>
    Strata Copilot
    Best Practices and Troubleshooting Scenarios
    Focus
    Download PDF
    Focus
    1. Home
    2. Prisma SD-WAN
    3. PagerDuty Notifier CloudBlade Integration
    4. Best Practices and Troubleshooting Scenarios
    Download PDF
    Prisma SD-WAN

    Best Practices and Troubleshooting Scenarios

    Table of Contents

    Best Practices and Troubleshooting Scenarios

    Refer to these guidelines to manage incidents efficiently and prevent data loss during scheduled maintenance troubleshoot effectively.
    Where Can I Use This?What Do I Need?
    • Strata Cloud Manager
    • Prisma SD-WAN license
    • PagerDuty Notifier CloudBlade
    Follow these guidelines to manage incidents efficiently and prevent data loss during scheduled maintenance.

    Polling Time Interval

    • Adjust polling time based on deployment size (CloudBlade processing varies).
    • Monitor cycle processing under MonitorGeneral InformationTime Taken.
    • Set polling intervals longer than the average processing time to:
      • Avoid irregular event forwarding.
      • Prevent system overload from back-to-back requests.

    Service Directory Mapping

    • Create a dedicated Service Directory in PagerDuty for Prisma SD-WAN incidents.
    • Use the dedicated directory to:
      • Simplify incident management.
      • Improve correlation of related incidents.

    Downtime Planning

    • Prisma SD-WAN: Disable the CloudBlade during planned maintenance as this may generate a high number of unwanted incidents. It also prevents historical checkpoints from storing automatically closed incidents.
    • PagerDuty: Pause the CloudBlade during PagerDuty maintenance and resume the CloudBlade after maintenance. This ensures the CloudBlade continues from the last saved checkpoint for historically closed incidents once the maintenance is complete.

    Common Troubleshooting Scenarios

    This section addresses common issues and questions you might encounter while using the CloudBlade for your Prisma SD-WAN integration.
    1. CloudBlade Installation and Service Directories
      • Error Scenario: After installing the CloudBlade, the UI does not show any Service Directories to choose from.
        Solution: The CloudBlade is asynchronous and may take some time to complete its first processing cycle. Service directories are fetched and cached only after a cycle is complete. Check the MonitorGeneral Information tab. If there is no data, this indicates that the first cycle is incomplete. Wait for a short time and if the issue persists, contact the Palo Alto support team.
    2. Managing Service Directories
      • Error Scenario: Status of existing incidents when the Service Directory in PagerDuty changes.
        Solution: Incidents created before the change will be resolved as usual. Any new incidents will be created on the new Service Directory.
      • Error Scenario: A Service Directory in PagerDuty is accidentally deleted.
        Solution: The CloudBlade stops creating new incidents for the deleted Service Directory, and it resolves any incidents that were already created as usual.
      • Error Scenario: You wish to skip incidents generated on a select site or device.
        Solution: Tag the site or device with PD_BLOCK (case insensitive). The CloudBlade ignores any incidents associated with a site or device that has this tag.
    3. CloudBlade Pausing and Downtime
      • Error Scenario: Behavior when the CloudBlade is paused for a long time period.
        Solution: The CloudBlade continues to read and process active incidents as usual, but it enforces a hard 90-day limit when tracking older, historically resolved incidents from Prisma SD-WAN.
      • Error Scenario: An active incident in Prisma SD-WAN has not been forwarded to PagerDuty.
        Solution: Go to MonitorSD-WAN Incidents State. Filter by the event's Correlation ID. If the incident is new and not found, wait for a cycle to process it.
    4. Handling API Issues
      • Error Scenario: API issues in PagerDuty.
        Solution: The CloudBlade reads the incidents in each cycle and attempts to create or resolve them in PagerDuty.
        • For open or active incidents, the CloudBlade will retry the operation in each cycle.
        • For unresolved or historical incidents, the CloudBlade will retry the operation three times before giving up.
      • Error Scenario: API issues in Prisma SD-WAN.
        Solution: The CloudBlade errors out and does not run if it cannot connect to the Prisma SD-WAN API.

    © 2025 Palo Alto Networks, Inc. All rights reserved.