# Incidents & Notifications > Loadster monitors open and close incidents whenever problems are detected, and notify you according to your notification policies. Source: https://loadster.com/manual/incidents-and-notifications/ When your Loadster monitor detects a problem it creates an **incident** that you and your team can track until the problem is resolved. Loadster can notify you when the incident is opened and resolved through your choice of channels. Available channels include email, robocalls, text messages, and integrations with your other apps and services. ## Incident Escalation & Resolution Loadster opens an incident whenever a monitor that was previously passing starts failing. It closes the incident when the monitor again starts passing. By default, a single failing cycle is enough to open an incident, and a single passing cycle is enough to resolve it. If this isn't appropriate for your situation, so you can change these thresholds with the escalation settings at the top of each monitor. _Figure: Configuring a monitor's escalation behavior_ If you only want an incident after multiple failed cycles, you can change the monitor's escalation threshold to only create an incident after 2, 3, or more consecutive failures. Similarly, you can change the resolution threshold to require multiple consecutive passing cycles before closing an incident. ## Notification Policies You can use **notification policies** to change how Loadster notifies you about monitoring incidents. Each monitor can have its own notification policies. ### Standard Notification Policies The standard notification policies are "Email & Integrations" and "None". The "Email & Integrations" standard policy simply emails everyone on your Loadster team when an incident is opened and again when it's resolved, and also calls any monitoring integrations that you have connected. The policy called "None", as you might have guessed, doesn't notify at all. The monitor will still track incidents and you'll be able to view these incidents on the dashboard, but you won't be notified about them. ### Custom Notification Policies In addition to the standard policies, you can create custom notification policies for better control over how and when you are notified of incidents. To create a custom notification policy, expand the *Notification Policies* drawer and add a new policy. ### Setting up notification policy actions A notification policy's *actions* are the actions it takes when an incident is opened and closed. You can add multiple actions to the same notification policy, and all actions will be triggered. Actions include: * Send an email * Send an SMS (text message) * Make a robocall (voice call) * Call an integration Keep in mind that text message and robocall notifications will only be made to users on your team who have entered and confirmed a phone number on the Account page. Everyone who wants to receive these has to enter their own phone number and opt in individually. There's no way to add phone numbers for your other team members because the phone carriers don't want to enable spam. Integrations can be configured on the Integrations page by any admin user on your team. Once an integration has been enabled you'll be able to use it in your notification policies. ### Customizing the notification policy schedule A notification policy also has a schedule of when to notify about incidents. It can notify immediately when an incident is opened, or only after it has been open for a set number of minutes. Sometimes monitors recover on their own, so you might not want to get notified of minor incidents unless the condition persists for a while. You can configure a notification policy to repeat the notifications while an incident remains open. If you enable multiple notifications, they'll repeat at increasing intervals with a sort of exponential back-off timing function. ### Maintenance windows **Maintenance windows** define scheduled time periods when you expect your site to be unavailable or degraded, such as during a weekly deployment or database migration. You can configure maintenance windows in the Monitoring tab of Settings. Each maintenance window can be linked to monitors, notification policies, or both. **Pausing monitors** — When you link a maintenance window to a monitor, the monitor won't run any cycles during that window. Any open incidents remain open until the monitor resumes after the window ends, and no new incidents will be created by that monitor since it won't be running. This is useful when you don't want planned downtime to affect your uptime stats. **Muting notifications** — When you link a maintenance window to a notification policy, monitors can still run normally but no alerts will be sent for incidents that occur during the window. This is useful when you want to continue tracking monitor results but don't want to be woken up or alerted during expected downtime. You can also associate maintenance windows from the monitor configuration page or from a notification policy's settings. ### Associating notification policies with your monitors After you've created a notification policy, don't forget to associate it with your monitors! You can do this by checking the checkbox in the list of notification policies in each monitor's escalation settings. It's possible to have multiple notification policies associated with the same monitor, although your life might be simpler if you keep it to just one at a time. Similarly, you can use the same custom notification policy with multiple monitors, and they will share the same notification actions and schedule. ## Incident History You can view open and resolved incidents on the Incidents page. Loadster keeps a detailed history of when and why each incident was opened and resolved. From the incident, you can jump to the failed monitor cycle that triggered the incident or the passed monitor cycle that resolved it. You can also view a timeline of notifications about that incident, so you can see who on your team has been notified and how. Similar to the main monitoring dashboard, you can filter your incidents by monitor, status, or tag.