Minimum Loki components to enable EKS Cluster Logging

Grafana Loki
1

Hi!

After installing Grafana Loki, I got these Pods running:

kubectl get pods -n loki                                                                                                                                                                                                                         <aws:DdTesting>
NAME                                    READY   STATUS    RESTARTS   AGE
loki-canary-ktbsb                       1/1     Running   0          136m
loki-canary-pbsk2                       1/1     Running   0          134m
loki-chunks-cache-0                     2/2     Running   0          105m
loki-compactor-0                        1/1     Running   0          105m
loki-distributor-75656bdc79-7p7nj       1/1     Running   0          105m
loki-gateway-6c59cdf9d6-gtnqc           1/1     Running   0          136m
loki-index-gateway-0                    1/1     Running   0          105m
loki-ingester-0                         1/1     Running   0          105m
loki-querier-5666db9f57-klk65           1/1     Running   0          105m
loki-query-frontend-7946cf88b9-hkd67    1/1     Running   0          105m
loki-query-scheduler-54bdb7ffdc-rxhq2   1/1     Running   0          105m
loki-results-cache-0                    2/2     Running   0          136m
loki-ruler-0                            1/1     Running   0          105m

Do I actually need all of them just to get logs from an Apps along with some Nodes files from /var/logs/? I am going to use Grafana to observe logging data.

2

There are three modes for running Loki, see Loki deployment modes | Grafana Loki documentation

It’s up to you to decide what works for you based on your work load. If you already have it running in distributed mode and you may have need to scale it in the future then perhaps stay on the same path. Otherwise you can always consider just running one single Loki container.

1 Like
3

Thank you, Tony!
This is exactly what I was looking for…