PRTG Manual: Setting up Notifications Based on Libraries: Example
This section describes how you can reorganize your devices or sensors by their function, and thereby create a new customizable view. This way, you can monitor entire sections of your network with a single set of notifications.
You can personalize these views and their notifications. Furthermore, you can grant and restrict permission to this view to other users or user groups to give them direct access to some of the information in PRTG.
Create a new library:
- Hover over Libraries in the main menu bar and select Add Library.
- Give the library a meaningful name and assign it a security context. Possibly add some tags for organization purposes.
- Optionally, assign access rights to your predefined user groups.
- Click Create to create the new library.
- The Management tab of the library opens.
Add objects to your library:
- Click Add Library Node.
- Define the Library Node Display Settings:
- Choose an object from which the sensors are taken (predefined: Root).
- Select Show a collection of filtered sensors in the library (max. 1000) to filter (the subtree) further.
Choose from the following filter criteria:
- Show specific sensor types only: Add check marks to sensor types that you want to include.
- Show sensors with a specific sensor status only: Add check marks to the sensor states that you want to include.
- Show objects with specific tags only: Enter tags to include certain sensors.
- Show objects with a specific priority only: Priority "1" through "5" is possible.
The libraries are dynamic and not static. When you add or remove sensors to your device tree that are of a type already specified here, the library automatically updates the displayed sensors. The same is true for the other three categories.
- Click Create to create the new library node.
- The new library node appears.
You can rearrange library nodes. To manually add objects from the device tree, proceed as follows: Select the Management tab, click and hold an object from the right-hand device tree, then drag over to the Drop objects here area. This transfers the entire content of the object to a new library node.
You can use notifications for libraries in very much the same way as notifications for sensors or devices.
- In your library, click the Notification Triggers tab.
- Choose one of the various notification trigger types and configure it:
Add State Trigger
Send notifications about sensor states over a specified time period.
Add Speed Trigger
Send notifications when there is too little or too much traffic over a specified time period.
Add Volume Trigger
Send notifications when a predefined amount of data in the specified interval is accumulated.
Add Threshold Trigger
Send notifications if the channel value of a sensor goes beyond or below a certain limit.
Add Change Trigger
Send change messages sent by your sensor (only certain sensors support change triggers).
You monitor hundreds of servers and want to keep an eye on CPU utilization. Using the channel limits in the sensor settings would cause an overwhelming amount of alerts because CPUs can easily spike to 100% load for one scanning interval, without necessarily indicating a problem.
Using libraries, it only takes you a couple of minutes to create a notification setup that notifies you when a CPU has been running at more than 90% load for more than 5 minutes.
We recommend the following steps:
- Create a custom library.
- Add CPU sensors (via tag or type).
- Add a threshold-based notification.
Add a state trigger to your library:
- Select the Notification Triggers tab (1).
- Hover over and select Add State Trigger (2).
- Enter the desired notification specifications (3), including:
- (4) initial parameters (triggering state, triggering state duration, notification type triggered),
- (5) escalation parameters (triggering state is always the same, ongoing duration that triggers escalation, the type of the escalation notification triggered, and how often it is repeated),
- (6) if the trigger parameters no longer apply, enter the type of notification to be triggered.
- Save the notification trigger. (7)
Additionally, you might want PRTG to keep sending you tickets when a lower load threshold is periodically exceeded. This low priority notification reminds you to keep an eye on a certain part of your network.
Add a threshold trigger by clicking Add Threshold Trigger and repeat the previous steps.
You can set up multiple triggers of the same type, so you can define increasingly serious notifications as well as progressing through different notification methods (email, SMS text, etc.). For system-critical network components, we recommend that you always define two different notification triggers (for example, email and escalation SMS text message).
The next steps are to review the notifications settings, summarization, and notification methods in the main menu bar under Setup | Account Settings | Notification Templates.
To make sure the notification system works as desired, you can trigger test notifications.
- Go to Setup | Account Settings | Notification Templates, enable the check box next to a notification and click , or click next to the notification status.
- Check if the test email and/or SMS text has arrived.
In test notification messages, placeholders are not resolved.
Troubleshooting: if you suspect a notification might not arrive, proceed as follows:
- Select Setup | Account Settings | Notification Templates from the main menu bar.
- Go to the notification action you have assigned in the library and click Edit.
- Review the notification setting so that:
- The notifications are in the Active status.
- The notifications have a schedule that does not conflict with the library you just set up.
- The summarization method collects alarms for a certain period. Also consider the Time Span for Summarizing Messages value.
- The user email addresses (phone numbers) specified for each user account are correct.
- Check the notification delivery settings under Setup | System Administration | Notification Delivery.
This only applies to PRTG on premises installations, not to PRTG hosted by Paessler instances.
- Review the logs for system events of the type Notification and the respective notification's name.
- Repeat these steps for all your notifications.
Sensors can simulate an error. Go to the library's Overview tab, right-click a sensor, and select Simulate Error Status from the context menu. You see that the sensor status quickly changes and sends the following message: The sensor shows a Down status because of a simulated error. To resolve this issue, right-click the sensor and select Resume from the context menu. (code: PE034). This also triggers the notification.