kubernetesopenshiftokdfedora-coreos

OKD 4.8 bootstrap failing due to DNS


I'm trying to install OKD:

klenkes74@fortuna ~ $ ./openshift-install version
./openshift-install 4.8.0-0.okd-2021-10-10-030117
built from commit e0b12188a18a84be721116ddda74bb3510bfe03f
release image quay.io/openshift/okd@sha256:1d3f75529b141333939987ba03bf4ad76d83ae31d3b17df9a12c1f1ef67feff2

After some problems with the setup (I install on hetzner.de dedicated servers, which mean, I have no control over DHCP, TFTboot, network in general), The systems are starting. But every system is comming up as 'localhost'. According to the documentation, the DHCP provided name is used (but DHCP seems to not provide a name here). In this case, there should be a reverse lookup from the IP. According to the DNS the interface is set up correctly, but perhaps it used interface 'lo' instead of the ehternet card 'enp8s'.

Checking DNS:

[core@master0 ~]$ dig @213.133.98.98 +noall +answer api-int.loki.kaiserpfalz-edv.de
api-int.loki.kaiserpfalz-edv.de. 55 IN  A   5.9.62.16
[core@master0 ~]$ dig @213.133.98.98 +noall +answer api.loki.kaiserpfalz-edv.de
api.loki.kaiserpfalz-edv.de. 60 IN  A   5.9.62.16
[core@master0 ~]$ dig @213.133.98.98 +noall +answer console.openshift.console.apps.loki.kaiserpfalz-edv.de
console.openshift.console.apps.loki.kaiserpfalz-edv.de. 60 IN A 5.9.62.16
[core@master0 ~]$ dig @213.133.98.98 +noall +answer -x 5.9.62.16
16.62.9.5.in-addr.arpa. 86400   IN  PTR fortuna.kaiserpfalz-edv.de.

[core@master0 ~]$ dig @213.133.98.98 +noall +answer master0.loki.kaiserpfalz-edv.de
master0.loki.kaiserpfalz-edv.de. 60 IN  A   188.40.69.186
[core@master0 ~]$ dig @213.133.98.98 +noall +answer -x 188.40.69.186
186.69.40.188.in-addr.arpa. 84697 IN    PTR master0.loki.kaiserpfalz-edv.de.

[core@master0 ~]$ dig @213.133.98.98 +noall +answer master1.loki.kaiserpfalz-edv.de
master1.loki.kaiserpfalz-edv.de. 60 IN  A   188.40.123.177
[core@master0 ~]$ dig @213.133.98.98 +noall +answer -x 188.40.123.177
177.123.40.188.in-addr.arpa. 34543 IN   PTR master1.loki.kaiserpfalz-edv.de.

[core@master0 ~]$ dig @213.133.98.98 +noall +answer master2.loki.kaiserpfalz-edv.de
master2.loki.kaiserpfalz-edv.de. 60 IN  A   188.40.69.184
[core@master0 ~]$ dig @213.133.98.98 +noall +answer -x 188.40.69.184
184.69.40.188.in-addr.arpa. 86400 IN    PTR master2.loki.kaiserpfalz-edv.de.

[core@master0 ~]$ dig @213.133.98.98 +noall +answer bootstrap.loki.kaiserpfalz-edv.de
bootstrap.loki.kaiserpfalz-edv.de. 60 IN A  162.55.101.99
[core@master0 ~]$ dig @213.133.98.98 +noall +answer -x 162.55.101.99
99.101.55.162.in-addr.arpa. 86400 IN    PTR bootstrap.loki.kaiserpfalz-edv.de.

Am I missing some configuration? My install-config.yaml is quite basic:

---
apiVersion: v1
metadata:
  name: loki
baseDomain: kaiserpfalz-edv.de
compute:
- hyperthreading: Enabled
  name: worker
  replicas: 0
  platform: {}
controlPlane:
  hyperthreading: Enabled
  name: master
  replicas: 3
  platform: {}
networking:
  clusterNetwork:
  - cidr: 10.128.0.0/14
    hostPrefix: 23
  networkType: OpenShiftSDN
  serviceNetwork:
  - 172.30.0.0/16
platform:
  none: {}
fips: false
pullSecret: <omitted>
sshKey <omitted>

Solution

  • The problem was the special environment of Hetzner. I had to log into the nodes as soon as they started and reconfigure the hostname. After the next reboot everything was fine.