An Associate Cloud Engineer needs to perform a controlled canary rollout of a new application revision across an existing Managed Instance Group (MIG) named `frontend-mig`. The deployment strategy requires testing the updated software on a fraction of the instances before fully updating the entire group. In what order should the engineer execute the following `gcloud` operations to complete this canary deployment safely?
- 1Create a new Compute Engine instance template (`frontend-template-v2`) containing the updated application image and configuration.
- 2Run `gcloud compute instance-groups managed rolling-action start-update frontend-mig` with `--canary-version` pointing to `frontend-template-v2` and a `--target-size` specifying a small percentage of instances.
- 3Monitor application performance and instance health using `gcloud compute instance-groups managed list-instances frontend-mig` to verify canary stability.
- 4Execute `gcloud compute instance-groups managed rolling-action start-update frontend-mig` specifying `frontend-template-v2` as the primary `--version` to rollout to 100% of the instances.
Cevap
The correct sequence of operations is: 1) Create the new instance template (`frontend-template-v2`), 2) Initiate the canary update by specifying `--canary-version` with a restricted target size, 3) Verify canary instance health using `list-instances`, and 4) Promote `frontend-template-v2` to the primary version for full deployment across all instances in the group.
The proper administrative workflow for a canary update on Compute Engine Managed Instance Groups requires creating the new instance template first. Next, the engineer initiates a rolling update using canary flags to deploy the template to a minor subset of instances. After verifying instance state and application health, the engineer promotes the new template to cover 100% of the MIG.
Adım Adım Çözüm
Anahtar Kavram
Managed Instance Group Rolling Updates and Canary Deployments