Description of problem: (From Clayton) multus is the first pod on a new node and the last to go ready because it pulls 5 images. it's taking about 1m15s. that means nodes go ready slowly because of multus. (it's the longest pole). the time is because init container images are pulled sequentially In order to address this, the daemonset for Multus should be separate from the ancillary CNI binary installation.
Tested and verified in 4.8.0-0.nightly-2021-06-02-025513 [weliang@weliang ~]$ oc get all NAME READY STATUS RESTARTS AGE pod/multus-additional-cni-plugins-5bln8 1/1 Running 0 128m pod/multus-additional-cni-plugins-6k2qb 1/1 Running 0 120m pod/multus-additional-cni-plugins-897zr 1/1 Running 0 128m pod/multus-additional-cni-plugins-f7bnl 1/1 Running 0 119m pod/multus-additional-cni-plugins-w92k7 1/1 Running 0 128m pod/multus-additional-cni-plugins-z4r5p 1/1 Running 0 119m pod/multus-admission-controller-bvzhl 2/2 Running 0 125m pod/multus-admission-controller-jtvhb 2/2 Running 0 125m pod/multus-admission-controller-kdt58 2/2 Running 0 125m pod/multus-cjf52 1/1 Running 0 128m pod/multus-g7bsg 1/1 Running 0 128m pod/multus-lkzmf 1/1 Running 0 119m pod/multus-mbfhr 1/1 Running 0 120m pod/multus-q67t5 1/1 Running 0 119m pod/multus-rs7sg 1/1 Running 0 128m pod/network-metrics-daemon-2qgss 2/2 Running 0 120m pod/network-metrics-daemon-cwxfj 2/2 Running 0 128m pod/network-metrics-daemon-ghs92 2/2 Running 0 128m pod/network-metrics-daemon-p6h2d 2/2 Running 0 128m pod/network-metrics-daemon-sj6zz 2/2 Running 0 119m pod/network-metrics-daemon-wnkvk 2/2 Running 0 119m NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE service/multus-admission-controller ClusterIP 172.30.173.199 <none> 443/TCP,8443/TCP 128m service/network-metrics-service ClusterIP None <none> 8443/TCP 128m NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE daemonset.apps/multus 6 6 6 6 6 kubernetes.io/os=linux 128m daemonset.apps/multus-additional-cni-plugins 6 6 6 6 6 kubernetes.io/os=linux 128m daemonset.apps/multus-admission-controller 3 3 3 3 3 node-role.kubernetes.io/master= 128m daemonset.apps/network-metrics-daemon 6 6 6 6 6 kubernetes.io/os=linux 128m [weliang@weliang ~]$
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 (Moderate: OpenShift Container Platform 4.8.2 bug fix and security update), 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/RHSA-2021:2438