Monitoring

Once everything is configured, you should now be able to access PWA by accessing your host on your browser. You should be prompted to the login page. You should signup / confirm your email address, then define host groups / testspecs, and construct new meshconfig using those test entries.

PWA reports the current health status via following API endpoint (for pwa-admin and pwa-cache)

https://<hostname>/pwa/api/pwa/health

{
    status: "ok",
    msg: "everything looks good",
    cache: {
        hosts: 255,
        update_time: 1486994021924
    }
}

You can configure your monitoring systems (Sensu, Nagios, etc..) to check for status and make sure it’s set to ‘ok’.

For pwa-pub instances, you should run separate test at http://<hostname>/pwa/pub/health (not https://)

{
    status: "ok"
}

Please note.. if you are running multiple instances of pwa-pub, then /pub/health is just from one of the instances (not all)

You should also monitor the authentication service status

https://<hostname>/pwa/api/auth/health

{
    status: "ok",
    headers: {...}
}

You can also monitor docker stdout/stderr log - similar to syslog:

sudo docker exec -it pwa-admin1 pm2 logs