OCP 4.10 Release Notes Tracker

[stevsmit]

Please leave comments here for anything that should be highlighted in the 4.10 release notes. If possible, provide a link to the Jira or BZ related to your item. Thank you!

A note about migrating to Jira
This tracker might be moved to Jira. In the event that happens, I will update accordingly and provide a direct link.

A note about known issues
All Bugzilla items that need to be documented as known issues in the release notes should have Doc Type = Known Issue and the Doc Text field populated accordingly. Before adding a known issue to this tracker:

1. Ensure the Doc Type and Doc Text fields are filled in correctly.
2. See if your issue shows up in the query we will use to populate the release notes.
3. If the fields are set correctly, and the issue is not in the query, add it here.

Resources

• Feature GA Status Tracker

[atiratree]

added support for podman config credentials in oc

• oc commands that use registry config (oc login, oc image, etc.) now obtain credentials from podman config locations if a registry entry cannot be found in the default docker config location.
• The order can be changed via REGISTRY_AUTH_PREFERENCE env variable to docker (current default - deprecated) or podman (prioritizes podman credentials over docker).
• REGISTRY_AUTH_FILE env variable can be used instead of --registry-config option

Deprecated features

• docker config location deprecated in oc. oc commands that use registry config currently obtain credentials from docker config location (usually ~/.docker/config.json) by default. This is deprecated and will be removed in 2 releases in favor of podman config locations.
• --registry-config or --to option in oc registry login command will stop accepting empty files (non existing files will still be valid use case) and writing to - (stdout) is also deprecated

PRs: openshift/oc#893, openshift/oc#983 and openshift/oc#1011

[soltysh]

oc 4.10 is the last release being built with support for rhel7, starting from 4.11 we will provide only rhel8 compatible builds. This is coming from the fact that oc is statically linking glibc. More information can be found in https://developers.redhat.com/blog/2020/10/08/migrating-c-and-c-applications-from-red-hat-enterprise-linux-version-7-to-version-8#

[mtulio]

On AWS OpenShift Container Platform 4.10 will prefer m6i instance types when available on the region falling back to m5 types when they're not. The sizes will keep the same: m6i.xlarge for control-plane and m6.large for compute nodes.

• PR: aws-ec2: add m6i as preferred instance type installer#5327
• Card: https://issues.redhat.com/browse/SPLAT-329

The default block storage provisioned by the installer on AWS nodes was changed from gp2 to gp3

• PR: aws: changing default EBS type on masters to gp3. installer#5239
• Cards: SPLAT-253 SPLAT-343

[ptrnull]

The release notes indicate that OpenShift 4.10 is released based on Kubernetes 1.22 but the FC release is using Kubernetes 1.23: https://openshift-release-artifacts.apps.ci.l2s4.p1.openshiftapps.com/4.10.0-fc.1/release.txt

[ctauchen] Updates for Kubernetes 1.23 are in the RN: https://docs.openshift.com/container-platform/4.10/release_notes/ocp-4-10-release-notes.html#ocp-4-10-about-this-release

[s-urbaniak]

Starting from OpenShift 4.10 control plane components have no fallback mechanisms included any more to fall back to the Common Name field on X.509 certificates when no Subject Alternative Names (SANs) are present. Any certificates must properly set the Subject Alternative Names field. This was initially announced in OpenShift 4.6, see https://docs.openshift.com/container-platform/4.6/release_notes/ocp-4-6-release-notes.html#ocp-4-6-tls-common-name.

[ctauchen] Added here: #41821

[SNiemann15]

Working together with @ktania46 and @alishaIBM to add IBM Z and Power content to the RNs:
https://docs.google.com/document/d/1LZqUJ86zmUt2Erh1mjhrACEbfBHEoE4z2Hgnr_MrugU/edit#heading=h.gi343gic2seb

PR for P/Z #40806

[jechen0648]

EgressIP on plubic cloud: https://issues.redhat.com/browse/SDN-1316 https://issues.redhat.com/browse/SDN-1315, https://issues.redhat.com/browse/SDN-1332,

[ctauchen] RN here: #39805

[duanwei33]

Strorge part:
CSI Azure Disk Driver Operator GA in 4.10 https://issues.redhat.com/browse/STOR-552
CSI OpenStack Cinder Driver Operator GA from 4.8 https://issues.redhat.com/browse/STOR-679
CSI AWS EFS Driver Operator GA in 4.10 https://issues.redhat.com/browse/STOR-362
CSI vSphere Driver Operator GA in 4.10 https://issues.redhat.com/browse/STOR-558
CSI Alicloud Disk operator GA in 4.10 https://issues.redhat.com/browse/STOR-574
CSI IBM Cloud Operator GA in 4.10 https://issues.redhat.com/browse/STOR-598
CSI Azure File Operator TP in 4.10 https://issues.redhat.com/browse/STOR-555
Automatic device discovery and provisioning with Local Storage Operator TP in 4.10

[rioliu-rh]

for config drift part, if pr#2910 cannot be merged before code freeze date Jan 28, we need to add bz2038240 to release notes as known issue

[quarterpin]

The ingress now adds support for IBM cloud environment: https://issues.redhat.com/browse/NE-619
The telemetry now includes metrics to track the number of active NLB routers in an AWS environment: https://issues.redhat.com/browse/NE-704

[xingxingxia]

1 PodSecurity is temporarily disabled in 4.10 and will be enabled in 4.11, see bug 2008462 . It is promoted to beta in 1.23, and 4.10 is k8s 1.23, so we can mention this (Discussed with Dev in Slack).
2 Add notes for OpenID IDP group consumption AUTH-8

[rioliu-rh]

for config drift part, if pr#2910 cannot be merged before code freeze date Jan 28, we need to add bz2038240 to release notes as known issue

PR is merged and BZ is verified. don't need to add it to release notes

[gpei]

For Installation in 4.10

• Allow thin provisioning for VMware vSphere IPI - https://issues.redhat.com/browse/CORS-1674
• Azure: IPI Support for Azure Stack Hub - https://issues.redhat.com/browse/CORS-1596
• IBM Public Cloud: Support IPI deployment - https://issues.redhat.com/browse/CORS-1513
• Support Alibaba IPI installer - https://issues.redhat.com/browse/SPLAT-193

[bgilbert]

coreos-installer now has iso customize and pxe customize commands for flexible customization of bare metal RHCOS installation from the live ISO/PXE images (COS-823).

[quarterpin]

List of features that are to be introduced with v4.10 for Network Edge component:
https://issues.redhat.com/browse/NE-542 - support to have ROUTER_ENABLE_COMPRESSION and ROUTER_COMPRESSION_MIME options in HAProxy
https://issues.redhat.com/browse/NE-624 - Support a Configurable syslog Maximum Message Size for HAProxy
https://issues.redhat.com/browse/NE-587 - Increase the Length of HAproxy Request URI in Logs

https://issues.redhat.com/browse/NE-505 - OpenShift Network Edge Support for Alibaba Cloud Platform
https://issues.redhat.com/browse/NE-621 - Support mutating .spec.endpointPublishingStrategy
https://issues.redhat.com/browse/NE-303 - Tech Preview introduction of ExternalDNS
https://issues.redhat.com/browse/NE-679 - Set CoreDNS forwarding policy via the DNS operator

https://issues.redhat.com/browse/NE-684 - Add support to update upstream resolvers in the pod [coreDNS]
https://issues.redhat.com/browse/NE-367 - [Observability] Add logLevel and operatorLogLevel APIs for DNS

[kasturinarra]

Scheduler policy is removed in 4.10 we could add that to the table. @ingvagabund could you please help confirm ? Thanks !!

[ingvagabund]

Yes, the scheduler Policy API was removed in 4.10.

[kasturinarra]

I think we should mention that "Simplified disconnected mirroring with oc" is going to be TP release in 4.10. @afflom @zhouying7780 could you please help confirm ? thanks !!

[stevsmit]

@afflom @zhouying7780 please confirm on #41590 if this is accurate. Thank you. cc @kasturinarra

[adambkaplan]

Storage:

Tech Preview feature - Shared Resource CSI Driver. This allows Secrets and ConfigMaps to be securely shared across namespaces. https://issues.redhat.com/browse/BUILD-293

[ctauchen] @rolfedh is taking care of this one.

[adambkaplan]

Builds:

Tech Preview - Builds can mount csi volumes. This can be used in conjunction with the Shared Resource CSI driver to use the cluster's Simple Content Access certificate within an OpenShift build. https://issues.redhat.com/browse/BUILD-274

[ctauchen] @rolfedh is taking care of this one.

[adambkaplan]

Jenkins (?):

The Jenkins Agent can be run as a sidecar container, allowing any container image to be run in a Jenkins pipeline with a correctly configured pod template and Jenkinsfile. https://issues.redhat.com/browse/JKNS-132

[ctauchen] @rolfedh is taking care of this one.

[chao007]

Storage:
IBM Cloud CSI Driver Operator should update to IBM VPC Block CSI Driver Operator

[ctauchen] Covered here: #38258

[ctauchen]

IBM Cloud CSI Driver Operator

@chao007 I'm having trouble finding information about this. Anything you can share to point me in the right direction? Thanks!

[sunzhaohua2]

Machine API part:

Cloud Controller Manager (CCM) for GCP (Tech Preview)-https://issues.redhat.com/browse/OCPCLOUD-976
Cloud Controller Manager (CCM) for vSphere(Tech Preview)-https://issues.redhat.com/browse/OCPCLOUD-1160-
Cloud Controller Manager (CCM) for Alibaba(GA)-https://issues.redhat.com/browse/OCPCLOUD-1122-

Alibaba integrate MAPI provider into OpenShift-https://issues.redhat.com/browse/OCPCLOUD-709
IBM Cloud integrate MAPI provider into OpenShift-https://issues.redhat.com/browse/OCPCLOUD-1229
Azure Stack Hub integrate MAPI provider into OpenShift -https://issues.redhat.com/browse/OCPCLOUD-1155

Enabling accelerated networking from the machine-api on Azure-https://issues.redhat.com/browse/OCPCLOUD-1203
Worker nodes with Ephemeral OS disks on Azure-https://issues.redhat.com/browse/OCPCLOUD-1303
Azure Availability Set Support-https://issues.redhat.com/browse/OCPCLOUD-1223
Enable GPU support on GCP-https://issues.redhat.com/browse/OCPCLOUD-1256
Machine Deletion Hooks-https://issues.redhat.com/browse/OCPCLOUD-948
Rebase autoscaler and CCMs on top of kubernetes 1.23-https://issues.redhat.com/browse/OCPCLOUD-1242

---

[@jeana-redhat]
2022.02.16: Thanks! Adding status below:

Already in Rel Notes:

• New CCM platform options (Alibaba & IBM GA, GCP & vSphere TP)
  

Docs done but holding Rel Notes for a single PR:

• Azure Stack Hub integrate MAPI provider into OpenShift https://issues.redhat.com/browse/OCPCLOUD-1155 (This will probably not actually merit a rel note on top of the existing installer announcements)
• Enabling accelerated networking from the machine-api on Azure https://issues.redhat.com/browse/OCPCLOUD-1203
• Worker nodes with Ephemeral OS disks on Azure https://issues.redhat.com/browse/OCPCLOUD-1303
• Azure Availability Set Support https://issues.redhat.com/browse/OCPCLOUD-1223

Docs in progress (will also be in single Rel Note PR):

• Enable GPU support on GCP https://issues.redhat.com/browse/OCPCLOUD-1256
• Alibaba integrate MAPI provider into OpenShift https://issues.redhat.com/browse/OCPCLOUD-709 (This will probably not actually merit a rel note on top of the existing installer announcements)
• IBM Cloud integrate MAPI provider into OpenShift https://issues.redhat.com/browse/OCPCLOUD-1229 (This will probably not actually merit a rel note on top of the existing installer announcements)

Scoped as having no doc impact:

• Machine Deletion Hooks https://issues.redhat.com/browse/OCPCLOUD-948
• Rebase autoscaler and CCMs on top of kubernetes 1.23 https://issues.redhat.com/browse/OCPCLOUD-1242

[chao007]

IBM Cloud CSI Driver Operator

@chao007 I'm having trouble finding information about this. Anything you can share to point me in the right direction? Thanks!

@ctauchen We only discuss it in here #39815 (comment)
And we can see the output from ocp cluster:

oc get pods -n openshift-cluster-csi-drivers | grep operator
ibm-vpc-block-csi-driver-operator-7b9594cb4c-mdk76   1/1     Running   0          20h

[ctauchen] This has been covered. See PR from @lpettyjo here: #38258

[zhaozhanqi]

https://issues.redhat.com/browse/SDN-1817 - Ability to modify the Cluster Network MTU Post Installation
this feature should mentioned in release note. cc @brahaney

[ctauchen] RN here: #39485

[yingwang-0320]

https://issues.redhat.com/browse/SDN-1353 - OVS kernel TC Flower HW Offload with CX-5: GA
Please add this feature in release note.

[ctauchen] RN here: #42128

[anuragthehatter]

It might have covered as a sub by above various Alibaba RNs but worth to mention
OVN support for Alibaba Platform https://issues.redhat.com/browse/SDN-1744
and OpenshiftSDN support for Alibaba Platform https://issues.redhat.com/browse/SDN-2366

[yunjiang29]

@ctauchen @stevsmit Bug 2051333 - [aws] records in public hosted zone and BYO private hosted zone were not deleted. will not be fixed in 4.10 GA, please add it to the release note. thanks

sample description: If you defined pre-existing Route53 private hosted zone when you deploy the cluster, the records in Route53 private and public hosted zone cannot be deleted when destroying the cluster.

cc @staebler

[yunjiang29]

@ctauchen @stevsmit Bug 2051334 - [AWS GovCloud] Cluster can not be installed on AWS GovCloud regions via terminal interactive UI. will not be fixed in 4.10 GA, please add it to the release note. thanks

sample description: Installing cluster into GovCloud regions from terminal interactive UI is not supported.

@staebler please confirm if terminal interactive UI is suitable here, thanks.

[mjpytlak]

@ctauchen @stevsmit Bug 2051334 - [AWS GovCloud] Cluster can not be installed on AWS GovCloud regions via terminal interactive UI. will not be fixed in 4.10 GA, please add it to the release note. thanks

sample description: Installing cluster into GovCloud regions from terminal interactive UI is not supported.

@staebler please confirm if terminal interactive UI is suitable here, thanks.

@yunjiang29 All supported versions of the OCP GovCloud doc, as well as the 4.10 doc, state that you must manually create/configure the install-config.yaml file to deploy a cluster. I believe adding a release note stating that installation program (terminal interactive UI) cannot be used might confuse matters for our end users. While the bug you reference is something that we might want to track for a possible enhancement, I do not believe a release note is required.

[staebler]

@ctauchen @stevsmit Bug 2051333 - [aws] records in public hosted zone and BYO private hosted zone were not deleted. will not be fixed in 4.10 GA, please add it to the release note. thanks

sample description: If you defined pre-existing Route53 private hosted zone when you deploy the cluster, the records in Route53 private and public hosted zone cannot be deleted when destroying the cluster.

cc @staebler

I have not been able to replicate this issue.

[bscott-rh]

@ctauchen @stevsmit Bug 2051333 - [aws] records in public hosted zone and BYO private hosted zone were not deleted. will not be fixed in 4.10 GA, please add it to the release note. thanks
sample description: If you defined pre-existing Route53 private hosted zone when you deploy the cluster, the records in Route53 private and public hosted zone cannot be deleted when destroying the cluster.
cc @staebler

I have not been able to replicate this issue.

@yunjiang29 In case we do need to add it to the release notes, how does this look:

There is a known issue with this release relating to AWS Route53 DNS records. If you define a private hosted Route53 zone when deploying a cluster, the Route53 records in public and private hosted zones cannot be deleted when destroying the cluster.

I will hold off on adding this to the release notes until the known issue is fully verified.

[yunjiang29]

@ctauchen @stevsmit Bug 2051334 - [AWS GovCloud] Cluster can not be installed on AWS GovCloud regions via terminal interactive UI. will not be fixed in 4.10 GA, please add it to the release note. thanks
sample description: Installing cluster into GovCloud regions from terminal interactive UI is not supported.
@staebler please confirm if terminal interactive UI is suitable here, thanks.

@yunjiang29 All supported versions of the OCP GovCloud doc, as well as the 4.10 doc, state that you must manually create/configure the install-config.yaml file to deploy a cluster. I believe adding a release note stating that installation program (terminal interactive UI) cannot be used might confuse matters for our end users. While the bug you reference is something that we might want to track for a possible enhancement, I do not believe a release note is required.

@mjpytlak, you are right, user should create install-config.yaml manually, release note is not required for this issue.

[yunjiang29]

@ctauchen @stevsmit Bug 2051333 - [aws] records in public hosted zone and BYO private hosted zone were not deleted. will not be fixed in 4.10 GA, please add it to the release note. thanks
sample description: If you defined pre-existing Route53 private hosted zone when you deploy the cluster, the records in Route53 private and public hosted zone cannot be deleted when destroying the cluster.
cc @staebler

I have not been able to replicate this issue.

@yunjiang29 In case we do need to add it to the release notes, how does this look:

There is a known issue with this release relating to AWS Route53 DNS records. If you define a private hosted Route53 zone when deploying a cluster, the Route53 records in public and private hosted zones cannot be deleted when destroying the cluster.

I will hold off on adding this to the release notes until the known issue is fully verified.

bscott-rh release note is not required for this issue, sorry about the confusing information.

[rioliu-rh]

config drift monitor is not part of machine-api, it is new feature of machine-config-operator

[openshift-bot]

Issues go stale after 90d of inactivity.

Mark the issue as fresh by commenting /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.
Exclude this issue from closing by commenting /lifecycle frozen.

If this issue is safe to close now please do so with /close.

/lifecycle stale

[openshift-bot]

Stale issues rot after 30d of inactivity.

Mark the issue as fresh by commenting /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.
Exclude this issue from closing by commenting /lifecycle frozen.

If this issue is safe to close now please do so with /close.

/lifecycle rotten
/remove-lifecycle stale

[openshift-bot]

Rotten issues close after 30d of inactivity.

Reopen the issue by commenting /reopen.
Mark the issue as fresh by commenting /remove-lifecycle rotten.
Exclude this issue from closing again by commenting /lifecycle frozen.

/close

[openshift-ci]

@openshift-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.

Reopen the issue by commenting /reopen.
Mark the issue as fresh by commenting /remove-lifecycle rotten.
Exclude this issue from closing again by commenting /lifecycle frozen.

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.