AI Gateway Multi-Region Cloud Run (production)
Production Change
Change Summary
Enable Multi-Region Cloud Run feature for AI Gateway in the production environment.
Change Details
- Services Impacted - ServiceAIGateway
- Change Technician - @fforster, @cfeick
- Change Reviewer - @igorwwwwwwwwwwwwwwwwwwww
- Time tracking - 90 minutes
- Downtime Component - none
Set Maintenance Mode in GitLab
If your change involves scheduled maintenance, add a step to set and unset maintenance mode per our runbooks. This will make sure SLA calculations adjust for the maintenance period.
Detailed steps for the change
Change Steps - steps to take to execute the change
Estimated Time to Complete (mins) - 90 minutes
-
Ping @sre-oncall
in#production
on Slack -
Set label changein-progress /label ~change::in-progress
-
Merge MR gitlab-org/modelops/applied-ml/code-suggestions/ai-assist!769 (merged), adding regions us-east4
,europe-west2
, andasia-northeast3
.-
Confirm deployment pipeline finished successfully
-
-
Verify all new regions are healthy: -
Verify the overall system is healthy by inspecting the global load balancer dashboard -
Verify all regions are serving traffic by inspecting the regional detail dashboard -
Verify all regions appear in the Regional Service Aggregated RPS graph.
Stable state: the system is stable at this point. We can roll back to this point and/or end the change here if needed.
-
Manually remove the us-east1
backend from the global load balancer URL map; this is an unfortunate manual step due to the issue documented at gitlab-com/gl-infra/platform/runway/team#111 (comment 1835120869). -
Verify the overall system continues to be healthy by inspecting the global load balancer dashboard -
Merge MR gitlab-org/modelops/applied-ml/code-suggestions/ai-assist!770 (merged), removing region us-east1
-
Confirm deployment pipeline finished successfully
-
-
Verify the overall system continues to be healthy by inspecting the global load balancer dashboard
Stable state: the system is stable at this point. We can roll back to this point and/or end the change here if needed.
-
Merge MR gitlab-org/modelops/applied-ml/code-suggestions/ai-assist!771 (merged), configuring regions to talk to the local Vertex AI endpoint. This needs to be done after removing us-east1
, because Vertex AI inus-east1
is incompatible.-
Confirm deployment pipeline finished successfully
-
-
Verify the overall system continues to be healthy by inspecting the global load balancer dashboard -
Verify all regions are serving traffic by inspecting the regional detail dashboard -
Spot-test code completion via VS Code on the local laptop. -
Set label changecomplete /label ~change::complete
-
Follow up in #production
to give the all-clear.
Rollback
Rollback steps - steps to be taken in the event of a need to rollback this change
Estimated Time to Complete (mins) - 15 minutes
- If the problem was caused by merging an MR:
-
Merge a rollback-MR of the previous MR, if applicable. This will trigger a Terraform run that will restore the last known good configuration.
-
- If the problem was caused by a manual change:
-
Trigger a new run of the deployment pipeline by creating a new (possibly empty) change and merging it. This will restore the last known good configuration.
-
-
Set label changeaborted /label ~change::aborted
Monitoring
Key metrics to observe
- Metric: global load balancer requests
- Location: runway: Runway Load Balancer Metrics
- What changes to this metric should prompt a rollback:
- Any change in the overall traffic pattern, e.g. increase in non-successful requests, sudden drop in overall requests.
Change Reviewer checklist
-
Check if the following applies: - The scheduled day and time of execution of the change is appropriate.
- The change plan is technically accurate.
- The change plan includes estimated timing values based on previous testing.
- The change plan includes a viable rollback plan.
- The specified metrics/monitoring dashboards provide sufficient visibility for the change.
-
Check if the following applies: - The complexity of the plan is appropriate for the corresponding risk of the change. (i.e. the plan contains clear details).
- The change plan includes success measures for all steps/milestones during the execution.
- The change adequately minimizes risk within the environment/service.
- The performance implications of executing the change are well-understood and documented.
- The specified metrics/monitoring dashboards provide sufficient visibility for the change.
- If not, is it possible (or necessary) to make changes to observability platforms for added visibility?
- The change has a primary and secondary SRE with knowledge of the details available during the change window.
- The change window has been agreed with Release Managers in advance of the change. If the change is planned for APAC hours, this issue has an agreed pre-change approval.
- The labels blocks deployments and/or blocks feature-flags are applied as necessary.
Change Technician checklist
-
Check if all items below are complete: - The change plan is technically accurate.
- This Change Issue is linked to the appropriate Issue and/or Epic
- Change has been tested in staging and results noted in a comment on this issue.
- A dry-run has been conducted and results noted in a comment on this issue.
- The change execution window respects the Production Change Lock periods.
- For C1 and C2 change issues, the change event is added to the GitLab Production calendar.
- For C1 and C2 change issues, the SRE on-call has been informed prior to change being rolled out. (In #production channel, mention
@sre-oncall
and this issue and await their acknowledgement.) - For C1 and C2 change issues, the SRE on-call provided approval with the eoc_approved label on the issue.
- For C1 and C2 change issues, the Infrastructure Manager provided approval with the manager_approved label on the issue.
- Release managers have been informed prior to any C1, C2, or blocks deployments change being rolled out. (In #production channel, mention
@release-managers
and this issue and await their acknowledgment.) - There are currently no active incidents that are severity1 or severity2
- If the change involves doing maintenance on a database host, an appropriate silence targeting the host(s) should be added for the duration of the change.
Edited by Florian Forster