Holt-Winters Threshold
You are viewing the ZenPack Archive
This page is part of the ZenPack Archive. Archived ZenPacks may not be compatible with your version of Zenoss Service Dynamics or Zenoss Cloud.
Commercial
This ZenPack is developed and supported by Zenoss Inc. Commercial ZenPacks are available to Zenoss commercial customers only. Contact Zenoss to request more information regarding this or any other ZenPacks. Click here to view all available Zenoss Commercial ZenPacks.
Organization
Zenoss, Inc.
Name
ZenPacks.zenoss.ZenHoltWinters
Holt-Winters Threshold ZenPack
The ZenHoltWinters ZenPack adds the ability to create threshold events when a device exceeds cyclical predicted values. The Holt-Winters exponential smoothing algorithm is used for this prediction.
Support
This ZenPack is included with commercial versions of Zenoss and enterprise support for this ZenPack is provided to Zenoss customers with an active subscription.
Background
The ZenPacks.zenoss.ZenHoltWinters ZenPack adds the ability to create threshold events when a device exceeds cyclical predicted values.
The Holt-Winters exponential smoothing algorithm is used for predictiond. For more information on RRD and Holt-Winters, refer to the rrdcreate command.
Warning: This ZenPack is not compatible with Zenoss 5 and later.
Warning: Zenoss platform relies on the existence of Holt-Winters RRAs within an RRD file. After adding Holt-Winters thresholds, the RRD files will need to be re-created so that the new configuration can occur. You will have to remove any existing RRD files so that new files can be created.
Removing RRD files will remove all historical information associated with these RRD files.
Prerequisites
Prerequisite | Restriction |
Product | Zenoss platform 4.x, Zenoss 2.2 or higher |
Required ZenPacks | ZenPacks.zenoss.ZenHoltWinters |
Add a Predictive Threshold
-
Navigate to the template that you want to modify.
-
From the Thresholds area, click (Add Threshold).
-
Provide a name for the new threshold and select the
HoltWintersFailure
threshold type, and then click Add. -
Choose the data source to which the threshold should be applied.
-
Specify the parameters for the prediction engine.
Predictive Threshold Data Source Threshold Options | | | |--------|---------------------------------------------------------------------------------------------------------| | Name | Description | | Rows | The number of points to use for predictive purposes. | | Alpha | A number from 0 to 1 that controls how quickly the model adapts to unexpected values. | | Beta | A number from 0 to 1 that controls how quickly the model adapts to changes in unexpected rates changes. | | Season | The number of primary data points in a season. Note that Rows must be at least as large as Season. |
-
Click Save to save your changes.
-
Remove the RRD file or files that correspond to the data source selected in a previous step.
cd $ZENHOME/perf/Devices rm device_names/DataSource_DataPoint.rrd
Note: Removing the RRD files does result in a loss of historical information.