Bug 1822770
| Summary: | Default openshift install requests too many CPU resources to install all components, requests of components on cluster are wrong | ||
|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | Jeremy Poulin <jpoulin> |
| Component: | kube-apiserver | Assignee: | Michal Fojtik <mfojtik> |
| Status: | CLOSED ERRATA | QA Contact: | Ke Wang <kewang> |
| Severity: | urgent | Docs Contact: | |
| Priority: | urgent | ||
| Version: | 4.3.z | CC: | adahiya, aos-bugs, bparees, ccoleman, esimard, Holger.Wolf, hwolf, jchaloup, jeder, lakshmi.ravichandran1, maszulik, mfojtik, miwilson, rphillips, sdodson, wjiang, wking, wvoesch, xxia, zyu |
| Target Milestone: | --- | Keywords: | ServiceDeliveryImpact |
| Target Release: | 4.3.z | ||
| Hardware: | s390x | ||
| OS: | Unspecified | ||
| Whiteboard: | multi-arch | ||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | 1812583 | Environment: | |
| Last Closed: | 2020-05-20 13:47:53 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Bug Depends On: | 1812709 | ||
| Bug Blocks: | |||
|
Comment 1
Scott Dodson
2020-04-13 18:00:50 UTC
Assigning over to Group B lead to coordinate backporting of all PRs on this bug's blocker bug. Ultimately it spans multiple components so just assigning as it had the majority of the work. Fixes are in: https://github.com/openshift/cluster-kube-apiserver-operator/pull/825 https://github.com/openshift/cluster-openshift-apiserver-operator/pull/351 https://github.com/openshift/cluster-kube-controller-manager-operator/pull/395 https://github.com/openshift/cluster-kube-scheduler-operator/pull/239 Verified with OCP 4.3.0-0.nightly-2020-05-13-220846, steps see below,
$ ns="openshift-kube-apiserver"
$ podname=$(oc get pods -n $ns | grep kube-apiserver | head -1 | cut -d " " -f1)
$ oc get pod -n $ns $podname -o json | jq .spec.containers[0].resources
{
"requests": {
"cpu": "300m",
"memory": "1Gi"
}
}
expected cpu request is "300m". Per https://bugzilla.redhat.com/show_bug.cgi?id=1812583#c2 ,and then the remaining components should have requests that consume no more than 270m,
- Check all kube-apiserver pods,
$ oc adm top pods -n openshift-kube-apiserver
NAME CPU(cores) MEMORY(bytes)
kube-apiserver-6fmwc-m-0.c.openshift-qe.internal 275m 1066Mi
kube-apiserver-6fmwc-m-1.c.openshift-qe.internal 180m 928Mi
kube-apiserver-6fmwc-m-2.c.openshift-qe.internal 188m 966Mi
- Check all requests on each master are less than 3 cores(3000m) as expected.
$ oc describe nodes -l node-role.kubernetes.io/master= | grep -i Allocated -A 5
Allocated resources:
(Total limits may be over 100 percent, i.e., overcommitted.)
Resource Requests Limits
-------- -------- ------
cpu 1592m (45%) 0 (0%)
memory 4279Mi (30%) 512Mi (3%)
--
Allocated resources:
(Total limits may be over 100 percent, i.e., overcommitted.)
Resource Requests Limits
-------- -------- ------
cpu 1712m (48%) 0 (0%)
memory 5049Mi (36%) 512Mi (3%)
--
Allocated resources:
(Total limits may be over 100 percent, i.e., overcommitted.)
Resource Requests Limits
-------- -------- ------
cpu 1672m (47%) 0 (0%)
memory 4659Mi (33%) 512Mi (3%)
All is well, so moved to verified.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2020:2129 |