用box往pod中注入故障,报错cgroups load failed, cgroups: cgroup deleted; 直接进入operator容器中,执行blade命令注入故障,成功
有大佬帮看看怎么回事呀,一整天卡在这个问题上。
[root@master1 opt]# kubectl describe pod chaosblade-operator-69b7bc4cbb-ltmsl -n kube-system Name: chaosblade-operator-69b7bc4cbb-ltmsl Namespace: kube-system Priority: 0 Node: node2/192.168.43.104 Start Time: Fri, 24 Mar 2023 14:01:03 +0800 Labels: name=chaosblade-operator part-of=chaosblade pod-template-hash=69b7bc4cbb Annotations: Status: Running IP: 10.244.2.89 IPs: IP: 10.244.2.89 Controlled By: ReplicaSet/chaosblade-operator-69b7bc4cbb Init Containers: chaosblade-tool: Container ID: docker://79473403220fa0fa198c09e893e780630808fabffc58291519c0ef7d6e4d16fe Image: chaosbladeio/chaosblade-tool:1.7.1 Image ID: docker-pullable://chaosbladeio/chaosblade-tool@sha256:8cdb3474b28a45a0d8390dcc865ae22b12be527224c2e0f456d7934bdb0dc3d0 Port: Host Port: Command: cp -R /opt/chaosblade /home State: Terminated Reason: Completed Exit Code: 0 Started: Fri, 24 Mar 2023 14:01:10 +0800 Finished: Fri, 24 Mar 2023 14:01:10 +0800 Ready: True Restart Count: 0 Environment: Mounts: /home from chaosblade (rw) /var/run/secrets/kubernetes.io/serviceaccount from chaosblade-token-ds8bq (ro) Containers: chaosblade-operator: Container ID: docker://204a657085eb53e49659e896dc0fc704d9025dbf294466261c475da2fd4f7baa Image: chaosbladeio/chaosblade-operator:1.7.1 Image ID: docker-pullable://chaosbladeio/chaosblade-operator@sha256:ae7dbbd0964ab36fb8bb474f9e1e53472321498c7cd76cecd4efdfa480026e3b Port: 9443/TCP Host Port: 0/TCP Command: chaosblade-operator Args: --chaosblade-image-repository=chaosbladeio/chaosblade-tool --chaosblade-version=1.7.1 --chaosblade-image-pull-policy=IfNotPresent --log-level=info --webhook-enable --daemonset-enable --remove-blade-interval=72h --chaosblade-namespace=kube-system State: Running Started: Fri, 24 Mar 2023 14:49:47 +0800 Ready: True Restart Count: 0 Environment: WATCH_NAMESPACE: POD_NAME: chaosblade-operator-69b7bc4cbb-ltmsl (v1:metadata.name) OPERATOR_NAME: chaosblade-operator Mounts: /opt from chaosblade (rw) /tmp/k8s-webhook-server/serving-certs from cert (ro) /var/run/secrets/kubernetes.io/serviceaccount from chaosblade-token-ds8bq (ro) Conditions: Type Status Initialized True Ready True ContainersReady True PodScheduled True Volumes: cert: Type: Secret (a volume populated by a Secret) SecretName: chaosblade-webhook-server-cert Optional: false chaosblade: Type: EmptyDir (a temporary directory that shares a pod's lifetime) Medium: SizeLimit: chaosblade-token-ds8bq: Type: Secret (a volume populated by a Secret) SecretName: chaosblade-token-ds8bq Optional: false QoS Class: BestEffort Node-Selectors: Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s node.kubernetes.io/unreachable:NoExecute for 300s Events: Type Reason Age From Message
Normal Pulled 26m kubelet, node2 Successfully pulled image "chaosbladeio/chaosblade-operator:1.7.1" Normal Created 26m kubelet, node2 Created container chaosblade-operator Normal Started 26m kubelet, node2 Started container chaosblade-operator
把这个文件的- mountPath: /host-sys,改成- mountPath: /sys,重新执行tar zcvf chaosblade-operator-1.7.1.tgz chaosblade-operator,helm卸载原有operator,用这个包helm安装新operator就可以了。这个在1.6.1 统一改成 /host-sys 了;/sys 会覆盖容器自己的;所以改成 /host-sys 了。
此答案来自钉钉群“ChaosBlade开源讨论群"
版权声明:本文内容由阿里云实名注册用户自发贡献,版权归原作者所有,阿里云开发者社区不拥有其著作权,亦不承担相应法律责任。具体规则请查看《阿里云开发者社区用户服务协议》和《阿里云开发者社区知识产权保护指引》。如果您发现本社区中有涉嫌抄袭的内容,填写侵权投诉表单进行举报,一经查实,本社区将立刻删除涉嫌侵权内容。