Merge pull request #29487 from johnwilkins/CNF950DU

New PR CNF950DU

Author: ahardin-rh

_topic_map.yml

@@ -1803,6 +1803,9 @@ Topics:
 - Name: Performance Addon Operator for low latency nodes
   File: cnf-performance-addon-operator-for-low-latency-nodes
   Distros: openshift-origin,openshift-enterprise,openshift-webscale
+- Name: Provisioning and deploying a distributed unit (DU)
+  File: cnf-provisioning-and-deploying-a-distributed-unit
+  Distros: openshift-webscale
 ---
 Name: Backup and restore
 Dir: backup_and_restore

images/135_OpenShift_Distributed_Unit_0121.svg

@@ -0,0 +1,255 @@
+// CNF-950 4.7 Deploying the DU infrastructure profile
+// Module included in the following assemblies:
+//
+// *cnf-provisioning-and-deploying-a-distributed-unit.adoc
+
+[id="scalability_and_performance/cnf-deploying-the-du-infrastructure-profile_{context}"]
+= Deploying the DU infrastructure profile
+
+[id="cnf-creating-the-performance-addon-operator-and-du-performance-profile_{context}"]
+== Creating the Performance Addon Operator and DU performance profile
+
+. Create and apply the performance profile, for example:
++
+[source,yaml]
+----
+cat <<EOF |  oc apply -f -
+apiVersion: performance.openshift.io/v1
+kind: PerformanceProfile
+metadata:
+# This profile is for typical lab DELL R640 with 2 x 26 pcores
+  name: perf-example
+spec:
+  additionalKernelArgs:
+  - nosmt
+  cpu:
+    # Temp. workaround for RT kernel bugs that consume too much
+    # CPU power: add more isolated cores
+    isolated: "1,3,5,7,9-51"  <1>
+    reserved: "0,2,4,6,8"     <2>
+  hugepages:
+    defaultHugepagesSize: 1G
+    pages:
+    - count: 16   <3>
+      size: 1G
+      node: 0
+  nodeSelector:
+    # Pay attention to the node label, create MCP accordingly
+    node-role.kubernetes.io/worker-cnf: ""
+  numa:
+    topologyPolicy: "restricted"
+  realTimeKernel:
+    # For CU should be false
+    enabled: true
+
+EOF
+----
+<1> Configure this line based upon the customer CPU hardware selected to run the DU.
+<2> Configure this line based upon the customer CPU hardware selected to run the DU.
+<3> Configure this line based upon the customer memory configuration selected to run the DU.
+
+. Use the following command to verify the changes have been applied to the cluster:
++
+[source,terminal]
+----
+$ oc wait mcp/worker-cnf --for condition="updated"
+----
+
+[id="cnf-creating-the-ptp-operator-and-slave-profile_{context}"]
+== Creating the PTP Operator and slave profile
+
+. Apply the PTP configuration, for example:
++
+[source,yaml]
+----
+cat <<EOF |  oc apply -f -
+apiVersion: ptp.openshift.io/v1
+kind: PtpConfig
+metadata:
+  name: slave
+  namespace: openshift-ptp
+spec:
+  profile:
+  - name: "slave"
+# The interface name is hardware-specific
+    interface: "eno1" <1>
+    ptp4lOpts: "-2 -s --summary_interval -4" <2>
+    phc2sysOpts: "-a -r -n 24" <3>
+    ptp4lConf: |
+      [global]
+      #
+      # Default Data Set
+      #
+      twoStepFlag 1
+      slaveOnly 0
+      priority1 128
+      priority2 128
+      domainNumber 24 <4>
+      #utc_offset 37
+      clockClass 248
+      clockAccuracy 0xFE
+      offsetScaledLogVariance 0xFFFF
+      free_running 0
+      freq_est_interval 1
+      dscp_event 0
+      dscp_general 0
+      dataset_comparison ieee1588
+      G.8275.defaultDS.localPriority 128
+      #
+      # Port Data Set
+      #
+      logAnnounceInterval -3 <5>
+      logSyncInterval -4 <5>
+      logMinDelayReqInterval -4 <5>
+      logMinPdelayReqInterval -4 <5>
+      announceReceiptTimeout 3 <5>
+      syncReceiptTimeout 0
+      delayAsymmetry 0
+      fault_reset_interval 4
+      neighborPropDelayThresh 20000000
+      masterOnly 0
+      G.8275.portDS.localPriority 128
+      #
+      # Run time options
+      #
+      assume_two_step 0
+      logging_level 6
+      path_trace_enabled 0
+      follow_up_info 0
+      hybrid_e2e 0
+      inhibit_multicast_service 0
+      net_sync_monitor 0
+      tc_spanning_tree 0
+      tx_timestamp_timeout 1
+      unicast_listen 0
+      unicast_master_table 0
+      unicast_req_duration 3600
+      use_syslog 1
+      verbose 0
+      summary_interval 0
+      kernel_leap 1
+      check_fup_sync 0
+      #
+      # Servo Options
+      #
+      pi_proportional_const 0.0
+      pi_integral_const 0.0
+      pi_proportional_scale 0.0
+      pi_proportional_exponent -0.3
+      pi_proportional_norm_max 0.7
+      pi_integral_scale 0.0
+      pi_integral_exponent 0.4
+      pi_integral_norm_max 0.3
+      step_threshold 0.0
+      first_step_threshold 0.00002
+      max_frequency 900000000
+      clock_servo pi
+      sanity_freq_limit 200000000
+      ntpshm_segment 0
+      #
+      # Transport options
+      #
+      transportSpecific 0x0
+      ptp_dst_mac 01:1B:19:00:00:00
+      p2p_dst_mac 01:80:C2:00:00:0E
+      udp_ttl 1
+      udp6_scope 0x0E
+      uds_address /var/run/ptp4l
+      #
+      # Default interface options
+      #
+      clock_type OC
+      network_transport UDPv4
+      delay_mechanism E2E
+      time_stamping hardware
+      tsproc_mode filter
+      delay_filter moving_median
+      delay_filter_length 10
+      egressLatency 0
+      ingressLatency 0
+      boundary_clock_jbod 0
+      #
+      # Clock description
+      #
+      productDescription ;;
+      revisionData ;;
+      manufacturerIdentity 00:00:00
+      userDescription ;
+      timeSource 0xA0
+  recommend:
+  - profile: "slave"
+    priority: 4
+    match:
+    - nodeLabel: "ptp/slave"
+
+EOF
+----
+<1> The interface selected needs to match the Linux interface name.
+<2> `-2` configures Ethernet encapsulation of PTP.  `--summary_interval -4` sets the logging interval.
+This is currently set to match `logSyncInterval -4`.
+<3> `-n 24` must match the `domainNumber 24`.
+<4> `domainNumber 24` must match the `-n 24`.
+<5> These variables are set to enable the G.8275.1 profile for PTP.
+
+[id="cnf-creating-the-sriov-operator-and-associated-profiles_{context}"]
+== Creating the SR-IOV Operator and associated profiles
+
+. Apply the SR-IOV network node policy, for example:
++
+[source,yaml]
+----
+cat <<EOF |  oc apply -f -
+apiVersion: sriovnetwork.openshift.io/v1
+kind: SriovNetworkNodePolicy
+metadata:
+  name: policy-mh-dpdk-site-1-fqdn-worker1
+  namespace: openshift-sriov-network-operator
+spec:
+# This works for Intel based NICs. <1>
+# For Mellanox please change to:
+#     deviceType: netdevice
+#     isRdma: true
+  deviceType: vfio-pci
+  isRdma: false
+  nicSelector:
+# The exact physical function name must match the hardware used
+    pfNames: ["ens1f1"] <2>
+  nodeSelector:
+    node-role.kubernetes.io/worker-cnf: ""
+    feature.node.kubernetes.io/network-sriov.capable: "true"
+  numVfs: 4
+  priority: 10
+  resourceName: mh_u_site_1_fqdn_worker1
+
+EOF
+----
+<1> This file works for Intel and must change for Mellanox, as described in
+_SR-IOV configuration notes_.
+<2> Must be updated with the specific device on the server.
+
+. Create the SR-IOV network, for example:
++
+[source,yaml]
+----
+cat <<EOF | oc apply -f -
+apiVersion: sriovnetwork.openshift.io/v1
+kind: SriovNetwork
+metadata:
+  name: mh-net-u-site-1-fqdn-worker1
+  namespace: openshift-sriov-network-operator
+spec:
+  ipam:  |
+    {
+    }
+  networkNamespace: mh-net-ns-site-1-fqdn-worker1
+  resourceName: mh_u_site_1_fqdn_worker1
+  vlan: 100  <1>
+---
+apiVersion: v1
+kind: Namespace
+metadata:
+    name: mh-net-ns-site-1-fqdn-worker1
+
+EOF
+----
+<1> Modify this line to match the DU’s networking.