[monitoring] resource consumption / capacity planning mimir

jkleinlercher commented 1 month ago

for mimir namespace the resourcequota and usage looks like this (per default)

kubectl get resourcequota -n mimir
NAME                    AGE   REQUEST                                               LIMIT
default-resourcequota   33m   requests.cpu: 2710m/4, requests.memory: 6532Mi/16Gi   limits.cpu: 11500m/14, limits.memory: 22728Mi/24Gi

we should investigate if this is really needed or if we can decrease (or should increase??) resource requests/limits. Also the reason could be the kyverno limitrange, maybe this is complete bullshit: https://github.com/suxess-it/sx-cnp-oss/blob/bd9c80a5d5ca5a3112c62f1df068af3beb15a62f/platform-apps/charts/kyverno/templates/policy-add-ns-quota.yaml#L55-L62

https://grafana.com/docs/mimir/latest/manage/run-production-environment/planning-capacity/

In the chart there are also some t-shirt values:

https://github.com/grafana/mimir/blob/main/operations/helm/charts/mimir-distributed/small.yaml https://github.com/grafana/mimir/blob/main/operations/helm/charts/mimir-distributed/large.yaml https://github.com/grafana/mimir/blob/main/operations/helm/charts/mimir-distributed/capped-large.yaml https://github.com/grafana/mimir/blob/main/operations/helm/charts/mimir-distributed/capped-small.yaml

jkleinlercher commented 1 month ago

also do capacity planning for disk space

jkleinlercher commented 6 days ago

also configuring a retention is needed

suxess-it / sx-cnp-oss

[monitoring] resource consumption / capacity planning mimir #360