Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1962305

Summary: Adjust memory requests of monitoring stack components
Product: OpenShift Container Platform Reporter: Damien Grisonnet <dgrisonn>
Component: MonitoringAssignee: Filip Petkovski <fpetkovs>
Status: CLOSED DUPLICATE QA Contact: Junqi Zhao <juzhao>
Severity: medium Docs Contact:
Priority: medium    
Version: 4.8CC: alegrand, anpicker, aos-bugs, erooth, kakkoyun, pkrupa
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2021-05-20 08:00:11 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
memory request during upgrade e2e test
none
memory request during operator e2e test none

Description Damien Grisonnet 2021-05-19 17:31:15 UTC 
Created attachment 1784879 [details]
memory request during upgrade e2e test

Description of problem:

With the recent changes from: https://github.com/openshift/cluster-monitoring-operator/pull/1158, the memory request of some of the containers of the monitoring stack were adjusted, but according to some recent tests on CI clusters, there seems to still be room for improvements.

I attached to this bugzilla, the results of two queries ran against the metrics from a e2e-agnostic-operator and e2e-agnostic-upgrade run in CI. Both of these were including the recent changes made in PR 1158. A negative result means that the memory was requested but not used, and a positive result means that more memory was used than the one that was requested. Results are in MiB per pod per container in the openshift-monitoring namespace.

From a first look, we should be able to:
- reduce memory request of thanos-sidecar to 25MiB
- align all kube-rbac-proxy memory requests to ~12MiB
- increase memory request of prometheus-operator, kube-state-metrics and prometheus-adapter

How reproducible:
The memory usage can be queried from Prometheus in a live cluster and can be easily seen.

Steps to Reproduce:
1. Create a cluster with cluster bot
2. Execute a query against prometheus to get the memory usage of pods in the openshift-monitoring space vs the actual memory usage of those pods. The exact query can be found in the attachments.
Comment 1 Damien Grisonnet 2021-05-19 17:32:18 UTC 
Created attachment 1784880 [details]
memory request during operator e2e test
Comment 2 Filip Petkovski 2021-05-20 08:00:11 UTC 

*** This bug has been marked as a duplicate of bug 1962261 ***