---
title: Managing Error Notifications
slug: zen-master/managing-error-notifications
docTags: 
createdAt: 2024-12-27T19:52:41.680Z
---

ZEN Master emails a user about an object only when all three of these are true:

- the user holds a role, directly or through a user group;
- that role's tag is on the object;
- that role has **Notify** turned on for the object's type (Resources, Sources, Channels, Targets or Reports).

Read and Write access do not send email on their own. Admin rights do not either. A user who can see and change an object gets no email about it unless one of their roles also has Notify.

**A new account sends no alert email until you turn on Notify.** The Default Admin role is created with Read and Write only. Edit the role, or add a new one, and turn on Notify for each object type you want email about. See [Adding Roles](docId:268Rk_Tt-vJbIwFUYl71S).

There is no recipient list and no mail server setting. Alerts go to the account email address of each user who qualifies. Incident emails are separate: they go to every user with Admin or Incident Manager rights on the account.

**Alert times are not fault times.** Each rule's escalation delay holds an alert back before it is raised. Its de-escalation delay, 5 minutes by default, keeps it open after the fault clears. So the time between an alert and its clear is not the length of the fault.

There are four methods for limiting the notifications that are sent out.

- You can *Mute* a specific object, stopping **all **notifications related to that object for **all **users.
- You can adjust the *Event Profile* that applies to the object, see [Configuring Events Profiles](docId\:CWshWpVMDch_ryL86IMTS). This enables you to adjust each rule in the profile individually, specifying the following settings:
  - **Enabled/Disabled** - Specify whether or not this rule is enabled for generating Events.
  - **Thresholds** - For rules that are governed by thresholds (e.g. CPU usage, low bitrate), specify the thresholds for generating warning and error events.
  - **Escalation** - Specify the number of events that cause escalation from "warning" to "error" Events and the window of time that is considered for escalation.
  - **Notifications** - Specify which Events trigger notifications.
- *Flapping Detection* automatically mutes an object temporarily when it repeatedly goes into an error state. You can configure the settings for when this takes effect.
- You can adjust the *thresholds* for certain Error and Warning notifications that relate to excessive CPU/GPU usage. This is done within the configuration for each specific Broadcaster Cluster, see [Adding a Broadcaster Cluster - Managed](docId:6sLx4TD__ZKr2XFhO8qZ6).






- [Muting Objects](docId:60Rcyl-FLgDG5tgWmIASl) 
- [Flapping Detection](docId\:Sha6sanwt1OKG_Z9VdxdD)