https://rancher.com/ logo
Join Slack
Powered by
another issue I posted on <#CBEB5A0H1> but looks l...
# general
c
another issue I posted on #CBEB5A0H1 but looks like its not related to vsphere so I'm postinig here when deploying a cluster, Rancher creates the VMs on vSphere but the cluster is stuck in "Updating" and the nodes stuck in "Reconciling". looks like the issue is with 
deployment.apps/calico-typha
Copy code
2026-07-03 18:54:08.350 [ERROR][1] typha/health.go 389: Health endpoint failed, trying to restart it... error=listen tcp: lookup localhost on 192.168.27.8:53: no such host
192.168.27.8 is my network DNS server why is calico-typha it searching for 
localhost
in there?! I just added 
"cluster-dns": "127.0.0.1",
to Networking to no avail One important information I'm getting the IP Address from a "Network Protocol Profile" on vsphere and using this cloud-config to apply it gist.github.com/gschanuel/4817b239bc3ef163f2ee011d575bd464 what am I missing here?
✅ 1
I changed the linux installation I was using Ubuntu Server Minimal (with some extras) Now with Ubuntu Server normal install it's working 🤦‍♂️ there is some dependency that is not announced on the docs I thought it was solved, but it's not I changed to cilium and at least I could get the cluster to be Active most of the time the kube-apiserver on all three controllers are restarting with the message
Copy code
Normal   Created    8m3s                    kubelet  spec.containers{kube-apiserver}: Container created
Normal   Started    8m2s                    kubelet  spec.containers{kube-apiserver}: Container started
Warning  Unhealthy  4m1s (x24 over 7m51s)   kubelet  spec.containers{kube-apiserver}: Startup probe failed: Unable to connect to the server: dial tcp: lookup localhost on 192.168.27.8:53: no such host
Normal   Killing    4m1s                    kubelet  spec.containers{kube-apiserver}: Container kube-apiserver failed startup probe, will be restarted
Warning  Unhealthy  2m12s (x7 over 3m12s)   kubelet  spec.containers{kube-apiserver}: Liveness probe failed: Unable to connect to the server: dial tcp: lookup localhost on 192.168.27.8:53: no such host
Warning  Unhealthy  2m6s (x18 over 3m18s)   kubelet  spec.containers{kube-apiserver}: Readiness probe failed: Unable to connect to the server: dial tcp: lookup localhost on 192.168.27.8:53: no such host
Normal   Killing    2m2s                    kubelet  spec.containers{kube-apiserver}: Container kube-apiserver failed liveness probe, will be restarted
Warning  Unhealthy  10s (x17 over 77s)      kubelet  spec.containers{kube-apiserver}: Readiness probe failed: Unable to connect to the server: dial tcp: lookup localhost on 192.168.27.8:53: no such host
Warning  Unhealthy  7s (x8 over 77s)        kubelet  spec.containers{kube-apiserver}: Liveness probe failed: Unable to connect to the server: dial tcp: lookup localhost on 192.168.27.8:53: no such host
Normal   Killing    7s                      kubelet  spec.containers{kube-apiserver}: Container kube-apiserver failed liveness probe, will be restarted
dumb dumb me... my /etc/hosts were empty 🤦‍♂️
Copy code
#cloud-config
manage_etc_hosts: false
runcmd:
  - |
    vmtoolsd --cmd 'info-get guestinfo.ovfEnv' > /tmp/ovfenv
    IP=$(sed -n 's/.*Property oe:key="guestinfo.interface.0.ip.0.address" oe:value="\([^"]*\).*/\1/p' /tmp/ovfenv)
    MASK=$(sed -n 's/.*Property oe:key="guestinfo.interface.0.ip.0.netmask" oe:value="\([^"]*\).*/\1/p' /tmp/ovfenv)
    GW=$(sed -n 's/.*Property oe:key="guestinfo.interface.0.route.0.gateway" oe:value="\([^"]*\).*/\1/p' /tmp/ovfenv)
    DNS=$(sed -n 's/.*Property oe:key="guestinfo.dns.servers" oe:value="\([^"]*\).*/\1/p' /tmp/ovfenv)

    IFACE=$(ip -o link | awk -F': ' '$2 !~ /lo/ {print $2; exit}')
    PREFIX=$(echo $MASK | awk -F. '{c=0; for(i=1;i<=4;i++){n=$i; while(n){c+=n%2; n=int(n/2)}}; print c}')
    cat > /etc/netplan/99-guestinfo.yaml <<EOF
    network:
      version: 2
      ethernets:
        $IFACE:
          dhcp4: false
          addresses: [$IP/$PREFIX]
          routes:
            - to: default
              via: $GW
          nameservers:
            addresses: [${DNS}]
            search: [rke.prd.internal]
    EOF
    chmod 600 /etc/netplan/99-guestinfo.yaml

    netplan apply

    HOSTNAME=$(hostname)
    ZONE="rke.prd.internal"

    hostnamectl set-hostname ${HOSTNAME}.${ZONE}

    cat > /etc/hosts <<EOF
      127.0.0.1   localhost
      ${IP}   ${HOSTNAME}.${ZONE} ${HOSTNAME}
    EOF
just updated my cloud-config and new clusters are Active, up and running in less than 8 minutes now
2 Views
Open in Slack
PreviousNext
Powered by