diff --git a/README.md b/README.md index cfeae8b..d037e2e 100644 --- a/README.md +++ b/README.md @@ -118,10 +118,19 @@ ![api_examples](https://user-images.githubusercontent.com/3319450/31204191-3e197804-a930-11e7-871e-2c469480b906.gif) # Alerting -* Glass alerts check once a minute for subnet utilization and every 5 seconds for lease-rate alerts - * Shared Network Utilization % Thresholds (Warning & Critical) (Default 0 and 95) - * Leases per minute rate (Default 50) - * 0 values = Off +* **Realtime Alerting** One of our biggest high value returns on this project was creating something that can keep us on top of any sort of outage issues. + * "Is our server writing leases even though the process is alive?" + * "Is any our shared networks filling up?" +* **Shared Subnet Utilization Alerting** (Checked once a minute) + * Customize thresholds + * Default 80 (Warning) + * Default 95 (Critical) +* **Leases Per Minute** (Checked once every 5 seconds) - If your disk fills up - or your SAN is unavailable - alerting your team on absolutely zero activity on your production server can allow you to be on the 8-ball. This is a step-above process alerting because you can have the dhcp server process running and no leases being written. + * **Default**: 50 + * Glass keeps track of a rolling average of leases being written on the minute, the counter can be seen realtime on the dashboard +* 0 value in **Glass Settings** or in your **glass_config.json** will turn the alerting functionality **OFF** + +## Alert Delivery Methods * Glass currently supports the following alerting delivery methods ## E-Mail