

Notes from meeting on 2024-08-14:

• monitoring can make use of Rhef's internal tools which would need to be open sourced