Service Maturity Model

Introduction

This page shows the output of our service maturity model for each service in our metrics catalog. The model itself is part of the metrics catalog, and uses information from the metrics catalog and the service catalog to score each service.

To achieve a particular level in the maturity model, a service must meet all the criteria for that level and all previous levels. Some criteria do not apply to all services (for instance, services like PgBouncer do not need development documentation).

Maturity score by service

❌ indicates the service does meet even the Level 1 criteria

Service	Level
ai-assisted	Level 3
ai-gateway	Level 2
api	Level 2
atlantis	Level 1
camoproxy	Level 2
ci-runners	Level 2
cloud-sql	Level 1
cloudflare	Level 1
consul	Level 1
customersdot	Level 3
errortracking	Level 2
ext-pvs	Level 3
external-dns	Level 1
frontend	Level 1
git	Level 2
gitaly	Level 3
gitlab-static	Level 1
glgo	Level 2
google-cloud-storage	Level 2
internal-api	Level 3
istio	Level 2
jaeger	Level 2
kas	Level 3
kube	Level 2
logging	Level 1
mailgun	Level 3
mailroom	Level 1
memorystore	Level 1
mimir	Level 2
monitoring	Level 2
nat	Level 1
nginx	Level 1
ops-gitlab-net	Level 3
packagecloud	Level 1
patroni	Level 2
patroni-ci	Level 2
patroni-embedding	Level 1
patroni-registry	Level 1
pgbouncer	Level 1
pgbouncer-ci	Level 1
pgbouncer-embedding	Level 1
pgbouncer-registry	Level 1
plantuml	Level 1
postgres-archive	Level 1
redis	Level 3
redis-cluster-cache	Level 3
redis-cluster-chat-cache	Level 3
redis-cluster-feature-flag	Level 3
redis-cluster-queues-meta	Level 3
redis-cluster-ratelimiting	Level 3
redis-cluster-repo-cache	Level 3
redis-cluster-shared-state	Level 3
redis-db-load-balancing	Level 3
redis-pubsub	Level 3
redis-registry-cache	Level 2
redis-sessions	Level 3
redis-sidekiq	Level 3
redis-tracechunks	Level 3
registry	Level 2
runway	Level 1
search	Level 1
sentry	Level 2
sidekiq	Level 2
thanos	Level 2
tracing	Level 2
vault	Level 2
web	Level 2
web-pages	Level 2
websockets	Level 2
woodhouse	Level 3

Maturity detail by service

Key:

• ✅ Service meets the criteria
• ❌ Service does not meet the criteria
• ➖ The criteria is skipped. Some maturity criteria make less sense for some services. For example, an infrastructure-facing service like Patroni is crucial to ops, but not related to our Development department, hence it does not require development guidelines.
• ⚪ We don’t measure the criteria yet. See https://gitlab.com/groups/gitlab-com/gl-infra/-/epics/560 for progress

ai-assisted detail

• Level 3
• Link to ai-assisted dashboard
• Service definition of ai-assisted

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	⚪ Not Implemented
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

ai-gateway detail

• Level 2
• Link to ai-gateway dashboard
• Service definition of ai-gateway

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Runway structured logs are temporarily available in Stackdriver
	Service exists in the dependency graph	➖ Reason: Runway services are deployed outside of the monolith
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	❌
	SRE guides exist in runbooks	❌
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

api detail

• Level 2
• Link to api dashboard
• Service definition of api

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6, 7, 8, 9
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7, 8
	SLA calculations driven from SLO metrics	⚪ Not Implemented
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

atlantis detail

• Level 1
• Link to atlantis dashboard
• Service definition of atlantis

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Atlantis is a work in progress, see https://gitlab.com/gitlab-com/gl-infra/reliability/-/issues/24613
	Service exists in the dependency graph	➖ Reason: Atlantis is a work in progress, see https://gitlab.com/gitlab-com/gl-infra/reliability/-/issues/24613
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: Atlantis is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

camoproxy detail

• Level 2
• Link to camoproxy dashboard
• Service definition of camoproxy

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	➖ Reason: Camoproxy does not interact directly with any declared services in our system
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

ci-runners detail

• Level 2
• Link to ci-runners dashboard
• Service definition of ci-runners

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11
	SLA calculations driven from SLO metrics	⚪ Not Implemented
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

cloud-sql detail

• Level 1
• Link to cloud-sql dashboard
• Service definition of cloud-sql

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Cloud SQL is a managed service of GCP. The logs are available in Stackdriver.
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	❌
	SLO monitoring: request rate	❌
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: Cloud SQL is an infrastructure component, powered by GCP
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

cloudflare detail

• Level 1
• Link to cloudflare dashboard
• Service definition of cloudflare

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Logs from CloudFlare are pushed to a GCS bucket by CloudFlare, and not ingested to ElasticSearch due to volume. See https://gitlab.com/gitlab-com/runbooks/-/blob/master/docs/cloudflare/logging.md for alternatives
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: WAF is an infrastructure component, powered by Cloudflare
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

consul detail

• Level 1
• Link to consul dashboard
• Service definition of consul

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: Consul is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

customersdot detail

• Level 3
• Link to customersdot dashboard
• Service definition of customersdot

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: All logs are available in Stackdriver
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

errortracking detail

• Level 2
• Link to errortracking dashboard
• Service definition of errortracking

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	❌
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

ext-pvs detail

• Level 3
• Link to ext-pvs dashboard
• Service definition of ext-pvs

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Runway structured logs are temporarily available in Stackdriver
	Service exists in the dependency graph	➖ Reason: Runway services are deployed outside of the monolith
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

external-dns detail

• Level 1
• Link to external-dns dashboard
• Service definition of external-dns

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Logs from external-dns are not ingested to ElasticSearch due to volume. Besides, the logs are also available in Stackdriver
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	❌
	SLO monitoring: request rate	❌
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: external-dns is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

frontend detail

• Level 1
• Link to frontend dashboard
• Service definition of frontend

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Logs from HAProxy are available in BigQuery, and not ingested to ElasticSearch due to volume.
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

git detail

• Level 2
• Link to git dashboard
• Service definition of git

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7, 8
	SLA calculations driven from SLO metrics	⚪ Not Implemented
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

gitaly detail

• Level 3
• Link to gitaly dashboard
• Service definition of gitaly

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

gitlab-static detail

• Level 1
• Link to gitlab-static dashboard
• Service definition of gitlab-static

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Logs from CloudFlare workers are available on-demand but they are not being ingested due to volume
	Service exists in the dependency graph	➖ Reason: This service is hosted by Cloudflare and does not depend on any other service
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	⚪ Not Implemented
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

glgo detail

• Level 2
• Link to glgo dashboard
• Service definition of glgo

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Runway structured logs are temporarily available in Stackdriver
	Service exists in the dependency graph	➖ Reason: Runway services are deployed outside of the monolith
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	❌
	SRE guides exist in runbooks	❌
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

google-cloud-storage detail

• Level 2
• Link to google-cloud-storage dashboard
• Service definition of google-cloud-storage

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Access logs of GCS and not enabled due to volume.
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

internal-api detail

• Level 3
• Link to internal-api dashboard
• Service definition of internal-api

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7
	SLA calculations driven from SLO metrics	⚪ Not Implemented
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

istio detail

• Level 2
• Link to istio dashboard
• Service definition of istio

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Istio service is not deployed in production
	Service exists in the dependency graph	➖ Reason: This service does not interfact directly with any other services
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: Istio is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

jaeger detail

• Level 2
• Link to jaeger dashboard
• Service definition of jaeger

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Jaeger service is not deployed in production
	Service exists in the dependency graph	➖ Reason: Jaeger is an independent internal observability tool
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

kas detail

• Level 3
• Link to kas dashboard
• Service definition of kas

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7, 8
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

kube detail

• Level 2
• Link to kube dashboard
• Service definition of kube

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1
	Service exists in the dependency graph	➖ Reason: This service is managed by GKE at the moment. It does not interfact directly with any other services
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: Application logic does not interact with kube
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

logging detail

• Level 1
• Link to logging dashboard
• Service definition of logging

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11
	Service exists in the dependency graph	➖ Reason: The logging platform consumes logs via fluentd, but does not interact directly with any other services
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

mailgun detail

• Level 3
• Link to mailgun dashboard
• Service definition of mailgun

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Mailgun is a vendor
	Service exists in the dependency graph	➖ Reason: Mailgun is a vendor
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	⚪ Not Implemented
	SRE guides exist in runbooks	⚪ Not Implemented
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

mailroom detail

• Level 1
• Link to mailroom dashboard
• Service definition of mailroom

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

memorystore detail

• Level 1
• Link to memorystore dashboard
• Service definition of memorystore

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Memorystore is a managed service of GCP. The logs are available in Stackdriver.
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	❌
	SLO monitoring: request rate	❌
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: Memorystore is an infrastructure component, powered by GCP
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

mimir detail

• Level 2
• Link to mimir dashboard
• Service definition of mimir

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6, 7, 8
	Service exists in the dependency graph	➖ Reason: Mimir is an independent internal observability tool. It fetches metrics from other services, but does not interact with them, functionally
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	⚪ Not Implemented
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

monitoring detail

• Level 2
• Link to monitoring dashboard
• Service definition of monitoring

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

nat detail

• Level 1
• Link to nat dashboard
• Service definition of nat

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: NAT is managed by GCP, thus the logs are avaiable in Stackdriver.
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: NAT is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

nginx detail

• Level 1
• Link to nginx dashboard
• Service definition of nginx

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Logs from nginx are not ingested to ElasticSearch due to volume. Usually, workhorse logs will cover the same ground. Besides, the logs are also available in Stackdriver
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: Application logic does not interact with nginx
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

ops-gitlab-net detail

• Level 3
• Link to ops-gitlab-net dashboard
• Service definition of ops-gitlab-net

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19
	Service exists in the dependency graph	➖ Reason: ops.gitlab.net is a standalone GitLab deployment
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	⚪ Not Implemented
	SRE guides exist in runbooks	⚪ Not Implemented
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

packagecloud detail

• Level 1
• Link to packagecloud dashboard
• Service definition of packagecloud

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	⚪ Not Implemented
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

patroni detail

• Level 2
• Link to patroni dashboard
• Service definition of patroni

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: patroni is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

patroni-ci detail

• Level 2
• Link to patroni-ci dashboard
• Service definition of patroni-ci

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: patroni is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

patroni-embedding detail

• Level 1
• Link to patroni-embedding dashboard
• Service definition of patroni-embedding

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: patroni is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

patroni-registry detail

• Level 1
• Link to patroni-registry dashboard
• Service definition of patroni-registry

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: patroni is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

pgbouncer detail

• Level 1
• Link to pgbouncer dashboard
• Service definition of pgbouncer

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: pgbouncer is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

pgbouncer-ci detail

• Level 1
• Link to pgbouncer-ci dashboard
• Service definition of pgbouncer-ci

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: pgbouncer is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

pgbouncer-embedding detail

• Level 1
• Link to pgbouncer-embedding dashboard
• Service definition of pgbouncer-embedding

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: pgbouncer is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

pgbouncer-registry detail

• Level 1
• Link to pgbouncer-registry dashboard
• Service definition of pgbouncer-registry

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: pgbouncer is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

plantuml detail

• Level 1
• Link to plantuml dashboard
• Service definition of plantuml

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: The logs are available in Stackdriver.
	Service exists in the dependency graph	➖ Reason: Platuml is a is a stateless web application that generates UML diagrams on the fly. The rendered markdown points to the platuml server in the frontends. It does not interact with any declared services
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

postgres-archive detail

• Level 1
• Link to postgres-archive dashboard
• Service definition of postgres-archive

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: postgres-archive is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis detail

• Level 3
• Link to redis dashboard
• Service definition of redis

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-cluster-cache detail

• Level 3
• Link to redis-cluster-cache dashboard
• Service definition of redis-cluster-cache

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-cluster-chat-cache detail

• Level 3
• Link to redis-cluster-chat-cache dashboard
• Service definition of redis-cluster-chat-cache

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-cluster-feature-flag detail

• Level 3
• Link to redis-cluster-feature-flag dashboard
• Service definition of redis-cluster-feature-flag

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-cluster-queues-meta detail

• Level 3
• Link to redis-cluster-queues-meta dashboard
• Service definition of redis-cluster-queues-meta

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-cluster-ratelimiting detail

• Level 3
• Link to redis-cluster-ratelimiting dashboard
• Service definition of redis-cluster-ratelimiting

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	⚪ Not Implemented
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-cluster-repo-cache detail

• Level 3
• Link to redis-cluster-repo-cache dashboard
• Service definition of redis-cluster-repo-cache

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-cluster-shared-state detail

• Level 3
• Link to redis-cluster-shared-state dashboard
• Service definition of redis-cluster-shared-state

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-db-load-balancing detail

• Level 3
• Link to redis-db-load-balancing dashboard
• Service definition of redis-db-load-balancing

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-pubsub detail

• Level 3
• Link to redis-pubsub dashboard
• Service definition of redis-pubsub

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-registry-cache detail

• Level 2
• Link to redis-registry-cache dashboard
• Service definition of redis-registry-cache

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	❌
	SRE guides exist in runbooks	❌
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-sessions detail

• Level 3
• Link to redis-sessions dashboard
• Service definition of redis-sessions

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-sidekiq detail

• Level 3
• Link to redis-sidekiq dashboard
• Service definition of redis-sidekiq

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

redis-tracechunks detail

• Level 3
• Link to redis-tracechunks dashboard
• Service definition of redis-tracechunks

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	➖ Reason: Metadata can't be injected in redis logs
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

registry detail

• Level 2
• Link to registry dashboard
• Service definition of registry

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13
	SLA calculations driven from SLO metrics	⚪ Not Implemented
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

runway detail

• Level 1
• Link to runway dashboard
• Service definition of runway

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Runway is a platform. The logs are available in Stackdriver.
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	❌
	SLO monitoring: error rate	❌
	SLO monitoring: request rate	❌
Level 3	Service health dashboards	✅ 1, 2, 3
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

search detail

• Level 1
• Link to search dashboard
• Service definition of search

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	❌
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

sentry detail

• Level 2
• Link to sentry dashboard
• Service definition of sentry

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: We are migrating our self-managed Sentry instance to the hosted one. For more information: https://gitlab.com/gitlab-com/gl-infra/reliability/-/issues/13963. Besides, Sentry logs are also available in Stackdriver.
	Service exists in the dependency graph	➖ Reason: Sentry is an independent internal observability tool
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

sidekiq detail

• Level 2
• Link to sidekiq dashboard
• Service definition of sidekiq

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6, 7
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
	SLA calculations driven from SLO metrics	⚪ Not Implemented
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

thanos detail

• Level 2
• Link to thanos dashboard
• Service definition of thanos

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
	Service exists in the dependency graph	➖ Reason: Thanos is an independent internal observability tool. It fetches metrics from other services, but does not interact with them, functionally
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	⚪ Not Implemented
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

tracing detail

• Level 2
• Link to tracing dashboard
• Service definition of tracing

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	❌
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

vault detail

• Level 2
• Link to vault dashboard
• Service definition of vault

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Vault is a pending project at the moment. There is no traffic at the moment. We'll add logs and metrics in https://gitlab.com/groups/gitlab-com/gl-infra/-/epics/739
	Service exists in the dependency graph	➖ Reason: Vault is a pending project at the moment. There is no traffic at the moment. The progress can be tracked at https://gitlab.com/groups/gitlab-com/gl-infra/-/epics/739
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	➖ Reason: Vault is an infrastructure component, developers do not interact with it
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

web detail

• Level 2
• Link to web dashboard
• Service definition of web

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6, 7, 8
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7
	SLA calculations driven from SLO metrics	⚪ Not Implemented
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

web-pages detail

• Level 2
• Link to web-pages dashboard
• Service definition of web-pages

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6, 7
	SLA calculations driven from SLO metrics	⚪ Not Implemented
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

websockets detail

• Level 2
• Link to websockets dashboard
• Service definition of websockets

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	✅ 1, 2, 3, 4, 5, 6
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1, 2, 3, 4, 5, 6
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	❌
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented

woodhouse detail

• Level 3
• Link to woodhouse dashboard
• Service definition of woodhouse

Level	Criterion	Passed
Level 1	Exists in the service catalog	✅ 1
	Structured logs available in Kibana	➖ Reason: Log volume is very low; tooling links to StackDriver provided which is sufficient for the purposes
	Service exists in the dependency graph	✅ 1
Level 2	SLO monitoring: apdex	✅ 1
	SLO monitoring: error rate	✅ 1
	SLO monitoring: request rate	✅ 1
Level 3	Service health dashboards	✅ 1
	SLA calculations driven from SLO metrics	➖ Reason: Service is not user facing
	All components include an apdex	✅ 1
	Logging includes metadata for measuring scalability	⚪ Not Implemented
	Developer guides exist in developer documentation	✅ 1
	SRE guides exist in runbooks	✅ 1
	Metrics on downstream service usage	⚪ Not Implemented
Level 4	Prepared Kibana dashboards	⚪ Not Implemented
	Dashboards linked from metrics catalogs	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented
Level 5	Long-term forecasting utilization and usage	⚪ Not Implemented
	70% of requests covered by at least one SLI	⚪ Not Implemented
	Automatic alert routing	⚪ Not Implemented