Scalability

Use this topic to learn about the scalability of tenant configuration DRC timeout and REST request timeout.

Scaled REST Request Timeout

When you run a scaled XCO tenant REST request (which takes more than 15 minutes), it fails with the following error:

Service is not available or internal server error has occurred, please try again later.

Run the show command to verify the successful completion of failed REST request.

Scaled DRC Timeout

Tenant DRC Behavior Tenant DRC Behavior in case of Scale Config
  • When you run the efa inventory drift-reconcile execute command, the tenant configuration drift and reconciliation starts with other services (for example, inventory, fabric, and policy).
  • During the tenant DRC, the tenant drift is identified, and then the same drift is reconciled. The timeout for both tenant configuration drift identification and reconciliation is 15 minutes.
  • The inventory drift-reconcile execute command waits for the tenant DRC response for 15 minutes. If the response is not received within 15 minutes, then the DRC fails with the reason code tenant-dr-timeout.
  • The functionality is applicable for manual DRC and auto DRC (triggers when the device is reloaded in maintenance mode).
  • In case of scaled tenant configuration drift (for example, 2000 VE or PO drifted and 100 VRFs drifted), the tenant can take more than 15 minutes for drift identification and reconciliation. Therefore, the DRC fails with the status tenant-dr-timeout.
  • The tenant configuration drift identification and reconciliation run to completion in the background even though the DRC fails with the status tenant-dr-timeout.
  • The subsequent DRC reflects the correct status of the drift.