Run checks against a Linux machine, export as Prometheus metrics
nix build
./result/bin/machine-check
# dry run - don't try to write to /run/metrics, just dump to stdout
./result/bin/machine-check dry-runDesigned to be run periodically via cron,
by default machine-check creates /run/metrics/machine-check.prom
containing results of all checks.
It is possible to configure certain checks like DNS check by providing a configuration file via environment variable:
MCCFG=./sample.config ./result/bin/machine-checkCheck out config.sample in this repository for a list of possible options.
# HELP zpool_status_errors pool errors
# TYPE zpool_status_errors gauge
zpool_status_errors{name="tank"} 0.0
# HELP zpool_status_health pool health
# TYPE zpool_status_health gauge
zpool_status_health{name="tank"} 0.0
# HELP zpool_status_parse_success parsing successful
# TYPE zpool_status_parse_success gauge
zpool_status_parse_success 0.0
# HELP zpool_status_status_reported pool reports status if 1
# TYPE zpool_status_status_reported gauge
zpool_status_status_reported{name="tank"} 1.0
# HELP zpool_status_success exitcode
# TYPE zpool_status_success gauge
zpool_status_success 0.0
# HELP zpool_list_success exitcode
# TYPE zpool_list_success gauge
zpool_list_success 1.0
Current checks:
- zpool list (capacity, health)
- zpool status (health, status, erros)
- birdc(6) show protocols
- dns
Future checks:
- ping
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent