openshift 4.20.12安装报错：
1 waitforceo.go:69] waiting on condition EtcdRunningInCluster in etcd CR /cluster to be True
google搜索显示如下：
The message "waiting on condition EtcdRunningInCluster in etcd CR /cluster to be True" indicates that the Kubernetes (specifically OpenShift, where this message is common) operator is waiting for the etcd cluster to successfully form and become healthy. This is a normal part of the bootstrapping process, but if it persists, it typically points to underlying network configuration, DNS, or load balancer issues preventing the etcd members from communicating correctly.
Common Causes and Troubleshooting Steps
If this message repeats for an extended period (e.g., more than 15-30 minutes), the cluster installation may have stalled due to one of the following reasons:
1. Network Configuration and DNS Issues
The most frequent cause is incorrect DNS configuration or a misconfigured load balancer, which prevents the etcd members from finding each other or the API server.
- Verify DNS Records:Ensure that all control plane (master) nodes can resolve the
api-intandapiURLs to the correct load balancer or virtual IP addresses, and that the load balancer is correctly forwarding traffic to the master nodes. - Check Load Balancer Health:Confirm that your load balancer (e.g., HAProxy) is running, configured correctly with all etcd member IPs/hostnames, and that firewall rules allow traffic on the necessary ports (typically 2379/2380 for etcd peer/client communication, and 6443 for the Kubernetes API).
2. Incorrect Member Information
If the installation was previously attempted and failed, etcd members might still be trying to connect to old peer addresses stored in their data directories.
- Clean up old data:If you are re-running an installation on the same hosts, you may need to delete the etcd data directory (commonly
/var/lib/etcd/member) on each etcd host to force a fresh cluster formation. The operator/orchestration tool should then rebuild the directory.
3. Resource Constraints
Insufficient CPU or memory on the bootstrap or control plane nodes can prevent essential services from starting up correctly.
- Check System Resources:Ensure your nodes meet the minimum system requirements for the specific Kubernetes/OpenShift version you are installing. Use tools like
toporhtopto monitor resource usage.
4. Certificate Issues
Problems with SSL/TLS certificates can prevent secure communication between etcd members.
- Review Logs:Check the detailed logs of the etcd operator or the etcd service itself using commands like
journalctl -u etcdor by using theoc adm must-gathercommand (in OpenShift) to look for certificate-related errors (e.g., "x509: certificate signed by unknown authority").
Next Steps
To troubleshoot effectively:
- Access detailed logsfrom the bootstrap process and the etcd service on the control plane nodes.
- Verify your networking setupagainst the official OpenShift documentation or the relevant Kubernetes project's installation guide.
- Ensure all prerequisitesregarding DNS, load balancing, and required resources are met.
根据Google提供的信息，根据我的实际情况，我判断是内存不够造成的。所以重新安装。
重新安装后，错误不再出现。
安装后的环境如下所示：
[root@bastion install]# oc get nodes
NAME STATUS ROLES AGE VERSION
master0.prod.cpd.com Ready control-plane,master 41m v1.33.6
master1.prod.cpd.com Ready control-plane,master 40m v1.33.6
master2.prod.cpd.com Ready control-plane,master 41m v1.33.6
worker1.prod.cpd.com Ready worker 20m v1.33.6
worker2.prod.cpd.com Ready worker 20m v1.33.6
worker3.prod.cpd.com Ready worker 20m v1.33.6
[root@bastion install]#
)
)
