2022-12-01: Gradually increase the number of maxreplicas for 2 Sidekiq shards

Production Change

Change Summary

Gradually increase the number of maxReplicas for 2 shards that are currently using the max number of replicas available sometimes. This is brought up in https://gitlab.com/gitlab-com/gl-infra/capacity-planning/-/issues/309#note_1116023566.

1. catchall from 300 to 310330 in steps of 10(https://gitlab.com/gitlab-com/gl-infra/capacity-planning/-/issues/309#note_1192955270)
2. urgent_cpu_bound from 157 to 172

Change Details

1. Services Impacted - ServiceSidekiq
2. Change Technician - @reprazent
3. Change Reviewer - @reprazent
4. Time tracking - N/A
5. Downtime Component - N/A

Detailed steps for the change

Change Steps - steps to take to execute the change

Estimated Time to Complete (mins) - Estimated Time to Complete in Minutes

• Set label changein-progress /label ~change::in-progress
• gitlab-com/gl-infra/k8s-workloads/gitlab-com!2164 (merged)
• gitlab-com/gl-infra/k8s-workloads/gitlab-com!2166 (merged)
• Set label changecomplete /label ~change::complete

Rollback

Rollback steps - steps to be taken in the event of a need to rollback this change

Estimated Time to Complete (mins) - Estimated Time to Complete in Minutes

• gitlab-com/gl-infra/k8s-workloads/gitlab-com!2166 (merged)
• gitlab-com/gl-infra/k8s-workloads/gitlab-com!2164 (merged)
• Set label changeaborted /label ~change::aborted

Monitoring

Key metrics to observe

• Metric: HPA Saturation dashboard

  • Location: https://dashboards.gitlab.net/d/alerts-sat_kube_horizontalpodautoscaler/alerts-kube_horizontalpodautoscaler_desired_replicas-saturation-detail?var-environment=gprd&var-type=sidekiq&var-stage=main&var-component=kube_horizontalpodautoscaler_desired_replicas&orgId=1&from=now-2d&to=now
  • What changes to this metric should prompt a rollback: This is the metric we're trying to improve. Saturation for catchall and urgent_cpu_bound will go down.

• Metric: pgbouncer backend connections

  • Location:
    • Primary: https://dashboards.gitlab.net/d/alerts-sat_pgbouncer_async_pool_primary/alerts-pgbouncer_async_primary_pool-saturation-detail?orgId=1
    • Secondary: https://dashboards.gitlab.net/d/alerts-sat_pgbouncer_async_pool_replica/alerts-pgbouncer_async_replica_pool-saturation-detail?orgId=1 this is unlikely to saturate
  • What: if there's no backend connections, we might have to increase the connection pools. Otherwise jobs will wait obtaining a connection, making it so we don't actually use the newly added pods

• Metric: pgbouncer client connections

  • Location:
    • Primary: https://dashboards.gitlab.net/d/alerts-sat_pgb_client_conn_primary/alerts-pgbouncer_client_conn_primary-saturation-detail?orgId=1
    • Secondary: https://dashboards.gitlab.net/d/alerts-sat_pgb_client_conn_replicas/alerts-pgbouncer_client_conn_replicas-saturation-detail?orgId=1
  • What: No client connections could result in errors when executing jobs when no connection is available. The pgbouncers for sidekiq seem to have plenty of headroom

• Metric: Redis primary CPU

  • Location: https://dashboards.gitlab.net/d/alerts-sat_redis_primary_cpu/alerts-redis_primary_cpu-saturation-detail?orgId=1&var-PROMETHEUS_DS=Global&var-environment=gprd&var-type=redis&var-stage=main
  • What: Check both persistent & cache which are close to saturation. The increased workload could push them over the edge, but it seems unlikely

• Metric: Kube pool max nodes

  • Location: https://dashboards.gitlab.net/d/alerts-sat_kube_pool_max_nodes/alerts-kube_pool_max_nodes-saturation-detail?orgId=1&var-PROMETHEUS_DS=Global&var-environment=gprd&var-type=sidekiq&var-stage=main
  • What: Make sure there's enough nodes available to schedule the new pods on.

Change Reviewer checklist

C4 C3 C2 C1:

• 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.

C2 C1:

• 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 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 (If needed! Cases include DB change) prior to 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.