What_s_New_in_OpenShift_Container_Platform_4.6.pdf

1
● Please direct your Q&A into the Q&A forum within Primetime or openshift.tv comments
● Any outstanding questions will be addressed at the end of the presentation or responses
will be facilitated after the brieﬁng
● This call is being recorded. The slide deck, recording, and Q&A will be provided after the call
What’s New in OpenShift 4.6 Field Brieﬁng

Table of Contents
● Introductory Content
● OpenShift 4.6 Spotlight Features
● Hosted OpenShift
● A broad ecosystem of workloads
● Cloud Native Development
○ Service Mesh
○ Serverless
○ Helm
○ Pipelines
○ GitOps
○ Code Ready / Dev Tools
● OpenShift Console
● Observability
● Core Platform
○ Install and upgrades
○ Control plane
○ RHEL CoreOS
○ Networking and Routing
○ Storage
● Telco
● Security and Compliance
● Multi-arch and Windows Containers

3
What’s New in OpenShift 4.6
OpenShift Product Management

What's new in OpenShift 4.6
NEW INSTALLER
PLATFORMS
WORKLOAD
STABILITY
CORE
PLATFORM
Bare Metal IPI
AWS and Azure Gov Clouds
Disconnected Update Intelligence
Remote Worker Nodes
Real Time and Low Latency
OVN (GA)
Compliance Operator (GA)
User Workload Monitoring (GA)
Log Forwarding API (GA)
Serverless Eventing (GA)
OpenShift 4.6
5

OpenShift Roadmap
APP
DEV
PLATFORM
APP
DEV
● OpenShift Pipelines (Tekton) GA
● OpenShift Builds (v2) TP
● Jenkins Operator TP
● Argo CD GA
● Schema based forms for Event Sources
● Improvements to GitOps experience
● Cluster Update Compatibility Checks
● Hybrid Operators with Operator-SDK
● Simplify Operator Lifecycle interactions
● IPv6 (single/dual stack on control plane)
● Enable user space pod int & API Library
● Utilize cgroups v2
● Azure Stack Hub support
● AWS C2S and China support
● Equinox Packet support
● IBM Cloud support
● Assisted Installer
● Network Enhancements derived from OVN
● Local storage support in OCS
● OpenShift Service Mesh Federation
● RHV UPI support
● GPU Sharing
OpenShift 4.7/4.8
● OSD GCP CCS & private clusters
● OSD CCS on-demand Marketplace billing
● OSD cluster autoscaling
● OSD custom domains, log forwarding
● ACM integration
● OSD / AMRO PCI Certification
H1 2021
● Improved getting started experience for devs
● OpenShift Serverless Eventing GA
● OpenShift Pipelines (Tekton) TP
● Jenkins Operator TP
● Monitor application workloads (GA)
● Operator dependency tools v2
● OpenShift Builds (v2) TP
OpenShift 4.6
● Amazon Red Hat OpenShift
● ARO Government (MAG) support
● OSD / AMRO Upgrade Scheduling
● OSD / AMRO Machine Pools
● AMRO Auto Scaling, BYO VPC
● BYOK disk encryption (AWS, Azure)
Q4 2020
● OVN GA, OVN Egress Firewall/Router/IP
● Bare metal (IPI) GA
● Remote worker nodes for Edge
● Realtime kernel (TP, RAN use-cases only)
● AWS GovCloud support
● Microsoft Azure Government (MAG) support
● VMware vSphere 7.0 support
● Improved cloud credential handling
● Disconnected OpenShift Update Service
● GCP & Azure spot instances
● CSI resize/snapshot GA
● Windows containers GA
● OAuth secure storage & inactivity timeout
● Enhanced RHCOS static networking UX
● Compliance Operator
PLATFORM
OpenShift 4.Next
APP
DEV
PLATFORM
MANAGED
H2 2021
● OpenShift Single node
● Utilize cgroups v2
● Microsoft Hyper-V (UPI) support
● Alibaba Cloud support
● Network Enhancements derived from OVN
● Local storage support in OCS
● OpenShift Service Mesh Multi-Cluster
● Next gen SmartNic architecture
● OSD / AMRO FedRAMP Certification
● Build, Operate, Transfer operational
model
● Windows containers
● GPU optimized VMs
● Workload Metrics Visualization
● Operator SDK: Python and Java Support
● Operators install/upgrade as a group
● Serverless Streaming
● Console integration with Tekton Hub
● Pipelines Notifications
● OpenShift Builds (v2) GA
● Jenkins Operator GA
MANAGED
MANAGED
7

What's New in OpenShift 4.6
Extended Update Support
OpenShift EUS and Layered Product and Add-ons
4.6 EUS
2020 2021 2022
MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG
Duration of the Platform EUS
Add-ons have a version that
is guaranteed to work for
Platform EUS OpenShift Logging
OpenShift Container Storage
Advanced Cluster Manager
OpenShift Serverless
OpenShift Pipelines
OpenShift Service Mesh
10
LAYERED UPGRADE
LAYERED UPGRADE
LAYERED UPGRADE
LAYERED UPGRADE
LAYERED UPGRADE

Kubernetes 1.19
Scheduling
● Customize the behavior of the Kube-scheduler
● Scheduler Profiles
● Pod Topology Spread constraints
Control Plane & Security
● Automatically track and act on the features not making
Stable
● Warning mechanism for use of deprecated APIs
● AppProtocol to Services and Endpoints
● Kubelet Client TLS Certificate bootstrap and rotation
● NodeRestriction admission controller
Misc
● Structured Logging proposal
CRI-O
1.19
Kubernetes
1.19
OpenShift
4.6
Blog: https://www.openshift.com/blog/kubernetes-1.19-arrives
11
Storage
● Immutable Secrets and ConfigMaps
● CSI Storage Capacity management (alpha)

12
OpenShift 4.6 Spotlight Features

13
*External Load Balancers (routing) and external DNS servers are provided by the user
OpenShift install
Bare Metal Nodes
OpenShift Cluster
▸ Installer provisions
• Networks
• Internal load balancers *
• Internal DNS *
• Red Hat CoreOS installation
• CoreOS ignition configs
• OpenShift nodes
• OpenShift cluster resources
Full stack automation (IPI) installation on Bare Metal
Deploying Red Hat OpenShift on Bare Metal on Installer-Provisioned Infrastructure (IPI)
Product Manager: Ramon Acedo Rodriguez
OpenShift on Bare Metal

14
OpenShift on Bare Metal
Full stack automation (IPI) installation on Bare Metal
Deploying Red Hat OpenShift on Bare Metal on Installer-Provisioned Infrastructure (IPI)
apiVersion: v1
basedomain: <domain>
metadata:
name: <cluster-name>
networking:
machineCIDR: <public-cidr>
networkType: OVNKubernetes
compute:
- name: worker
replicas: 2
controlPlane:
name: master
replicas: 3
platform:
baremetal: {}
platform:
baremetal:
apiVIP: <api-ip>
ingressVIP: <wildcard-ip>
provisioningNetworkInterface: <NIC1>
provisioningNetworkCIDR: <CIDR>
hosts:
- name: openshift-master-0
role: master
bmc:
address: ipmi://<out-of-band-ip>
username: <user>
password: <password>
bootMACAddress: <NIC1-mac-address>
hardwareProfile: default
- name: openshift-master-1
role: master
bmc:
address: ipmi://<out-of-band-ip>
username: <user>
password: <password>
bootMACAddress: <NIC1-mac-address>
hardwareProfile: default
Bare Metal Management
Powered by Metal3
and OpenStack Ironic under
the hood
Host Power Management
Redﬁsh, IPMI, iDrac, iLo.
Provisioning over the network
Installation over DHCP/PXE or Virtual Media
Disconnected Installations
RHCOS image cache and disconnected registry
Metal3
OpenStack Ironic

AWS GovCloud
Deploy OpenShift to AWS GovCloud regions
● Government customers and their Partners can now deploy
OpenShift to the AWS GovCloud ‘US-East’ & ‘US-West’
regions.
● AWS GovCloud (US) is specifically designed for US
government agencies at the federal, state, and local level, as
well as contractors, educational institutions, and other U.S.
customers that need to run sensitive workloads in the cloud.
● RHEL CoreOS AMI publishing is not available in the
GovCloud regions, so users must upload their own prior to
installing OpenShift via:
○ ‘aws ec2 import-snapshot’ & ‘aws ec2 register-image’
● Installation of OpenShift on AWS GovCloud is similar to
existing deployment methods for other AWS regions, but
the AWS region and RHEL CoreOS AMI ID must be manually
configured in install-config.yaml.
Generally Available
Product Manager: Katherine Dubé
15
% aws ec2 describe-regions --output text
REGIONS ec2.us-gov-west-1.amazonaws.com opt-in-not-required us-gov-west-1
REGIONS ec2.us-gov-east-1.amazonaws.com opt-in-not-required us-gov-east-1
% grep -B 1 -A 2 "aws:" mycluster/install-config.yaml
platform:
aws:
region: us-gov-west-1
amiID: ami-9dbf86fc
% ./openshift-install create cluster --dir mycluster
INFO Credentials loaded from default AWS environment variables
INFO Consuming Common Manifests from target directory
INFO Consuming Worker Machines from target directory
INFO Consuming Openshift Manifests from target directory
INFO Consuming OpenShift Install (Manifests) from target directory
INFO Consuming Master Machines from target directory
INFO Creating infrastructure resources…
INFO Waiting up to 20m0s for the Kubernetes API at
https://api.mycluster.example.com:6443...
INFO API v1.19.0+f5121a6 up
INFO Waiting up to 30m0s for bootstrapping to complete...
INFO Destroying the bootstrap resources...
INFO Waiting up to 40m0s for the cluster at https://api.mycluster.example.com:6443
to initialize...
INFO Waiting up to 10m0s for the openshift-console route to be created...
INFO Install complete!
INFO To access the cluster as the system:admin user when using 'oc', run 'export
KUBECONFIG=/Users/userid/openshift-install/mycluster/auth/kubeconfig'
INFO Access the OpenShift web-console here:
https://console-openshift-console.apps.mycluster.example.com
INFO Login to the console with user: "kubeadmin", and password:
"5char-5char-5char-5char"
INFO Time elapsed: 40m10s

Microsoft Azure Government (MAG)
Deploy OpenShift to Microsoft Azure Government
● Government customers and their Partners can now
deploy OpenShift to the Microsoft Azure Government
(MAG) dedicated instance.
● MAG is comprised of six government-only datacenter
regions, all granted an Impacted Level 5 Provisional
Authorization.
● Installation of OpenShift to MAG is similar to existing
deployment methods for other Azure regions, but the
‘cloudName’ field must be set to
‘AzureUSGovernmentCloud’ in the install-config.
Generally Available
% az cloud set --name AzureUSGovernment
Switched active cloud to 'AzureUSGovernment'.
Active subscription switched to 'Production (291bba3f-e0a5-47bc-a099-3bdcb2a50a05)'.
% az account list-locations -o table
DisplayName Name RegionalDisplayName
-------------- ------------- ---------------------
Global global Global
USDoD Central usdodcentral (US) USDoD Central
USDoD East usdodeast (US) USDoD East
USGov Arizona usgovarizona (US) USGov Arizona
USGov Iowa usgoviowa (US) USGov Iowa
USGov Texas usgovtexas (US) USGov Texas
USGov Virginia usgovvirginia (US) USGov Virginia
% ./openshift-install explain installconfig.platform.azure.cloudName
RESOURCE: <string>
cloudName is the name of the Azure cloud environment which can be used to configure the Azure SDK
with the appropriate Azure API endpoints. If empty, the value is equal to "AzurePublicCloud".
% export AZURE_AUTH_LOCATION=/Users/userid/.azure/osServicePrincipal-mag.json ; ./openshift-install
create cluster --dir mycluster
INFO Credentials loaded from file "/Users/userid/.azure/osServicePrincipal-mag.json"
INFO Consuming Common Manifests from target directory
INFO Consuming Worker Machines from target directory
INFO Consuming Openshift Manifests from target directory
INFO Consuming OpenShift Install (Manifests) from target directory
INFO Consuming Master Machines from target directory
INFO Creating infrastructure resources…
INFO Waiting up to 20m0s for the Kubernetes API at https://api.mycluster.example.com:6443...
INFO API v1.19.0+f5121a6 up
INFO Waiting up to 30m0s for bootstrapping to complete...
INFO Destroying the bootstrap resources...
INFO Waiting up to 40m0s for the cluster at https://api.mycluster.example.com:6443 to initialize...
INFO Waiting up to 10m0s for the openshift-console route to be created...
INFO To access the cluster as the system:admin user when using 'oc', run 'export
KUBECONFIG=/Users/userid/openshift-install/mycluster/auth/kubeconfig'
https://console-openshift-console.apps.mycluster.example.com
INFO Login to the console with user: "kubeadmin", and password: "5char-5char-5char-5char"
INFO Time elapsed: 40m10s
16

17
Update manager for your clusters in restricted or disconnected
networks
● OpenShift Update Service (OSUS) is the on-premise release of
Red Hat’s hosted update service
● Supports the publishing of upgrade graph information to clusters in
restricted networks
● Provides clusters with a list of next recommended update versions
based on the current version installed on the cluster
● Comprised of two services:
○ Graph Builder: Fetches OpenShift release payload information
(primary metadata) from any container registry (compatible with
Docker registry V2 API) and builds a directed acyclic graph (DAG)
representing valid upgrade edges
○ Policy Engine: Responsible for selectively serving updates to
every cluster by altering a client’s view of the graph with a set of
filters
● GA release planned for post-4.6 and will be distributed on
Operator Hub as an optional add-on operator
● Blog post announcing OpenShift Update Service
OpenShift Update Service
Local Container
Registry in
Restricted Network
OpenShift
Update
Service
Graph Builder
Policy Engine
OpenShift Cluster
in Restricted Network
Cluster Version
Operator (CVO)
Scrape Release
Images from
Registry
Read graph data
(secondary
metadata)
Edge
Add/Remove
Cluster Version
Operator (CVO)
OpenShift Cluster
in Restricted Network
Generally Available

Speciﬁcations for Remote Worker Nodes
Zone-1
S
Zone-2
W
W
Zone-3
W
W
W
Red Hat OpenShift
Supervisors reside in a
central location, with
reliably-connected
workers distributed at
edge sites sharing a
control plane.
SUPERVISORS
WORKER
Tolerant of disruption
● Admin can configure status
update frequency
● Zones with disruption
budget
● Tolerations
● DaemonSet & Static Pods
stay running
S
W
Product Manager: Tushar Katarki
18
W

Open Virtual Network (OVN)
● Next-gen Kubernetes CNI plugin (ovn-kubernetes)
● OCP 4.6 GA (non-default, default TBD)
● Install-time option or post-install (bare metal only) migration
Why?
● Consolidates Red Hat SDN efforts across products
● Advanced Telco and enterprise-grade features
● Flexible SDN architecture for faster feature development
● Large upstream community (Linux Foundation project)
● Red Hat leadership in upstream OVS & OVN communities
● Manages overlays and physical network connectivity
● Flexible security policies via ACLs and security groups
● Distributed L3 routing, L2/L3 Gateways to other networks
● IPv4 and IPv6 capability
● Integration with TOR and other "physical" gateways
● Native support for NAT, load balancing and IPAM
● Windows “Hybrid Overlay” service for pod-to-pod traffic between
Windows and Linux cluster nodes.
OpenShift SDN OVN Kubernetes
veth pairs veth pairs
OVS bridge OVS bridge
Central controller / host-ipam Central controller / host-ipam
VXLAN tunnels Geneve tunnels
OVS flows for NetworkPolicy OVS flows for NetworkPolicy
IPTables for services OVN LBs for services
IPTables for NAT OVS for NAT
Product Manager: Marc Curry
19
Goal: Develop and support a modern, maintainable, community-based, open-source Kubernetes CNI network plugin for OpenShift
that complements the existing capabilities of OVS to add native support for virtual network abstractions.
Technology Highlights Comparison

OpenShift Compliance Operator: Declarative Security Compliance
=
Install, upgrade,
reconcile, conﬁg
Describe intent
with declarative
conﬁg
Monitor, scale,
troubleshoot,
backup
Summarize
Observe
ComplianceSuite
Scan (results)
1 A compliance profile is
selected
2
The operator runs the scan
for the profile against
nodes, collect results, and
(optionally) performs
remeditations
3 Accreditors or Auditors can
examine the scan results
for compliance status,
After review, if desired,
remediations can be
manually applied by the
cluster-admin.
ComplianceCheckResult
ComplianceRemediations
Security and Compliance
Product Manager: Kirsten Newcomer
With 4.6, a limited set of RHCOS checks will
be implemented. Additional compliance
checks will be delivered roughly every 2
months.
20

Leverage our existing Monitoring infrastructure to
monitor your own workloads.
● Enable a dedicated monitoring stack managed by us.
● Configure monitoring for your custom services or
infrastructure services not covered by the out-of-the-box
cluster monitoring stack.
● Access metrics and alert information through a single,
multi-tenant interface.
○ Note: You can explore and manage both from the
developer perspective inside the OpenShift Console.
● Not in scope for this release are things like adding your own
dashboards to the console, creating new rules inside
platform-specific namespaces (e.g. openshift-*),
tenant-based routing configuration for Alertmanager, and a
few more.
● Monitoring your sample application Quick Start available to
show users how to access basic monitoring features
Product Manager: Christian Heidenreich
21
Monitor your own services
Generally Available
1. Enable dedicated monitoring by setting ‘enableUserWorkload’ to ‘true’ inside
the cluster-monitoring-config ConfigMap.
apiVersion: v1
kind: ConfigMap
metadata:
name: cluster-monitoring-config
namespace: openshift-monitoring
data:
config.yaml: |
enableUserWorkload: true
2. Configure a ServiceMonitor CR inside a user-defined namespace where app is
running that exposes a /metrics endpoint.
3. Go to the Developer Perspective, switch to your namespace and look for your
metrics (it can take a bit time to have our infra picking up everything)

Abstract Fluentd configuration by introduce new log
forwarding API to improve support and experience for
customers.
● Introduce a new, cluster-wide ClusterLogForwarder CRD (API)
that replaces needs to configure log forwarding via Fluentd
ConfigMap.
● The API helps to reduce probability to misconfigure Fluentd
and helps bringing in more stability into the Logging stack.
● Features include: Audit log collection and forwarding, Kafka
support, namespace- and source-based routing, tagging, as
well as improvements to the existing log forwarding features
(e.g. syslog RFC5424 support).
● WARNING: We will not automagically migrate old Tech
Preview CRs into a GA CR.
Infra
App
Audit
Forward logs to
different systems
based on their
“inputSource”.
inputSource=app
inputSource=audit
apiVersion: "logging.openshift.io/v1"
kind: "ClusterLogForwarder"
spec:
outputs:
- name: MyLogs
type: Syslog
syslog:
Facility: Local0
url: localstore.example.com:9200
pipelines:
- inputs: [Infrastructure,
Application, Audit]
outputs: [MyLogs]
Introduce new log forwarding API
Generally Available
23

CONFIDENTIAL designator
V0000000
Eventing
■ Brokers
✓ Built-in Event Filtering
✓ Routing based on event types or attributes
✓ Multiple event types
✓ Multi-tenant
■ Channels
✓ Event Fanout to multiple subscribers
✓ Same event type
✓ Single-tenant
Generally Available
Coming with OpenShift Serverless 1.11
24
Product Manager: William Markito & Naina Singh

V0000000
Eventing User Experience
Generally Available
25
Camel-K Connectors
● Connect your applications
with AWS Kinesis, AWS SQS,
Slack, JIRA, Telegram,
SalesForce and more...
Red Hat AMQ Streams
● Integration with Apache
Kafka for reliable event
delivery with Channels
and Broker support.

26
Robust. Proven. Award-winning.

Red Hat Advanced Cluster Management for Kubernetes
Multi-cluster lifecycle
management
Policy driven governance,
risk, and compliance
Advanced application
lifecycle management
Observability for your Clusters and Apps
● GA provisioning of OpenShift on vSphere
● GA provisioning of OpenShift on Bare Metal
● Open Source Policy Repository
● Enhanced OPA integration
● Simplified Application Experience
● Portfolio Integration with Ansible Automation Platform -
● Cluster Health monitoring with Thanos
● Multi-cluster health optimization with Grafana
What’s new with 2.1
27

F18017-190601
RHACM Hub
Managed Clusters
28
Integration Architecture Overview for Application Life Cycle
Red Hat Openshift Platform
RHACM Klusterlet
Red Hat Openshift
Platform
Red Hat Ansible
Automation Platform IT Systems
Security
Network
Application
CM
APP A
APP A
Kubernetes
resources
Channel
1
2
3
4
2
Kubernetes Job
1
3
4
Managed Clusters install resources
based on channel it subscribed
ACM hub call Ansible Tower with
Template Job ID deﬁne in
Application Pre & Post Action
Ansible Tower executes Job
ACM hub receives feedback from
Job execution and show all
Kubernetes resources in topology
including Ansible Job status
Pre &
Post
+

29
Managed OpenShift
Get the best of OpenShift without being on call

31
31
New Managed OpenShift Pricing
Product Managers: Patrick Strick, Jacob Lucky, Andrew Cathrow
WORKER NODES
MULTI-AZ
SINGLE-AZ
4 vCPU SUBSCRIPTION PRICE
On-demand (hourly) $0.171
1 Year $1,000
3 Year $2,000
4 vCPU
24x7 Premium Support
99.95% Uptime SLA
$0.03 per hour
New Minimum Cluster Size (OSD)
vCPU Based Pricing
Cluster Fee
https://www.openshift.com/pricing/

32
New Feature Highlights
● UI for cluster upgrade scheduling
● Custom Machine Pools (AZ aware
Machine Sets)
● Customer notifications tied to Cluster
History Log
● BYOK Disk Encryption on AWS CCS
32
OpenShift Dedicated & Amazon Red Hat OpenShift
Product Manager: Patrick Strick and Andrew Cathrow

Microsoft Azure Government (MAG)
○ Deploy managed OpenShift clusters on
Azure’s government cloud
Egress lockdown
○ Documented outbound IP/DNS
requirements to secure outbound traffic via
firewall
BYOK disk encryption for PV’s and OS disk
Larger VM sizes, including dedicated instances
Cluster create GUI in Azure Portal
Azure Red Hat OpenShift
Product Manager: Jacob Lucky
33

34
A broad ecosystem of workloads
Services allow for a
SaaS experience on your own infrastructure
Relational DBs
NoSQL DBs
Storage
Messaging
Security
Monitoring
AL/ML
Big Data
DevOps

New Operator Bundle Format
Product Manager: Daniel Messer
The Bundle format uses standard container technology for
shipping the metadata and allows developers to publish their
own Operator update streams in catalogs. This is very similar to
how OCI artifact spec plans to ship non-runnable image artifacts
through registries.
Changes to building custom catalogs
● Using opm was optional, now it is mandatory
● Much easier UX to add/remove/update catalog content
OpenShift now has per-version Operator catalogs
● Teams can ship to very intentional ranges of OCP versions
● 4.1 to 4.5 will continue to share a single catalog
35
Operator objects:
Deployment/STS, Roles, RoleBindings, ServiceAccount,
CRDs
Metadata:
icon, channels, dependencies, related images, CR
examples, links
Operator Bundle
Supplemental objects:
ConﬁgMap, Secrets, HPA, PDBs SCCs, PriorityClass, ...
Operator Lifecycle:
Full OLM feature set
Simplified Lifecycle:
Create & Recreate
Drives Resolution, Updates and
Catalog UIs
opm index add
--bundles quay.io/username/my-bundle:0.0.1 # add this bundle
--tag quay.io/username/my-index:1.0.0 # to this catalog

● Helm 3.3 GA
● Support for multiple Helm
repositories in Developer Catalog
● Select chart version on install
● Form-based values.yaml
● Displays charts compatible with
OpenShift version (kubeVersion)
Product Manager: Karena Angell
Helm 3 on OpenShift 4.6
36

V0000000
Red Hat Application Services
38
Red Hat Runtimes
● Quarkus - GA of Native Compilation Support, OpenShift Extension GA and new Spring compatibilities
● Data Grid 8.1 - Cross-site cluster support and auto-scaling on OpenShift
● Red Hat Build of OpenJDK Support for the Java Flight Recorder - OpenJDK 8
● Spring Boot 2.2 - New AMQ Starters, GA of Reactive support and Kubernetes Java annotations.
Red Hat Integration
● 3scale API Management - Improved manageability with operator for Air-Gapped deployment, Monitoring
& backup/restore. Accelerated API performance with content caching, and new policies for API Gateway.
● Fuse - Air-Gapped deployment, OpenShift AuthN/AuthZ for Console, and Spring Boot 2 support for Fuse
on OpenShift.
● Camel K for Serverless (TP) - now integrated to OpenShift Developer Console to leverage the huge
Camel connector catalog for apps based on Camel K and Knative Eventing.
Red Hat Process Automation
● OptaPlanner - Support for new rotation screen in Optaweb Employee Rostering
● Dashboard Builder - Stand alone Dashbuilder: Support for multiple dashboards, Runtime
REST api, React components
Product Manager: Karena Angell (on behalf of the Red Hat Application Services team)
Events
APIs EIPs
Data

V0000000
Migration Toolkit for Applications
39
● Review Java Apps - review source code or
decompile binaries and find ways to make them
more JEE compliant, and container friendly.
● OpenJDK, Container and Linux rules -
discover fixes to be applied to your app to
increase its mobility
● Camel 2 to 3 Rules - review your Camel 2 rules
and find out how to convert them to Camel 3
(more container friendly).
● Web,CLI, Maven and IDE - use the tool in any
your preferred context, from CI/CD pipelines , to
maven builds and in within your development
environment. Easy to deploy on OpenShift.
Product Manager: Miguel Pérez Colino
MTA 5.0 Launched
red.ht/mta

Modernized workloads, support mixed applications consisting of VMs, containers, and serverless
VMs Containers
Red Hat OpenShift Container Platform
Red Hat Enterprise Linux CoreOS
Physical machine
OpenShift Virtualization
40
What’s new in OpenShift Virtualization (2.5)
Core
● Deploy CNV on a subset of cluster nodes
● Import from VMware - cold or offline migration
● Robust VM baseline performance
Network
● Support of bonding modes 2 (balance-xor) and 4 (802.3ad)
● Added CNI certification test suite for VMs
Storage
● Improved dev workflow with default OS images & templates
● Fast DataVolume CDI cloning via CSI Snapshots
● Offline VM Snapshots
● Import ContainerDisks to persistent storage more efficiently
Product Manager: Peter Lauterbach, Rob Young

43
Product Manager: Jamie Longmuir and Mauricio "Maltron" Leal
OpenShift Service Mesh 2.0
Key Features & Updates
● Version 2.0 to GA in November 2020
● Upgrades Istio to version 1.6
● Simplifies architecture based on a single
Istio daemon (“Istiod”)
● Improves key and certificate rotation with
Secret Discovery Service
● Improves metrics collection with
Telemetry V2 architecture.
● Introduces WebAssembly extensions as a
“Tech Preview” feature.

● Consolidates the Istio control plane components (Pilot,
Galley, Citadel) into a single binary known as istiod.
○ Simplifies installation, upgrades and management
of the Control Plane.
○ Reduces the Control Plane’s resource usage,
startup time and improves performance.
● Secret Discovery Service (SDS) provides a more secure
and performant mechanism for delivering certificates to
Envoy side car proxies.
○ Removes the use of Kubernetes Secrets.
○ Enables 3rd party cert manager integrations.
● New Telemetry V2 architecture substantially reduces
metrics collection latency.
Istio 1.6 - Architectural Changes
44
Pilot Citadel Galley
istiod
Control Plane
Service A Service B
Envoy Envoy
Data Plane
Discovery
Configuration
Certificates
Ingress Egress
Mesh Traffic

● New ServiceMeshControlPlane resource (v2)
to simplify configuration.
● Kiali:
○ Distributed traces are visualized and
accessible in the service graph.
○ New wizards make it easier to configure
timeouts, retries and fault injection
scenarios.
● Jaeger:
○ Support for external ElasticSearch clusters.
○ OpenTelemetry collector in Tech Preview
enabling vendor-neutral instrumentation.
User Experience Enhancements
46

V0000000
OPENSHIFT SERVERLESS What's new in OpenShift 4.6
Serverless & the Portfolio
✓ OpenShift Service Mesh Support [doc]
■ Support for JWT Auth[doc]
■ Custom Domains for Knative Services[doc]
✓ OpenShift Pipelines Templates and Tasks
✓ CLI Commands for Eventing
Service
Mesh
Serverless Pipelines
50
Serverless & Pipelines Experience

V0000000
51
Powerful CLI experience
✓ Local Developer Experience
✓ Based on Buildpacks
✓ Deploy as Knative Service
✓ Project templates
✓ Support for Cloud Events/HTTP
✓ Runtimes:
Functions
$ kn faas help
Usage:
faas [command]
Available Commands:
build Build an existing Function project as an OCI image
completion Generate bash/zsh completion scripts
create Create a new Function, including initialization of
local files and deployment
delete Delete a Function deployment
deploy Deploy an existing Function project to a cluster
describe Describes the Function
help Help about any command
init Initialize a new Function project
list Lists deployed Functions
run Runs the Function locally
update Update a deployed Function
version Print version. With --verbose the build date stamp
and commit hash are included if available.
Developer Preview

V0000000
52
Functions
Developer Preview

● Pipeline templates for serverless when importing application (+Add)
● Pipeline templates use workspaces instead of PipelineResources
● Default workspace per PipelineRun or globally
● Expanded Task library
○ Helm tasks
○ Skopeo tasks
○ Trigger Jenkins jobs from Tekton
● Support for disconnected clusters
● Pipeline metrics in cluster monitoring
● Pipeline Quick Start tours in Dev Console
● Enhancements in Tekton CLI: workspaces, results, ...
Tech Preview
Product Manager: Siamak Sadeghianfar
OpenShift Pipelines 1.2*
56
* Available through the OpenShift Pipelines operator “preview” channel

● Start pipeline wizard
● Add trigger wizard
● Open Tekton docs from
YAML
● Restart pipeline action
Tekton Pipelines in
IntelliJ & Visual Studio Code
57

OpenShift GitOps
(new add-on)
Tech Preview Q4CY20
● Enable teams to adopt a declarative GitOps approach
to multi-cluster configuration and continuous delivery
● OpenShift GitOps is complementary to OpenShift
Pipelines and includes
○ Argo CD
○ GitOps Application Manager CLI
○ Integrated into Dev Console (App Stages)
● Included in OpenShift SKUs
Desired
State
Cluster
State
Observe
State
Take
Action
OpenShift
GitOps
58

Traditional and
Kubernetes-native
CI/CD
OpenShift
Builds
OpenShift
OpenShift
Pipelines
OpenShift
GitOps
Build container images
from source code using
Kubernetes tools
A Comprehensive DevOps Platform for Hybrid Cloud
Declarative GitOps for
multi-cluster
continuous delivery
61

kind: ServiceBinding
metadata:
name: binding-request
spec:
application:
name: cool-app
resource: deployments
group: apps
version: v1
services:
- group: postgresql.baiju.dev
version: v1alpha1
kind: Database
name: cool-db
Service Binding Operator
Tech Preview
● Automate configuring applications to find the
coordinates of the backing service (database, mq, etc)
○ Operator services
○ Helm Charts
○ Any k8s resource
● Injects service coordinates into Deployments,
DeploymentConfig, Knative Service and more
● Requires services to advertise injectable configuration
via annotation present on k8s resources
Deployment:
cool-app
Database CR:
cool-db
application
Service
Inject env vars
63

V0000000
Targeted for Nov 4
● Support for IBM Z (v2.4)- run on
OpenShift on IBM Z
● Single host proxy - route ingress to
all components from single host
● Support OpenShift-trusted CA
bundle (v2.4)
● Experimental support for IntelliJ
as IDE - community edition with
steps to use customer’s licensed
version
Product Manager: Parag Dave
CodeReady Workspaces 2.5
64

V0000000
Released September 24th!
Product Manager: Serena Nichols
odo 2.0 - OpenShift’s Dev-Focused CLI
65
$ odo create nodejs --starter
Start quickly using linked samples
$ odo catalog list components
Odo Devfile Components:
NAME DESCRIPTION REGISTRY
java-maven Upstream Maven and OpenJDK 11 DefaultDevfileRegistry
java-openliberty Open Liberty microservice in Java DefaultDevfileRegistry
java-quarkus Upstream Quarkus with Java+GraalVM DefaultDevfileRegistry
java-springboot Spring Boot® using Java DefaultDevfileRegistry
nodejs Stack with NodeJS 12 DefaultDevfileRegistry
Core language support via a
common/shared model with
Eclipse Che with devfile stack
definitions
$ odo catalog list services
Operators available in the cluster
NAME CRDs
etcdoperator.v0.9.4 EtcdCluster, EtcdBackup, EtcdRestore
$ odo service create
etcdoperator.v0.9.4/EtcdCluster
Works with core Kubernetes!
- Creation of operands
- Binding of services
$ odo debug
Easily connect for debugging

V0000000
OCP 4.6 update - Oct 22
● Regular releases to pick up 4.5
z-streams and fresh certs
● Resource requirements - no changes
for 4.6, worked on future
improvements
● VS Code OpenShift Connector
extended to work with starting and
using CodeReady Containers
Product Manager: Steve Speicher
CodeReady Containers: OpenShift on your Laptop
66

68
OpenShift Console
OpenShift = Kubernetes
Managing
Kubernetes
Extending
Kubernetes
Learning
Kubernetes
Developing on
Kubernetes

Guide users to recommended update
paths and available channels.
● Make it easier to find information on
channels and versions
● Provide recommended update paths
Recommendation Alerts
● Three new recommendation alerts were
added to inform users when:
○ a new patch becomes available
○ a new minor release becomes
available
○ new channels become available
Provide transparency into the update
process with an in progress checklist
● Inform on Operator and Node Progress
● Surface conditions
Over the air goodness!
Product Manager: Ali Mobrem,
Generally Available
69

Combine an “init custom resource”
creation with Operator installation flow
● Easily see the installation status with a new
"Installing..." Operator screen.
● A custom resource contains initialization
setups to be created during the Operator
installation.
Show when a k8s resource “owned by”
or “related to” an Operator / Operand
● OLM managed Operator: Easily see if the
resource is managed by the Operator or an
Operand instance.
● Cluster Operator: A list of resources
associate with the Operator.
Group Operand’s properties per CRD’s
schema structure
● Easily understand and see the spec/status
properties of the CR instance.
● Easily learn schema info on property’s
popover directly on this UI.
Managing Operators at ease
70
Product Manager: Ali Mobrem, Tony Wu

Default Perspective --and-- Guided Tour
● Non privileged users are brought to
Developer perspective by default upon initial
login
● A Guided Tour has been added to the
Developer Perspective to help with
discoverability
Getting started with samples
● Developer get started quickly with samples
Quick Starts
● Guides customers with interactive
documentation tours
● Helps customers to discover and enable
value added services
● Reduces the time it takes to get customers
up and running
● Educates users on how to maximize usage
of the UI
● Accessible on both the Administrator and
Developer perspectives
Getting started experience
Product Manager: Ali Mobrem, Serena Nichols
72

Connectivity mode
- Allows developers to focus on the
composition of their application, both on how
it’s managed as well as how things are
connected.
Parity between List & Graphical
- Display Options
- Filters
- Find
Admin’s Project-> Workload tab has an
increased feature set
Consumption mode
- Allows developers to focus solely on
components consuming resources.
- Thus, no connectors are shown (Service
Binding, Visual, Traffic, Triggers, etc), nor
groupings. Pod count is shown by default.
Application topology
73

Empower developers with visibility of
their application across all
environments
● Dedicated Application Stages view
● View all app groupings
● Drill into app grouping details to get
visibility into the composition and status
of the applications/workloads deployed
across environments
Visibility of apps across environments
74
Dev Preview

Expose selected Fluentd performance optimization
parameters in the ClusterLogging API.
● Not relevant to most users, default settings should give good
general performance.
● Ultimately we want great performance "out of the box" with
no user intervention. However, today we can't always
predict/detect the best settings; customers have had to
adjust fluentd parameters to get good performance.
● All possible settings relate to optimizing the forwarding
process, meaning when logs leave Fluentd to either our
internal storage or a configured 3rd party system.
● Settings include retries, memory usage and the flushing
output behaviour.
“Tune” Fluentd
77
1. After installing OpenShift Logging, apply the following YAML.
apiVersion: logging.openshift.io/v1
kind: ClusterLogging
metadata:
name: instance
namespace: openshift-logging
spec:
forwarder:
fluentd:
buffer:
chunkLimitSize: 8m
flushInterval: 5s
flushMode: interval
flushThreadCount: 3
overflowAction: throw_exception
retryMaxInterval: "300s"
retryType: periodic
retryWait: 1s
totalLimitSize: 32m

Improve our current Monitoring capabilities to better
help admins to gain insights into OpenShift Logging.
● Introduce dashboards into the OpenShift Console (admin
perspective) that shows the most critical data points for
admins to proactively research problems.
○ Two new dashboards: OpenShift Logging (central
overview look) and Elasticsearch.
○ Access from Monitoring -> Dashboards and select
either from the dropdown list.
● Enrich and/or improve current alerting rules to cover "you
must page me at 3am" scenarios.
● Overhaul metrics where necessary.
○ Note: Removed all index level metrics since they
introduced an abnormal amount of metrics which
ended up exploding our Monitoring solution. We will
reintroduce some + improvements in a future release.
Logging “Observability”
78

4.6 Supported Providers
Generally Available
Full Stack Automation (IPI) Pre-existing Infrastructure (UPI)
Bare Metal
Product Manager(s): Katherine Dubé (AWS, Azure, GCP), Maria Bracho (VMware), Peter Lauterbach (RHV & OCP Virtualization), Ramon Acedo Rodriguez (OSP, BM), & Duncan Hardie (IBM Z & Power)
IBM Power Systems
80
Bare Metal
New addition in OCP 4.6 Now supports deploying
to VMware vSphere 7.0

81
OpenShift on OpenStack
Supported OSP releases with OCP 4.6
Red Hat OpenStack Platform 13
Red Hat OpenStack Platform 16.1
New with OCP 4.6 on OSP
● OpenStack Bare Metal (Ironic) integration
● Installer support for specifying OpenStack Availability Zones
● Floating IPs no longer required
OpenShift on OpenStack

Enhancements to RHV full stack installer
What’s new in OCP 4.6
● Dynamically provision storage to OCP cluster with
RHV CSI operator
● Improved control of workloads and resources by
auto-scaling workers nodes
● Support for Disconnected / restricted installs
Supported RHV releases with OCP 4.6
● RHV 4.4.2+
● Customers running OCP 4.5 on RHV 4.3 must
upgrade to RHV 4.4.2+ before upgrading to
OCP 4.6
Upcoming work in future releases
● OCP on RHV UPI moved to OCP 4.7
Product Manager: Peter Lauterbach
Generally Available
$ ./openshift-install create cluster --dir ./demo
? SSH Public Key /home/user_id/.ssh/id_rsa.pub
? Platform ovirt
? Enter oVirt’s api endpoint URL admin:pw123
https://rhv-env.virtlab.example.com/ovirt-engine/api
? Is the installed oVirt certificate trusted? Yes
? Enter oVirt’s CA bundle xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
? Enter ovirt-engine username admin@internal
? Enter passsword xxxxxxxxxxxxx
? Select oVirt cluster Default
? Select oVirt storage domain hosted_storage
? Select oVirt network ovirtmgmt
? Enter the internal API virtual IP 10.35.1.19
? Enter the internal DNS virtual IP 10.35.1.21
? Enter the ingress IP 10.35.1.20
? Base Domain example.com
? Cluster Name demo
? Pull Secret [? for help] xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
INFO Creating infrastructure resources...
INFO API v1.17.1 up
https://console-openshift-console.apps.demo.example.com
INFO Login to the console with user: kubeadmin, password: xxxxx-xxxxx-xxxxx-xxxxx
82

New Credential Modes for OpenShift Installation
Specify how CredentialsRequests are satisfied
● Allows users to define how CredentialsRequest are handled on
behalf of OpenShift components requiring cloud API access.
● Three new modes can now be specified for deployments on
AWS, Azure, and GCP:
○ Mint: Creates new credentials with a subset of the overall
permissions as specified by the CredentialsRequest.
○ Passthrough: Uses the provided credentials “as is” for each
OpenShift component’s CredentialsRequest.
○ Manual: CredentialsRequests must be manually handled by
the user (useful for cases where access to the IAM endpoint
has been restricted.)
● If the field is set to any of the above values, then the installer
will not attempt to check the credential permissions prior to
installing OpenShift.
○ Important for situations where the credential policy
checking can’t adequately validate the user credentials
(when using SCP on AWS.)
Generally Available
Product Manager: Maria Bracho / Katherine Dubé
% ./openshift-install explain installconfig.credentialsMode
KIND: InstallConfig
VERSION: v1
RESOURCE: <string>
CredentialsMode is used to explicitly set the mode with which CredentialRequests are
satisfied.
If this field is set, then the installer will not attempt to query the cloud
permissions before attempting installation. If the field is not set or empty, then the
installer will perform its normal verification that the credentials provided are
sufficient to perform an installation.
There are three possible values for this field, but the valid values are dependent upon
the platform being used. "Mint": create new credentials with a subset of the overall
permissions for each CredentialsRequest "Passthrough": copy the credentials with all of
the overall permissions for each CredentialsRequest "Manual": CredentialsRequests must
be handled manually by the user
For each of the following platforms, the field can set to the specified values. For all
other platforms, the field must not be set.
83

AWS Custom Endpoint Support
Define custom API endpoints for private AWS regions
● Adds a new field ‘serviceEndpoints’ in install-config.yaml,
which contains a list of custom endpoints for overriding the
default service endpoints of AWS services.
● Custom API endpoints can be specified for EC2, S3, IAM,
Elastic Load Balancing, Tagging, Route 53, and STS AWS
services.
● Only required for cases were alternative AWS endpoints (like
FIPS) need to be used.
○ Note: Not needed for deploying to known regions (which are found
in the AWS SDK.)
● List of AWS service endpoints can be found here:
https://docs.aws.amazon.com/general/latest/gr/aws-service-information.html
Generally Available
apiVersion: v1
baseDomain: example.com
compute:
- architecture: amd64
hyperthreading: Enabled
name: worker
Platform: {}
replicas: 3
controlPlane:
architecture: amd64
name: master
platform: {}
replicas: 3
metadata:
creationTimestamp: null
name: mycluster
networking:
clusterNetwork:
- cidr: 10.18.0.0/14
hostPrefix: 23
machineNetwork:
- cidr: 10.0.0.0/16
networkType: OpenShiftSDN
serviceNetwork:
- 172.30.0.0/16
platform:
aws:
Region: us-east-2
amiID: ami-0f4ecf819275850dd
serviceEndpoints:
- service: ec2
url: https://ec2-fips.us-east-2.amazonaws.com
- service: s3
url: https://<account-id>.s3-control.us-east-2.amazonaws.com
publish: External
84

User Defined Routing on Azure
Define custom API endpoints for private Azure regions
● Today, internal clusters on Azure always use Public Standard Load
Balancers for Internet egress. This means public IPs and public load
balancers are required, which many customers don’t want to use for
internal clusters.
● User Defined Routing allows the users to choose their own
outbound routing for Internet access enabling them to leverage
pre-existing setups instead of defaulting to the per-cluster
OpenShift recommended way.
● Users are only allowed to change the outbound type when using
pre-existing networking since outbound routing needs to be setup
by user prior to installing the cluster.
● Adds a new egress strategy ‘UserDefinedRouting’ to the
‘outboundType’ field in the install-config
Generally Available
apiVersion: v1
compute:
name: worker
platform: {}
replicas: 3
controlPlane:
architecture: amd64
name: master
platform: {}
replicas: 3
metadata:
name: mycluster
networking:
clusterNetwork:
- cidr: 10.128.0.0/14
hostPrefix: 23
machineNetwork:
- cidr: 10.0.0.0/16
serviceNetwork:
- 172.30.0.0/16
platform:
azure:
baseDomainResourceGroupName: os4-common
cloudName: AzurePublicCloud
outboundType: UserDefinedRouting
region: eastus
publish: External
pullSecret: <secret>
85

Specify Disk Type & Size for Control Plane & Compute Nodes on Azure & GCP
Configure both disk type and size based on node requirements
● Support for configuring disk type and size on control plane and
compute nodes has been extended to Azure & GCP.
● Introduces two new fields ‘osDisk.diskSizeGB’ & ‘osDisk.diskType’ in
the install-config
● For Azure, supported disk types include:
"Standard_LRS","Premium_LRS", & "StandardSSD_LRS"
○ Note: For control plane nodes only “Premium_LRS” &
“StandardSSD_LR” can be configured.
● For GCP, supported disk types include: "pd-ssd" & "pd-standard"
○ Note: For control plane nodes only “pd-ssd” can be configured.
Generally Available
apiVersion: v1
compute:
name: worker
platform:
- osDisk:
DiskSizeGB: 120
DiskType: pd-standard
replicas: 3
controlPlane:
architecture: amd64
name: master
platform:
- osDisk:
DiskSizeGB: 120
DiskType: pd-ssd
replicas: 3
metadata:
name: mycluster
networking:
clusterNetwork:
- cidr: 10.128.0.0/14
hostPrefix: 23
machineNetwork:
- cidr: 10.0.0.0/16
serviceNetwork:
- 172.30.0.0/16
platform:
gcp:
projectID: openshift-production
region: us-central1
publish: External
86

Improved Recovery Time After Hard Shutdown of Master Node
After a hard shutdown of a master node, the result of a
failure or not, the OpenShift APIs would become
unavailable for a lengthy period of time (15min+) while
the endpoints were reconciled and the cluster detected
and adapted to the loss of the node.
For OpenShift 4.6, the recovery time of the control plane
was dramatically improved, in most cases, to ~90s.
89

Pod Topology Spread Constraints
kind: Pod
apiVersion: v1
metadata:
name: mypod
labels:
foo: bar
spec:
topologySpreadConstraints:
- maxSkew: 1
topologyKey: zone
whenUnsatisfiable: DoNotSchedule
labelSelector:
matchLabels:
foo: bar
Node 1 Node 2
Pod Pod
Zone = Zone A
Node 1 Node 2
Pod
New
Pod
Zone = Zone B
Control how Pods are spread across the cluster
among failure-domains such as regions, zones,
nodes, and other user-defined topology domains.
Help to achieve high availability as well as efficient
resource utilization
Product Manager: Tushar Katarki
90

Cluster Infrastructure updates
OCP CLUSTER INFRASTRUCTURE
● Expanding Spot Instance support
○ Azure: machine API support for spot instances
○ GCP: machine API support for Preemptible VM
instances
● Security and Compliance
○ AWS: Support for custom endpoints and
air-gapped regions
○ Azure: Support for GovCloud
● Usability
○ AWS Machine API Support of more than one block
device
○ Get validation/defaulting for providerSpec APIs
apiVersion: machine.openshift.io/v1beta1
spec:
metadata:
providerSpec:
spotMarketOptions:
maxPrice: "0.06"
MachineSet
Product Manager: Duncan Hardie
Generally Available
91

93
Kube-Native Operating System
Product Manager: Mark Russell
RHCOS
4.6
EUS
● Aligned for full life cycle with
RHEL 8.2.z EUS stream
● Stable 4.18 kernel ABI allowlist
● Deploy /var on a separate disk
● Extension system with usbguard

94
Updated CoreOS Image & Installer
Key Features
● Hardware and interface name discovery
● Preserve existing data partitions option
● Automatic 4K-sector drive detection
● Easily embed custom ignition configuration into
custom ISOs for installation in environments with
restricted networking
● Live PXE and Live ISO environment
Red Hat Enterprise Linux CoreOS 46.82.20200928174-0 (Ootpa) 4.6
SSH host key: SHA256:mmPpxnYfcrXsMng0c72dEm6GqoM5Bx/eOP3bm1DsuV4 (ECDSA)
SSH host key: SHA256:Nb30rUtSbanzeLyT4quS1tnH1116aFFZGZrmNWJMidQ (ED25519)
SSH host key: SHA256:u1wL1agK+UIGNLn5iBU8+bHBryk3QWGgNpZ8KfofZFa (RSA)
enp1s0: 192.168.122.51 fw80::5054::ff:fe6a:add7
enp6s0: 192.168.122.145 fe80::5054::ff:fe78:befe
localhost login: core (automatic login)
###########################################################################
Welcome to the CoreOS live environment. This system is running completely
from memory, making it a good candidate for hardware discovery and
installing persistently to disk. Here is an example of running an install
to disk via coreos—installer:
sudo coreos—installer install /dev/sda
—— ignition—url https://example.com/example.ign
You may configure networking via ‘sudo nmcli’ or ‘sudo nmtui’ and have
that configuration persist into the installed system by passing the
‘——copy—network’ argument to ‘coreos—installer install’. Please run
‘coreos—installer install ——help’ for more information on the possible
install options.
###########################################################################
[core@localhost ~]$

95
Improved Networking UX
For Bare Metal
● Use nmtui or nmcli from the Live Installer
environment
● Pass your live config by invoking the RHCOS
installer with the --copy-network argument
For VMware
● The new RHCOS VMware OVA file accepts static
networking in the guestinfo fields
● Pass dracut ip= syntax to configure static
networking through the vSphere web console
or API

SR-IOV Enhancements
Infiniband Support
● High-throughput low-latency communication
standard for high-perf internode message passing
● Configured via SR-IOV Operator and is enabled on
Mellanox CX-4/5/6 cards
IPAM Plug-in: whereabouts
● A CNI plug-in providing IPAM for other (Multus) CNI
plugins, e.g. DHCP
● Assigns IP addresses dynamically across the cluster,
and without DHCP, and allows overlapping IP ranges
● Stores IP address allocations via Kubernetes API
Infiniband Configuration Overview
1. Install SR-IOV operator
2. Create a SriovNetworkNodePolicy CR
3. Create an SR-IOV network
4. Create a pod with the Infiniband device and network
apiVersion: sriovnetwork.openshift.io/v1
kind: SriovNetworkNodePolicy
metadata:
name: policy-ib-net-1
namespace: openshift-sriov-network-operator
spec:
resourceName: ibnic1
nodeSelector:
feature.node.kubernetes.io/network-sriov.capable:
"true"
numVfs: 4
nicSelector:
vendor: "15b3"
deviceID: "101b"
rootDevices: ['0000:19:00.0']
linkType: ib
isRdma: true
{
"ipam": {
"type": "whereabouts",
"range": "<range>",
"exclude": ["<exclude_part>, ..."],
}
}
97

Additional Networking Enhancements
Switch to System OVS
● OVS previously ran in a cluster pod, resulting in existing network flow disruption upon cluster upgrades/restarts
● OVS now runs on the RHCOS host, and remains active during cluster upgrades/restarts
● Requires node reboot to update the OVS version
Extended serviceNodePortRange (UPI only)
Allows expansion of the default service node port range (30000-32767) for services of type NodePort for customers
that implement a large number of node ports, if the corresponding ports are opened at the infrastructure layer..
Increased Maximum Number of Rules per EgressFirewall Policy
The number of rules in a single EgressFirewall policy was insufficient for some deployments, and was raised from a
maximum of 50 to 1000.
oc patch network cluster -p '{"spec":{"serviceNodePortRange": "30000-33000"}}' --type=merge
98

Configuration Enhancements
HTTP Forwarded Header Policy
Use Case: A developer that configures an
application-specific proxy that injects
X-Forwarded-For and wants an IngressController to
pass the header through unmodified for the
application's Route.
HTTP Header Capture
Configure OpenShift to log specific HTTP request and
response headers for Routes, to ensure security
compliance and increase observability.
99
HTTP Cookie Capture
Configure OpenShift to log specific, named HTTP
cookies, to ensure security compliance and enable
business analytics.
Ingress TLS Termination Policy
Ingresses can now specify reencrypt or passthrough
policy:
● "reencrypt" decrypts and re-encrypts HTTP
traffic when forwarding it.
● "passthrough" passes traffic through without
terminating TLS.
HTTP Path Rewriting
Support for a Route annotation to configure path
rewriting. On incoming requests, the Route’s spec.path
is replaced with the rewrite target before forwarding.
HTTP Unique-Id Header
Configure an IngressController to inject an HTTP
header with a unique request id into each HTTP request
before forwarding the request to the application, so
that I can trace HTTP requests and increase
observability.

Configure IngressController to Use AWS NLB
By default, an IngressController resource will use an AWS
Classic Load Balancer when the endpoint publishing
strategy is “type: LoadBalancerService
” and the
Infrastructure resource platform status is “type: AWS”.
Simply by specifying the AWS provider parameter “type:
NLB” the IngressController resource will instead use an
AWS Network Load Balancer (NLB).
apiVersion: operator.openshift.io/v1
kind: IngressController
metadata:
name: $MY_INGRESS_CONTROLLER
namespace: openshift-ingress-operator
spec:
replicas: 1
domain: $MY_UNIQUE_INGRESS_DOMAIN
endpointPublishingStrategy:
type: LoadBalancerService
loadBalancer:
scope: External
providerParameters:
type: AWS
aws:
type: NLB
100

Storage updates
OCP STORAGE
OCP Supported
AWS EBS Fibre Channel
Azure File & Disk HostPath
GCE PD Local Volume
VMware vSphere Disk Raw Block
NFS iSCSI
Supported via OCS
File , Block, Raw Block, Object
Supported via OSP
Cinder
● No change on support for intree drivers
● CSI Operators
○ CSI Operator Library
○ Move to CSO managing CSI Operators
○ Indicate support of fsGroup
● CSI Capabilities
○ Crash Consistent Snapshots (Tech preview)
■ Fully supported when used with OCS or
CNV
● Enabling OCS via Local Storage Operator
○ Auto-provision of PVs
○ Continuous inventory of local disks
102

● Encryption support for the entire cluster
● Crash Consistent Snapshots, Clones
● Compression and Replica 2 for block storage
● Object namespaces - single view for multiple object
storage buckets.
● Improved bare metal deployment with LSO
○ Auto-provision of PVs
○ Continuous inventory of local disks
○ Easy local drive filtering
● Additional platforms - IBM Z/Power (by IBM)
OpenShift Container Storage updates
OCP STORAGE
Out of the box support
Block, File, Object
Platforms
AWS Azure (Tech Preview)
Bare metal Google Cloud (Tech Preview)
VMWare Azure (Tech Preview)
IBM Z/Power (by IBM) Oct 2020 - RHV (Tech Preview)
Nov 2020 - OSP (Tech Preview)
Deployment modes
Disconnected environment and Proxied environments
103

A Real Time Kernel is a Red Hat Enterprise Linux kernel
that is modified to maintain low latency, consistent
response time and workload determinism.
This feature allows workloads to run uninterrupted by the
Operating System.
● Allow the installation of the Real Time Kernel on RHEL
CoreOS nodes.
● Allow the cluster administrator to provide a
PerformanceProfile that defines:
○ A number of CPU cores dedicated to
“housekeeping” tasks.
○ A number of CPU cores dedicated for workloads
(CPU Pinning).
● NUMA alignment for devices, memory and cores used by
Low Latency Workloads.
Real Time Kernel and Low Latency Workloads for RAN
Product Manager: Robert Love
106
Real Time Kernel
0 1 2 3 4 5 6
A B C D E
CPU Cores:
RAN Workloads:
Cores Dedicated For Workloads
Core Dedicated to OS
“housekeeping”

Cloud-native Network Functions Tests (CNF Tests)
The CNF Tests container image allows service providers to validate that their cluster has been
provisioned and configured correctly ready to run CNFs. The documentation resides here.
It validates the following additional performance-related functionality is configured and available on
the cluster:
● Precision Time Protocol (PTP)
● Single-root input/output virtualization (SR-IOV)
● Stream Control Transmission Protocol (SCTP)
● Data Plane Development Kit (DPDK)
● Performance AddOn Operator (PAO)
Product Manager: Robert Love
107

Openshift File Integrity Operator
=
Notify
Enable
FileIntegrity
Checking
Monitor
Summarize
Observe
AIDE
AIDE Configuration
Scan Nodes
1 The operator scans the
selected nodes to populate
the AIDE database
2
Repeat scans collect
results, and check against
the AIDE database.
3 Admins can examine the
scan results for status
Deploy AIDE Pods
Notification
(fileIntegrityNodeStatus)
Roadmap
109

RH ACM and Compliance
=
Install, upgrade,
reconcile, conﬁg
Describe intent
with declarative
conﬁg
Monitor, scale,
troubleshoot,
backup
Maintain
Observe
apiVersion: machineconfiguration.openshift.io/v1
kind: ContainerRuntimeConfig
metadata:
name: set-log-and-pid
spec:
machineConfigPoolSelector:
matchLabels:
debug-crio: config-log-and-pid
containerRuntimeConfig:
pidsLimit: 2048
logLevel: debug
2 Red Hat curates cluster configs,
including RHCOS configs to meet
security profiles, like CIS or
NIST-800-53
1 A user requests a new
cluster
3 OpenShift operators apply updates;
he Machine Config Operator applies
the selected secure machine config
for RHCOS updates
Metrics are sent to Red
Hat Insights for analysis
via secured HTTPS.
4
Roadmap
115

Security/Auth Improvements: Customize Audit Config
Control the amount of information that is logged to the node audit logs by choosing the audit log policy profile to
use.
● Default: Logs only metadata for read and write requests; does not log request bodies. This is the default
policy.
● WriteRequestBodies: In addition to logging metadata for all requests, logs request bodies for every write
request to the API servers (create, update, patch). This profile has more resource overhead than the Default
profile.
● AllRequestBodies: In addition to logging metadata for all requests, logs request bodies for every read and
write request to the API servers (get, list, create, update, patch). This profile has the most resource overhead.
apiVersion: config.openshift.io/v1
kind: APIServer
metadata:
...
spec:
audit:
profile: WriteRequestBodies
116

You can conﬁgure OAuth tokens to expire after a set period of inactivity. By default, no token inactivity timeout is
set.
Add the spec.tokenConfig.accessTokenInactivityTimeout ﬁeld and set your timeout value:
apiVersion: config.openshift.io/v1
kind: OAuth
metadata:
...
spec:
tokenConfig:
accessTokenInactivityTimeout: 400s
Security/Auth Improvements: Token inactivity timeout for OAuth Server
Example output
error: You must be logged in to the server (Unauthorized)
117

OAuth access token and OAuth authorize token object names are now stored as non-sensitive object names.
Previously, secret information was used as the OAuth access token and OAuth authorize token object names. When
etcd is encrypted, only the value is encrypted, so this sensitive information was not encrypted.
Security/Auth Improvements: Secure OAuth Resource Storage
If you are upgrading your cluster to OpenShift Container Platform 4.6, old tokens from OpenShift Container
Platform 4.5 will still have the secret information exposed in the object name. By default, the expiration for tokens is
24 hours, but this setting can be changed by administrators. Sensitive data can still be exposed until all old tokens
have either expired or have been deleted by an administrator.
OAuth Server
Access token
{
“scope”: “..”,
“client_id”: “..”,
“exp”: “..”,
}
OAuth Access token
Encrypted
118

Windows Community Operator
● Community distribution of the Windows Machine Config Operator will be available in mid to late October
● The Windows Machine Config Operator is the entry point for OpenShift customers who want to run Windows
workloads on their clusters.
● The intent of this feature is to allow a cluster administrator to add a Windows compute node as a day 2
operation with a prescribed configuration to an installer provisioned OpenShift 4.6 cluster and enable
scheduling of Windows workloads.
● Prerequisite: OpenShift 4.6+ cluster configured with hybrid OVN Kubernetes networking.
● Tested on AWS and Azure. vSphere CI tests on-going
● Red Hat certified operator will be generally available in December
Community Operator Red Hat Operator
Location In Cluster OperatorHub Red Hat Marketplace
Available date Mid Oct Mid Dec
Platforms supported AWS, Azure AWS, Azure, vSphere (possibly)
Refresh cycle Every 1-2 months Every OCP Y stream
120

Windows Machine Config Operator (WMCO) workflow
WMCO
Transfer binaries
This includes Windows
Machine Config Bootstrapper
Configure kubelet
Remotely execute WMCB to
configure kubelet
Run hybrid-overlay
Create OpenShift HNS
network
Configure CNI
Configure kubelet for CNI
plugin
Set up kube-proxy
Maintains network rules on
nodes allowing outside
communication
WMCO WMCO WMCO WMCO
121

Multi-architecture updates
OCP MULTI-ARCHITECTURE
● Align IBM Power and IBM Z GA with x86
● Storage being expanded
○ Local Storage Operator
○ Fibre Channel
○ HostPath
○ Raw Block
○ iSCSI
○ 4k Disk support
● Logging now supported
● OpenShift Core (CVO
Operators)
● UPI installer
● OVS/OVN (networking)
● RHEL7 Based container
support
● RHEL CoreOS (host nodes)
● Ansible Engine
● Red Hat Software Collections
● AdoptOpenJDK with OpenJ9
● Single Sign-On (Z only)
Supported
● OpenShift Cluster Monitoring
(Prometheus, Grafana)
● Node Tuning Operator
● OpenShift Jenkins
● OpenShift Logging
(elasticSearch, kibana)
● Machine Configuration
Operator (used in IPI installs)
● Node Feature Discovery
Operator
● Red Hat Runtimes (Z only)
Extra content ported
122

linkedin.com/company/red-hat
youtube.com/user/RedHatVideos
facebook.com/redhatinc
twitter.com/RedHat
123
Thank you

What_s_New_in_OpenShift_Container_Platform_4.6.pdf

More Related Content

Similar to What_s_New_in_OpenShift_Container_Platform_4.6.pdf (20)

Recently uploaded (20)

What_s_New_in_OpenShift_Container_Platform_4.6.pdf