Loki Configuration Guide
Purpose: For platform engineers, operators, shows how to configure Loki in cluster repositories that consume the openCenter base.
Use this guide when a cluster repo needs to tune Loki or wire log collection into the observability stack.
What the Base Deploys
The base service deploys:
-
HelmRelease/loki -
base chart values from the service
helm-values/directory -
optional
Secret/loki-values-override
The base does not create cluster-specific storage credentials or log shipping pipelines.
Common Cluster-Specific Configuration
Most clusters need to configure:
-
object storage backend and credentials
-
retention and compaction behavior
-
log ingestion path from OpenTelemetry or other collectors
-
tenant strategy if multi-tenancy is needed
-
ingress or gateway exposure for Grafana-to-Loki access, if externalized
Override Values Pattern
Example override.yaml:
loki:
commonConfig:
replication_factor: 2
storage:
type: s3
bucketNames:
chunks: loki-chunks
ruler: loki-ruler
admin: loki-admin
Integration Notes
-
If the platform uses
opentelemetry-kube-stack, collector exporters usually point at the Loki gateway or write endpoint. -
Keep label design disciplined. Excessive cardinality is a common reason Loki becomes expensive or unstable.
Verification
kubectl get helmrelease -n observability loki
kubectl get pods -n observability -l app.kubernetes.io/name=loki
kubectl logs -n observability deploy/loki-read
Healthy signs:
-
write, read, and backend components are
Running -
ingestion succeeds from collectors
-
Grafana queries return expected logs