Bug 1809238

Note: even after manually approving the CSRs the worker nodes do not get into Ready state because ovnkube-node keeps looping through:

[root@worker-0 core]# crictl logs 0eba49f3ad7c8
+ [[ -f /env/worker-0.ocp-edge-cluster.qe.lab.redhat.com ]]
+ cp -f /usr/libexec/cni/ovn-k8s-cni-overlay /cni-bin-dir/
+ ovn_config_namespace=openshift-ovn-kubernetes
+ retries=0
+ true
++ kubectl get ep -n openshift-ovn-kubernetes ovnkube-db -o 'jsonpath={.subsets[0].addresses[0].ip}'
+ db_ip=
+ [[ -n '' ]]
+ ((  retries += 1  ))
+ [[ 1 -gt 40 ]]
+ echo 'waiting for db endpoint'
waiting for db endpoint
+ sleep 5
+ true
++ kubectl get ep -n openshift-ovn-kubernetes ovnkube-db -o 'jsonpath={.subsets[0].addresses[0].ip}'
+ db_ip=
+ [[ -n '' ]]
+ ((  retries += 1  ))
waiting for db endpoint
+ [[ 2 -gt 40 ]]
+ echo 'waiting for db endpoint'
+ sleep 5
+ true
++ kubectl get ep -n openshift-ovn-kubernetes ovnkube-db -o 'jsonpath={.subsets[0].addresses[0].ip}'
+ db_ip=
+ [[ -n '' ]]
+ ((  retries += 1  ))
waiting for db endpoint

So is it one testblocker ?

(In reply to Wei Sun from comment #2)
> So is it one testblocker ?

Yes, it's a test blocker.

The ovnkube-node issue is a known bug which is fixed by: https://github.com/openshift/cluster-kube-controller-manager-operator/pull/359

I would test again, as at least that issue should be gone.  As for the CSR approval thing, there's nothing I can do without either logs or access to a cluster in this state.

Marius gave me access to a cluster showing the problem.

The cluster-machine-approver shows:

I0311 18:10:32.635909       1 csr_check.go:418] retrieving serving cert from worker-0.ocp-edge-cluster.qe.lab.redhat.com ([fd2e:6f44:5dd8:c956::13b]:10250)
W0311 18:10:32.636926       1 csr_check.go:178] Failed to retrieve current serving cert: remote error: tls: internal error
I0311 18:10:32.636948       1 csr_check.go:183] Falling back to machine-api authorization for worker-0.ocp-edge-cluster.qe.lab.redhat.com
I0311 18:10:32.636959       1 main.go:181] CSR csr-nlgss not authorized: No target machine for node "worker-0.ocp-edge-cluster.qe.lab.redhat.com"
I0311 18:10:32.636966       1 main.go:217] Error syncing csr csr-nlgss: No target machine for node "worker-0.ocp-edge-cluster.qe.lab.redhat.com"

... meaning that it failed to map a Node back to a corresponding Machine.  This usually happens because our addresses don't match up correctly.

The addresses on the worker-0 Node are:

  addresses:
  - address: fd2e:6f44:5dd8:c956::13b
    type: InternalIP
  - address: worker-0.ocp-edge-cluster.qe.lab.redhat.com
    type: Hostname

The addresses on the Machine are:

    addresses: 
    - address: 172.22.0.59
      type: InternalIP
    - address: ""
      type: InternalIP
    - address: worker-0.ocp-edge-cluster.qe.lab.redhat.com
      type: Hostname
    - address: worker-0.ocp-edge-cluster.qe.lab.redhat.com
      type: InternalDNS

and the relevant info from the BareMetalHost:

    hostname: worker-0.ocp-edge-cluster.qe.lab.redhat.com
    nics:
    - ip: 172.22.0.59
      mac: 52:54:00:09:9d:d2
      model: 0x1af4 0x0001
      name: enp4s0
      pxe: true
      speedGbps: 0
      vlanId: 0
    - ip: ""
      mac: 52:54:00:50:57:ca
      model: 0x1af4 0x0001
      name: enp5s0
      pxe: false
      speedGbps: 0
      vlanId: 0


The problem here is that we failed to collect the host's IPv6 address during Ironic introspection.  That interface has a blank ip field on the BareMetalHost (which got copied to the Machine).

We need to determine what went wrong during introspection.  Either the host didn't get an IP (or didn't even ask for one), or it failed to report it back to ironic.

I removed the TestBlocker as deployment passes now(tested on 4.4.0-0.ci-2020-03-11-095511) and we can use as a workaround manually approving the certificates:

oc get csr | grep Pending | awk {'print $1'} | xargs oc adm certificate approve

Marius, if you still have this environment, can you please provide the raw introspection data for the interfaces for one of the workers, e.g similar to:

$ oc describe bmh openshift-worker-0 -n openshift-machine-api | grep -A1 Provisioning
  Provisioning:
    ID:  cfc24b87-d922-492a-a504-22a5d13057c3

$ curl http://172.22.0.3:5050/v1/introspection/cfc24b87-d922-492a-a504-22a5d13057c3/data | jq .all_interfaces,.interfaces
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  3091  100  3091    0     0   754k      0 --:--:-- --:--:-- --:--:--  754k
{
  "enp2s0": {
    "ip": "192.168.111.23",
    "mac": "00:24:28:5f:06:1e",
    "client_id": null,
    "pxe": false
  },
  "enp1s0": {
    "ip": "172.22.0.90",
    "mac": "00:24:28:5f:06:1c",
    "client_id": null,
    "pxe": true
  }
}
{
  "enp1s0": {
    "ip": "172.22.0.90",
    "mac": "00:24:28:5f:06:1c",
    "client_id": null,
    "pxe": true
  }
}

I didn't have the same environment anymore but this is what I got from a new one which shows the same issue:


oc describe bmh openshift-worker-0 -n openshift-machine-api | grep -A1 Provisioning
  Provisioning:
    ID:  a8c6d4fd-74a2-4f7e-b57e-4b900c63b58a
--
  Normal  ProvisioningStarted   66m   metal3-baremetal-controller  Image provisioning started for http://[fd00:1101::3]:6180/images/rhcos-44.81.202002241126-0-openstack.x86_64.qcow2/rhcos-44.81.202002241126-0-compressed.x86_64.qcow2
  Normal  ProvisioningComplete  62m   metal3-baremetal-controller  Image provisioning completed for http://[fd00:1101::3]:6180/images/rhcos-44.81.202002241126-0-openstack.x86_64.qcow2/rhcos-44.81.202002241126-0-compressed.x86_64.qcow2
  Normal  Registered            20m   metal3-baremetal-controller  Registered new host

[kni@provisionhost-0 ~]$ curl -g http://[fd00:1101::3]:5050/v1/introspection/a8c6d4fd-74a2-4f7e-b57e-4b900c63b58a/data 
{"error":{"message":"Introspection data not found for node a8c6d4fd-74a2-4f7e-b57e-4b900c63b58a, processed=True"}}


oc -n openshift-machine-api get bmh/openshift-worker-0 -o yaml 
apiVersion: metal3.io/v1alpha1
kind: BareMetalHost
metadata:
  creationTimestamp: "2020-03-12T15:17:51Z"
  finalizers:
  - baremetalhost.metal3.io
  generation: 2
  name: openshift-worker-0
  namespace: openshift-machine-api
  resourceVersion: "53866"
  selfLink: /apis/metal3.io/v1alpha1/namespaces/openshift-machine-api/baremetalhosts/openshift-worker-0
  uid: 7230aad7-e738-4a25-8f63-74159839b001
spec:
  bmc:
    address: redfish://[fd2e:6f44:5dd8:c956::1]:8000/redfish/v1/Systems/934a8d73-0eef-4e01-80b8-07ae714f2eb1
    credentialsName: openshift-worker-0-bmc-secret
    disableCertificateVerification: true
  bootMACAddress: 52:54:00:51:2a:08
  consumerRef:
    apiVersion: machine.openshift.io/v1beta1
    kind: Machine
    name: ocp-edge-cluster-worker-0-fzttk
    namespace: openshift-machine-api
  hardwareProfile: unknown
  image:
    checksum: http://[fd00:1101::3]:6180/images/rhcos-44.81.202002241126-0-openstack.x86_64.qcow2/rhcos-44.81.202002241126-0-compressed.x86_64.qcow2.md5sum
    url: http://[fd00:1101::3]:6180/images/rhcos-44.81.202002241126-0-openstack.x86_64.qcow2/rhcos-44.81.202002241126-0-compressed.x86_64.qcow2
  online: true
  userData:
    name: worker-user-data
    namespace: openshift-machine-api
status:
  errorMessage: ""
  goodCredentials:
    credentials:
      name: openshift-worker-0-bmc-secret
      namespace: openshift-machine-api
    credentialsVersion: "7733"
  hardware:
    cpu:
      arch: x86_64
      clockMegahertz: 2199.996
      count: 8
      flags:
      - 3dnowprefetch
      - abm
      - adx
      - aes
      - apic
      - arat
      - arch_capabilities
      - arch_perfmon
      - avx
      - avx2
      - bmi1
      - bmi2
      - clflush
      - cmov
      - constant_tsc
      - cpuid
      - cpuid_fault
      - cx16
      - cx8
      - de
      - ept
      - erms
      - f16c
      - flexpriority
      - fma
      - fpu
      - fsgsbase
      - fxsr
      - hle
      - hypervisor
      - invpcid
      - invpcid_single
      - lahf_lm
      - lm
      - mca
      - mce
      - mmx
      - movbe
      - msr
      - mtrr
      - nopl
      - nx
      - pae
      - pat
      - pcid
      - pclmulqdq
      - pdpe1gb
      - pge
      - pni
      - popcnt
      - pse
      - pse36
      - pti
      - rdrand
      - rdseed
      - rdtscp
      - rep_good
      - rtm
      - sep
      - smap
      - smep
      - ss
      - sse
      - sse2
      - sse4_1
      - sse4_2
      - ssse3
      - syscall
      - tpr_shadow
      - tsc
      - tsc_adjust
      - tsc_deadline_timer
      - tsc_known_freq
      - umip
      - vme
      - vmx
      - vnmi
      - vpid
      - x2apic
      - xsave
      - xsaveopt
      - xtopology
      model: Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz
    firmware:
      bios:
        date: ""
        vendor: ""
        version: ""
    hostname: worker-0.ocp-edge-cluster.qe.lab.redhat.com
    nics:
    - ip: ""
      mac: 52:54:00:4d:38:bc
      model: 0x1af4 0x0001
      name: enp5s0
      pxe: false
      speedGbps: 0
      vlanId: 0
    - ip: ""
      mac: 52:54:00:51:2a:08
      model: 0x1af4 0x0001
      name: enp4s0
      pxe: true
      speedGbps: 0
      vlanId: 0
    ramMebibytes: 16384
    storage:
    - hctl: "0:0:0:0"
      model: QEMU HARDDISK
      name: /dev/sda
      rotational: true
      serialNumber: drive-scsi0-0-0-0
      sizeBytes: 55834574848
      vendor: QEMU
    systemVendor:
      manufacturer: Red Hat
      productName: KVM
      serialNumber: ""
  hardwareProfile: unknown
  lastUpdated: "2020-03-12T16:47:59Z"
  operationHistory:
    deprovision:
      end: null
      start: null
    inspect:
      end: "2020-03-12T15:36:53Z"
      start: "2020-03-12T15:33:38Z"
    provision:
      end: "2020-03-12T15:43:53Z"
      start: "2020-03-12T15:40:08Z"
    register:
      end: "2020-03-12T15:33:38Z"
      start: "2020-03-12T15:33:18Z"
  operationalStatus: OK
  poweredOn: true
  provisioning:
    ID: a8c6d4fd-74a2-4f7e-b57e-4b900c63b58a
    image:
      checksum: http://[fd00:1101::3]:6180/images/rhcos-44.81.202002241126-0-openstack.x86_64.qcow2/rhcos-44.81.202002241126-0-compressed.x86_64.qcow2.md5sum
      url: http://[fd00:1101::3]:6180/images/rhcos-44.81.202002241126-0-openstack.x86_64.qcow2/rhcos-44.81.202002241126-0-compressed.x86_64.qcow2
    state: provisioned
  triedCredentials:
    credentials:
      name: openshift-worker-0-bmc-secret
      namespace: openshift-machine-api
    credentialsVersion: "7733"

Created attachment 1669718 [details]
metal3-ironic-inspector.log

Attaching metal3-ironic-inspector log

*** This bug has been marked as a duplicate of bug 1816121 ***