Work a Core Assurance Observation
Observations
A record of a condition in your environment that needs attention. Observations appear on the Observations page and in the Observation Feed widget, and they move through lifecycle states. For example, Core Assurance creates an observation when the percentage of online endpoints that fail a test is at or above the trigger set in the test profile. from Core Assurance appear on the Observations page and in the Observation Feed widget, with your other observations. Work an observation to find out which test endpoints
A user device, such as a laptop or desktop computer, that has the Core Assurance module installed and runs synthetic tests. Each test endpoint receives test profiles from the cloud, runs them on schedule, and sends its results back. are affected, what is failing, and when the problem is over.
To work a Core Assurance observation:
-
Click Observations in the left menu. The Live Feed page opens.
-
Click Open to show only open observations, then click the observation. The Observation Details page opens.
-
While the observation is open, the header shows Ongoing in place of an end time.
-
-
Note the related application, the related test, when the observation started, and how long the observation has been open.
-
Review the Supporting Artifacts card, if the card lists any artifacts. An artifact is a piece of supporting information, such as an analysis result, that a service, an agent, or a rule attached to the observation. Each artifact shows its name, description, type, producer, and when the artifact was produced. To see the full content of an artifact, click Show full artifact.
-
Review the Actions card, if the card lists any actions. Each action is a suggested next step. Click an action to open a chat panel where Riverbed Q works on the suggested step with you. See Agents.
-
Open the related test profile
A reusable Core Assurance definition of a synthetic test. A test profile sets what to test, what counts as a passing run, when the test runs, which test endpoints run it, and what to do when enough endpoints fail., then review its runs on the Test Results and Test Details pages. See Determine Whether Failures Are Isolated or Widespread. -
Identify which endpoints are failing and look for error patterns. For example, check whether the failures are timeouts, HTTP (Hypertext Transfer Protocol
Hypertext Transfer Protocol. A protocol used for transmitting data over the internet, commonly used in webhook runbooks and integrations with external systems.) status errors, DNS (Domain Name System) errors, or network path problems. See Core Assurance Failure Codes. -
Fix the cause outside Core Assurance, as needed. For example, address the network, the application, or DNS, or roll back a recent change.
-
Confirm recovery. The failure percentage declines, the profile moves to Dampening, and the observation closes when the dampening period
A Core Assurance cool-down period that starts when the failure percentage for a test profile is no longer above its trigger. New failed runs are still logged during the dampening period, but no new action is taken. The related observation closes when the period ends. ends.