k8s 1.2已部署locally, single-node docker
我做错了吗？这是否适用于其他所有人，或者在我的k8s部署中是否有问题？
按照ConfigMaps指南中的示例，应该在下面创建/etc/config/special.how，但不是：
[root@totoro brs-kubernetes]# kubectl create -f example.yaml
configmap "special-config" created
pod "dapi-test-pod" created
[root@totoro brs-kubernetes]# kubectl exec -it dapi-test-pod -- sh
/ # cd /etc/config/
/etc/config # ls
/etc/config # ls -alh
total 4
drwxrwxrwt 2 root root 40 Mar 23 18:47 .
drwxr-xr-x 7 root root 4.0K Mar 23 18:47 ..
/etc/config #
example.yaml
apiVersion: v1
kind: ConfigMap
metadata:
name: special-config
namespace: default
data:
special.how: very
special.type: charm
---
apiVersion: v1
kind: Pod
metadata:
name: dapi-test-pod
spec:
containers:
- name: test-container
image: gcr.io/google_containers/busybox
command: ["sleep", "100"]
volumeMounts:
- name: config-volume
mountPath: /etc/config
volumes:
- name: config-volume
configMap:
name: special-config
items:
- key: special.how
path: how.file
restartPolicy: Never
一致性测试失败的摘要如下（要求由jayunit100运行）。完整运行此gist。
Summarizing 7 Failures:
[Fail] ConfigMap [It] updates should be reflected in volume [Conformance]
/home/schou/dev/kubernetes/_output/local/go/src/k8s.io/kubernetes/test/e2e/configmap.go:262
[Fail] Downward API volume [It] should provide podname only [Conformance]
/home/schou/dev/kubernetes/_output/local/go/src/k8s.io/kubernetes/test/e2e/util.go:1637
[Fail] Downward API volume [It] should update labels on modification [Conformance]
/home/schou/dev/kubernetes/_output/local/go/src/k8s.io/kubernetes/test/e2e/downwardapi_volume.go:82
[Fail] ConfigMap [It] should be consumable from pods in volume with mappings [Conformance]
/home/schou/dev/kubernetes/_output/local/go/src/k8s.io/kubernetes/test/e2e/util.go:1637
[Fail] Networking [It] should function for intra-pod communication [Conformance]
/home/schou/dev/kubernetes/_output/local/go/src/k8s.io/kubernetes/test/e2e/networking.go:121
[Fail] Downward API volume [It] should update annotations on modification [Conformance]
/home/schou/dev/kubernetes/_output/local/go/src/k8s.io/kubernetes/test/e2e/downwardapi_volume.go:119
[Fail] ConfigMap [It] should be consumable from pods in volume [Conformance]
/home/schou/dev/kubernetes/_output/local/go/src/k8s.io/kubernetes/test/e2e/util.go:1637
Ran 93 of 265 Specs in 2875.468 seconds
FAIL! -- 86 Passed | 7 Failed | 0 Pending | 172 Skipped --- FAIL: TestE2E (2875.48s)
FAIL
findmnt的输出：
[schou@totoro single-node]$ findmnt
TARGET SOURCE FSTYPE OPTIONS
/ /dev/mapper/fedora-root
│ ext4 rw,relatime,data=ordere
├─/sys sysfs sysfs rw,nosuid,nodev,noexec,
│ ├─/sys/kernel/security securityfs securit rw,nosuid,nodev,noexec,
│ ├─/sys/fs/cgroup tmpfs tmpfs ro,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/systemd cgroup cgroup rw,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/cpuset cgroup cgroup rw,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/net_cls,net_prio cgroup cgroup rw,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/memory cgroup cgroup rw,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/hugetlb cgroup cgroup rw,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/cpu,cpuacct cgroup cgroup rw,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/perf_event cgroup cgroup rw,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/pids cgroup cgroup rw,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/blkio cgroup cgroup rw,nosuid,nodev,noexec,
│ │ ├─/sys/fs/cgroup/freezer cgroup cgroup rw,nosuid,nodev,noexec,
│ │ └─/sys/fs/cgroup/devices cgroup cgroup rw,nosuid,nodev,noexec,
│ ├─/sys/fs/pstore pstore pstore rw,nosuid,nodev,noexec,
│ ├─/sys/firmware/efi/efivars efivarfs efivarf rw,nosuid,nodev,noexec,
│ ├─/sys/kernel/debug debugfs debugfs rw,relatime
│ ├─/sys/kernel/config configfs configf rw,relatime
│ └─/sys/fs/fuse/connections fusectl fusectl rw,relatime
├─/proc proc proc rw,nosuid,nodev,noexec,
│ ├─/proc/sys/fs/binfmt_misc systemd-1 autofs rw,relatime,fd=32,pgrp=
│ └─/proc/fs/nfsd nfsd nfsd rw,relatime
├─/dev devtmpfs devtmpf rw,nosuid,size=8175208k
│ ├─/dev/shm tmpfs tmpfs rw,nosuid,nodev
│ ├─/dev/pts devpts devpts rw,nosuid,noexec,relati
│ ├─/dev/mqueue mqueue mqueue rw,relatime
│ └─/dev/hugepages hugetlbfs hugetlb rw,relatime
├─/run tmpfs tmpfs rw,nosuid,nodev,mode=75
│ ├─/run/user/42 tmpfs tmpfs rw,nosuid,nodev,relatim
│ │ └─/run/user/42/gvfs gvfsd-fuse fuse.gv rw,nosuid,nodev,relatim
│ └─/run/user/1000 tmpfs tmpfs rw,nosuid,nodev,relatim
│ └─/run/user/1000/gvfs gvfsd-fuse fuse.gv rw,nosuid,nodev,relatim
├─/tmp tmpfs tmpfs rw
├─/boot /dev/sda2 ext4 rw,relatime,data=ordere
│ └─/boot/efi /dev/sda1 vfat rw,relatime,fmask=0077,
├─/var/lib/nfs/rpc_pipefs sunrpc rpc_pip rw,relatime
├─/var/lib/kubelet/pods/fd20f710-fb82-11e5-ab9f-0862662cf845/volumes/kubernetes.io~secret/default-token-qggyv
│ tmpfs tmpfs rw,relatime
├─/var/lib/kubelet/pods/2f652e15-fb83-11e5-ab9f-0862662cf845/volumes/kubernetes.io~configmap/config-volume
│ tmpfs tmpfs rw,relatime
└─/var/lib/kubelet/pods/2f652e15-fb83-11e5-ab9f-0862662cf845/volumes/kubernetes.io~secret/default-token-6bzfe
tmpfs tmpfs rw,relatime
[schou@totoro single-node]$
答案 0 :(得分：8)
感谢@Paul Morie帮助我诊断和修复此问题（来自github issue）：
bingo，/ var / lib / kubelet的安装传播模式是私有的。尝试将kubelet dir的mount标志更改为
-v /var/lib/kubelet:/var/lib/kubelet:rw,shared
我还必须在我的docker systemd文件中将MountFlags=slave更改为MountFlags=shared。